Prediction model of interval grey number based on DGM(1,1)

In grey system theory,the studies in the field of grey prediction model are focused on real number sequences,rather than grey number ones.Hereby,a prediction model based on interval grey number sequences is proposed.By mining the geometric features of interval grey number sequences on a two-dimensional surface,all the interval grey numbers are converted into real numbers by means of certain algorithm,and then the prediction model is established based on those real number sequences.The entire process avoids the algebraic operations of grey number,and the prediction problem of interval grey number is usefully solved.Ultimately,through an example＇s program simulation,the validity and practicability of this novel model are verified.

NARX-GA-Elman Method for Mach Number Prediction of Wind Tunnel Flow Field

Mach number is a key metric in the evaluation of wind tunnel flow field performance.This complex process of wind tunnel test mainly has the problems of nonlinearity and time lag.In order to overcome the problems and control the Mach number stability,this paper proposes a new method of Mach number prediction based on a nonlinear autoregressive exogenous-genetic algorithm-Elman(NARX-GA-Elman)model,which adopts NARX as the basic framework,determines the order of the input variables by using the false nearest neighbor(FNN),and uses the dynamic nonlinear network Elman to fit the model,and finally uses the global optimization algorithm GA to optimize the weight thresholds in the model to establish the Mach number prediction model with optimal performance under single working condition.By comparing with the traditional algorithm,the prediction accuracy of the model is improved by 61.5%,and the control accuracy is improved by 55.7%,which demonstrates that the model has very high prediction accuracy and good stability performance.

Prediction of Kappa Number in Eucalyptus Kraft Pulp Continuous Digester Using the Box &Jenkins Methodology

The quality of the resulting pulping continuous digesters is monitored by measuring the Kappa number, which is a reference of residual lignin. The control of the kappa number is carried out mainly in the top of the digester, therefore it is important to get some indication of this analysis beforehand. In this context, the aim of this work was to obtain a prediction model of the kappa number in advance to the laboratory results. This paper proposes a new approach using the Box & Jenkins methodology to develop a dynamic model for predicting the kappa number from a Kamyr continuous digester from an eucalyptus Kraft pulp mill in Brazil. With a database of 1500 observations over a period of 30 days of operation, some ARMA models were studied, leading to the choice of ARMA (1, 2) as the best forecasting model. After fitting the model, we performed validation with a new set of data from 30 days of operation, achieving a model of 2.7% mean absolute percent error.

Mach Number Prediction for a Wind Tunnel Based on the CNN-LSTM-Attention Method

The test section’s Mach number in wind tunnel testing is a significant metric for evaluating system performance.The quality of the flow field in the wind tunnel is contingent upon the system's capacity to maintain stability across various working conditions.The process flow in wind tunnel testing is inherently complex,resulting in a system characterized by nonlinearity,time lag,and multiple working conditions.To implement the predictive control algorithm,a precise Mach number prediction model must be created.Therefore,this report studies the method for Mach number prediction modelling in wind tunnel flow fields with various working conditions.Firstly,this paper introduces a continuous transonic wind tunnel.The key physical quantities affecting the flow field of the wind tunnel are determined by analyzing its structure and blowing process.Secondly,considering the nonlinear and time-lag characteristics of the wind tunnel system,a CNN-LSTM model is employed to establish the Mach number prediction model by combining the 1D-CNN algorithm with the LSTM model,which has long and short-term memory functions.Then,the attention mechanism is incorporated into the CNN-LSTM prediction model to enable the model to focus more on data with greater information importance,thereby enhancing the model's training effectiveness.The application results ultimately demonstrate the efficacy of the proposed approach.

Study on Prediction Model of Number of Rainstorm Days in Summer Based on C5.0 Decision Tree Algorithm

Based on the data of daily precipitation in Lianyungang area from 1951 to 2012 and various climate signal data from the National Climate Center website and the NOAA website,a model for predicting whether the number of rainstorm days in summer in Lianyungang area is large was established by the classical C5. 0 decision tree algorithm. The data samples in 48 years( accounting for about 80% of total number of samples)was as the training set of a model,and the training accuracy rate of the model was 95. 83%. The data samples in the remaining 14 years( accounting for about 20% of total number of samples) were used as the test set of the model to test the model,and the test accuracy of the model was 85. 71%. The results showed that the prediction model of number of rainstorm days in summer constructed by C5. 0 algorithm had high accuracy and was easy to explain. Moreover,it is convenient for meteorological staff to use directly. At the same time,this study provides a new idea for short-term climate prediction of number of rainstorm days in summer.

Predictions of Galapagos Volcanic Eruption, El Niño, Ecuadorian Earthquake, Global Volcanic Eruption and Forest Fire by Sunspot Number

The earthquake number was correlated (R2 = 0.8781) with the volcano number in the East Pacific and Central American countries. The year of volcanic eruptions in the Galapagos Hot Spot (GHS) was proportional to the year of El Ni&#241o events (R2 = 0.9939) as well as to the year of earthquake events (R2 = 0.9943) in Ecuador, Colombia, and Peru from 1977 to 2016. Therefore, there can be El Ni&#241o and earthquake events in Ecuador, Colombia and Peru if there are strong volcanic eruptions either from aboveground volcanoes or undersea seamounts in the GHS during the maximal sunspot number’s period. Global volcanic eruptions were occurred during the maximal temperature departure, the latter being reversely proportional (R2 = 0.4512) to the sunspot number from 1980 to 2019. Forest fires occurred in middle latitude countries (South Korea, California, western Russia, Australia) around 35 degree from Pacific Equator during La Ni&#241a events at the time of the minimal sunspot number. Since there were intense forest fires in South Korea, Australia and California in 2019, more are expected in 2030 due to the 11 year cyclical variation in the sunspot number. The sequence of forest fires can be schematically summarized as below. 1) La Ni&#241a event in Equator causes the low vapor pressure of water due to the low sea surface temperature (SST);2) There is the pumping of freshwater from the warm Pacific Ocean to cold Equator with simultaneous transfer of heat (warm Pacific Ocean to cold Equator) and mass (water enriched Pacific Ocean to water lean Equator);3) Strong winds from dry zone pass Mountains to cause the forest fires in the residential land with bushes and trees. 4) According to the Bernoulli’s principle, the pressure gradient between the high Mountains and the low Ocean, induces the decrease of static velocity gradient for vigorously upwards flares in the Mountains and the residential land during the forest fires, which may be why it is difficult to extinguish the forest fires until burnt up bushes and trees in the Mountains and residential land with serious damages. Most effective solution to forest fires is to change the events from La Ni&#241a to El Ni&#241o for reductions of temperature and pressure gradients by Gay-Lussac’s law. Such a transition may be induced by artificial submarine volcanic eruptions among volcanoes of Fernandina, Sierra Negra and Wolf in the GHS for Korea and California. As for Australia, artificial volcanic eruptions can be caused in huge underwater volcano chain in Tasmania. The strong damages of forest fires in South Korea, Australia and California were observed in the period of the minimal sunspot number, as were in October of 2019 to February of 2020. Simultaneous transfers of momentum (velocity), heat (temperature) and mass (freshwater) were occurred from the dry zone or the hot Deserts via the Mountains and the residential land with bushes and trees to the cold Sea or Ocean. It is expected the lowest SST during the year of the minimal sunspot number to induce the strong damages of forest fires, as was in South Korea, Australia and California during the years from 2019 to 2020. CO2 emissions in Australia showed 34.5% increase, 70% coal power and no nuclear power to induce bushfires because of increases of ozone hole area and UV radiation for the hot land in the period of La Ni&#241a. Fossil fuel operation should be cut down to decrease the ozone hole area and UV radiation for weak bushfires in Australia. A few coming years may still have bushfires in Australia although their damages may not be as terrible as was in 2020. A simple remedy can be the reduction of CO2 emissions as low as possible. It was proposed that the maximal sunspot number induced El Ni&#241o event, GHS volcanic eruption and Ecuadorian earthquakes, while the minimal sunspot number induced La Ni&#241a events in Ecuador, Colombia, Peru and intense forest fires in middle latitude countries of South Korea (forest fires), Australia (bushfires) and California (wildfires).

The Genesis of Prime Numbers—Revealing the Underlying Periodicity of Prime Numbers

Prime numbers are the integers that cannot be divided exactly by another integer other than one and itself. Prime numbers are notoriously disobedient to rules: they seem to be randomly distributed among natural numbers with no laws except that of chance. Questions about prime numbers have been perplexing mathematicians over centuries. How to efficiently predict greater prime numbers has been a great challenge for many. Most of the previous studies focus on how many prime numbers there are in certain ranges or patterns of the first or last digits of prime numbers. Honestly, although these patterns are true, they help little with accurately predicting new prime numbers, as a deviation at any digit is enough to annihilate the primality of a number. The author demonstrates the periodicity and inter-relationship underlying all prime numbers that makes the occurrence of all prime numbers predictable. This knowledge helps to fish all prime numbers within one net and will help to speed up the related research.

Effects of Blade Number on Characteristics of Centrifugal Pumps

The blade number of impeller is an important design parameter of pumps, which affects the characteristics of pump heavily. At present, the investigation focuses mostly on the performance characteristics of axis flow pumps, the influence of blade number on inner flow filed and characteristics of centrifugal pump has not been understood completely. Therefore, the methods of numerical simulation and experimental verification are used to investigate the effects of blade number on flow field and characteristics of a centrifugal pump. The model pump has a design specific speed of 92.7 and an impeller with 5 blades. The blade number is varied to 4, 6, 7 with the casing and other geometric parameters keep constant. The inner flow fields and characteristics of the centrifugal pumps with different blade number are simulated and predicted in non-cavitation and cavitation conditions by using commercial code FLUENT. The impellers with different blade number are made by using rapid prototyping, and their characteristics are tested in an open loop. The comparison between prediction values and experimental results indicates that the prediction results are satisfied. The maximum discrepancy of prediction results for head, efficiency and required net positive suction head are 4.83%, 3.9% and 0.36 m, respectively. The flow analysis displays that blade number change has an important effect on the area of low pressure region behind the blade inlet and jet-wake structure in impellers. With the increase of blade number, the head of the model pumps increases too, the variable regulation of efficiency and cavitation characteristics are complicated, but there are optimum values of blade number for each one. The research results are helpful for hydraulic design of centrifugal pump.

基于季节ARIMA模型对某三级综合性医院门诊量的预测研究

目的 通过建立季节ARIMA模型,对浙江省某三级综合性医院门诊量进行预测,为医院合理配备门诊人力资源提供依据。方法 以2013年1—6月浙江省某医院门诊量数据为基线,利用SPSS软件构建季节ARIMA模型,对2023年7—12月的门诊量进行预测,通过对比门诊量实测值,评价季节ARIMA模型预测门诊人次的精度。结果 该综合性医院门诊量呈现逐年上升趋势,并呈现周期性波动的特征。拟合的最优季节ARIMA模型为ARIMA(0,1,1)(1,0,1)12,BIC(贝叶斯信息准则)为5.273,MAPE(平均绝对百分误差)为14.265,R2(模块决定系数)为0.408,总体相对误差为1.83%,预测结果良好。结论 季节ARIMA模型较好地模拟了该三级综合性医院门诊量在时间序列上的变化趋势,为该院门诊量的短期预测提供理论依据。

暴雨洪涝灾害转移安置人数的组合预测模型研究

为了更加科学精准地预测暴雨洪涝灾害下需要转移安置的人数,收集2011—2018年全国范围内严重暴雨洪涝灾害案例,通过Pearson相关性分析检验转移安置人数与表征暴雨洪涝灾害严重程度影响因素之间的关系;分别使用基于主成分分析的回归模型和支持向量机(SVM)预测暴雨洪涝灾害下需要转移安置人数,并以2种方法的结果为基础,提出1种组合预测方法对暴雨洪涝灾害转移人数进行修正。研究结果表明:组合预测法的MSE、MAE均小于回归预测和SVM模型预测。使用组合预测方法对洪涝灾害转移安置人数进行预测,可以充分结合单一预测模型的优势,提高组合预测模型的预测精度和泛化能力。研究结果可为确定暴雨洪涝灾害的避难需求并制定避难疏散计划提供参考。

基于图像特征的森林地表凋落物载量分析

【目的】森林地表凋落物载量值影响林火的发生和森林火灾所表现的一系列火行为特征等,准确获取地表凋落物载量值十分重要。图像特征欧拉数能够表征图像中对象的多少,分析欧拉数与载量之间的关系,并建立基于图像欧拉数的载量预测模型对于载量研究具有重要意义。【方法】以贵州省典型林分柳杉林和毛竹林内凋落物为研究对象,通过野外林分和载量调查、拍摄凋落物图片和图片特征处理,分析图像特征欧拉数与地表凋落物载量之间的关系,建立基于图像欧拉数的载量预测模型,并检验模型精度。【结果】1)选择不同阈值对图像二值化处理后,提取得到的欧拉数并不是都与载量存在相关性,阈值为0.1对图像二值化后的图像欧拉数与两种凋落物载量呈极显著相关;2)随着图像欧拉数的增加,柳杉和毛竹林地表凋落物载量整体呈下降趋势;3)选择线性回归建立基于图像特征欧拉数的凋落物载量预测模型,柳杉和毛竹林凋落物载量的预测模型绝对误差分别为1.60和1.72 t·hm^(-2),相对误差分别为20.03%和20.71%,柳杉林地表凋落物载量的预测效果要优于毛竹林。【结论】本研究验证了基于图像特征预测森林地表凋落物载量的可行性,为准确获取载量研究提供新思路,对于火险预报和科学林火管理具有重要意义。

基于图神经网络与迁移学习的流行病例数预测

预测流行病的病例数对研究流行病学和保障卫生安全至关重要,但现有的研究工作很少考虑到实时移动性数据等因素,这一问题给病例数的预测研究带来了挑战。因此,在图神经网络GNN的基础上提出一种新型计算框架-信息聚合网络IAN,既考虑地区病例数据特征,也考虑地区之间的人口移动性数据特征。为了优化各个国家的前期预测模型,在该框架的基础上加入迁移学习方法TL。在四个欧洲国家数据集上的实验结果表明,IAN以及IAN-TL明显优于传统方法,能够有效地降低预测误差。

气浮电动缸功率匹配分析与控制算法设计

针对承担重载的电驱机构存在运行功率大、机械结构易磨损等问题,论文介绍了一种气浮电动的联合驱动方案,首先分析了气浮电动缸的工作原理及电、气驱动力和功率匹配情况,然后把电动系统功率作为预测变量,以降低电动系统功率为目的,设计了气动系统模型预测控制(MPC)算法。仿真结果显示,应用MPC算法后的气动系统分担了较大驱动力,降低电动系统功率效果显著。

建筑光伏储能系统优化管理策略及影响因素研究

在建筑能源系统中引入光伏已成为可持续能源的有效解决方案之一。然而光伏发电受时间和环境影响明显,其变化规律很难与建筑自身用能需求完全匹配。为了解决该问题,在峰谷电价背景下,提出一种新型的基于发电和用电预测的建筑光伏储能系统优化管理策略,并重点研究实际运行工况下,发电量和用电量预测精度、建筑层数、季节类型对优化管理策略性能的影响。首先基于极端梯度提升(XGBoost)算法,开发了建筑用电量和光伏发电量逐时预测模型。在此基础上,针对光伏、蓄电池、电网及建筑4个模块,利用遗传算法优化未来24h蓄电池逐时充放电量,从而实现系统日用电成本最小化。研究发现,在发电量与用电量比较接近时,优化管理策略相比传统策略有显著的性能提升,特别适用于不高于2层的建筑,有望在城镇中广泛推广。而对于城市高层建筑,在光伏发电量远低于建筑用电量的场景下,优化管理策略虽然优势并不显著,但效果仍优于传统策略。研究还发现,对于不高于2层的建筑,即使发、用电量预测精度不高,优化管理策略依然性能良好,且随着预测精度的上升性能会进一步提升。

目标个数不规则变化的动态多目标优化算法

文中提出了一种基于混合策略的初始种群预测算法(A Hybrid Strategy Based Initial Population Rrediction Algorithm,HIPPA)来解决目标个数随时间不规则变化的动态多目标优化问题。HIPPA依据目标个数判断环境是否发生变化,根据不同的目标个数划分环境类型。在种群初始化阶段,初始种群由3种机制产生。首先,利用历史种群信息训练改进的神经网络算法,生成一部分初始种群。其次,改进的精英策略利用历史种群信息生成一部分初始种群。最后,使用改进的随机策略生成一部分种群,以保持种群的多样性。本文使用基准实验F1-F5验证所提算法的有效性,并将结果与其他动态优化算法对比。实验结果表明,HIPPA可以更加有效地解决目标个数随时间不规则变化的动态多目标优化问题。

基于改进TabNet的糖尿病预测模型

针对糖尿病数据特征冗余、类别数不平衡导致的预测困难问题,提出一种基于改进TabNet的糖尿病预测模型dTabNet(dual TabNet)。首先,通过dTabNet得到量化的特征全局贡献指标,同时构造3层全连接层替代特征转换器和注意力转换器中的单层全连接层,以增强对数据的表示能力;其次,设计特征选择模块,以去除糖尿病数据集中的冗余和无关特征,从而提高学习的效率;最后,根据特征选择模块输出的特征子集预测糖尿病。此外,引入Focal loss函数优化损失函数,以解决糖尿病数据类别数不平衡的问题;之后,利用贝叶斯优化算法对dTabNet进行超参数优化。在预处理后的真实糖尿病检测数据集上评估所提模型。实验结果表明,dTabNet针对糖尿病预测的准确率达到了90.9%,曲线下面积(AUC)值达到了94.7%,相较于TabNet的准确率和AUC值分别提高了4.0和2.2个百分点。这表明所提模型在数据冗余、类别数量不平衡的糖尿病数据上能有效地完成糖尿病预测。

基于改进相关系数和IGOWLA算子的三角模糊数组合预测模型

随着经济系统复杂程度日益提高,用三角模糊数能更好地反映不确定信息.为保证预测结果的内部整体性,提出将三角模糊数序列的3个界点值转化为等价信息的面积中心、中界值、重心并构建新的序列.随后以预测精度为诱导因子,构建诱导广义有序加权对数平均(IGOWLA)算子,在此基础上选取改进后的Pearson相关系数作为最优准则,构建多目标组合预测模型,并引入偏好系数将模型进一步转化为单目标规划模型,通过实例分析证明了组合预测模型的有效性.结果表明:该组合预测模型可以在保证内部整体性的基础上有效地提高预测精度.

神仙豆腐的货架期预测模型

[目的]研究贮藏温度对神仙豆腐品质和微生物菌落总数的影响规律,建立基于品质指标和菌落总数的神仙豆腐货架期保鲜预测模型。[方法]将神仙豆腐分别在1,4,10,15,20,25℃的温度下贮藏,定期测定硬度、黏性、弹性、咀嚼性、持水性以及菌落总数,研究温度对各指标的影响,确定各指标的最佳拟合方程。根据神仙豆腐的品质指标及菌落总数,结合Arrhenius模型和Belehradck模型,建立神仙豆腐货架期预测模型。[结果]在不同的贮藏温度下,神仙豆腐的弹性指数无明显变化,但对硬度、黏性、咀嚼性、持水性及菌落总数有显著影响。零级反应方程对神仙豆腐的持水性变化拟合效果最佳,平均拟合程度为0.978;对硬度、黏性和咀嚼性的影响,一级反应的拟合最佳,平均拟合程度分别为0.967,0.904,0.977;Logistics方程对神仙豆腐菌落总数变化的拟合效果最佳,平均拟合为0.992。建立的硬度、黏性、弹性、咀嚼性、持水性及菌落总数模型对3℃贮藏条件下神仙豆腐货架期预测的相对误差分别为7.50%,8.13%,4.38%,5.00%,3.75%,预测效果好。[结论]模型可以有效预测1~25℃贮藏条件下神仙豆腐的货架期。

共情护理辅助预见性护理对采血室儿童静脉采血穿刺成功率、不合格血标本数的影响

目的 研究共情护理辅助预见性护理对采血室儿童静脉采血穿刺成功率、不合格血标本数的影响。方法 回顾性分析我院2022年9月至2022年11月282例静脉采血患儿的的临床资料,根据护理方式不同分为对照A组(94例)、对照B组(94例)、观察组(94例),对照A组予以常规护理,在此基础上,对照B组予以预见性护理,观察组予以共情护理辅助预见性护理。比较3组一次采血成功率、不合格血标本数、疼痛行为反应(FLACC)、儿童哭闹时长、采血时长、穿刺次数、静脉穿刺并发症、家属护理满意度。结果 观察组、对照B组一次采血成功率95.74%(90/94)、91.49%(86/94)高于对照A组81.91%(77/94),不合格血标本数0.86%(2/233)、1.29%(3/232)低于对照A组5.22%(12/230)(P<0.05);观察组儿童哭闹时长、采血时长短于对照A组、对照B组,且对照B组短于对照A组,穿刺次数少于对照A组、对照B组,且对照B组少于对照A组(P<0.05);观察组FLACC量表体位、面部表情、腿部活动、可安慰度、哭闹条目评分低于对照A组、对照B组,且对照B组低于对照A组(P<0.05);观察组、对照B组静脉穿刺并发症2.13%(2/94)、3.19%(3/94)低于对照A组12.77%(12/94)(P<0.05);观察组家属护理满意度98.94%(93/94)高于对照A组80.85%(76/94)、对照B组90.43%(85/94),且对照B组高于对照A组(P<0.05)。结论 共情护理辅助预见性护理能减轻静脉采血患儿疼痛程度,提高一次穿刺成功率,减少不合格血标本数,也能提高家属护理满意度。

基于感染人数预测的应急仓库订货数量研究

文章基于LSTM模型预测感染人数,构建了物资需求模型,推导出应急仓库订货数量。文章首先从应急物资分类入手,为构建订货模型铺垫。然后,针对预测感染人数这一问题的特点选定LSTM作为预测模型,并搭建LSTM模型网络。最后综合各方面影响因素,建立订货数量计算公式。文章提供了重大突发公共卫生事件下单周期应急仓库订货数量量化方法,为应急物资供应链仓储订货提供了科学的数据支撑。