To address the challenge of detecting abnormal navigation trajectories of ves-sels and to provide informational support for safe navigation and military decision-making,this paper proposes a suspicious vessel early warning algorithm CALS that leverages deep learning and sliding window techniques.Utilizing one year's worth of AIS data from Chinese maritime areas,Convolutional Neural Networks(CNN)is employed to extract multidimen-sional features related to vessel movement parameters.The AIS data is formatted temporally and input into a Long Short-Term Memory(LSTM)network,where an attention mechanism is introduced to assign weights to the hidden layers.This approach enables us to differentiate the impact levels of various time points on predicting vessel movement parameters,thereby optimizing the prediction model.Ultimately,a sliding window algorithm is implement for anomaly detection in streaming data,achieving timely warnings for abnormal ship trajecto-ries.In the experiments,8576 AIS messages are selected from vessels operating in Chinese maritime areas on a specific day in 2021 for testing purposes.The results indicate that the neural network prediction model incorporating the attention mechanism significantly out-performs both LSTM and CNN-LSTM models regarding prediction accuracy,with an early warning accuracy consistently around 90％.
This article mainly studies the pricing and inverse problem of European options for fractional order Black Scholes equations under the generalized CEV model.Firstly,the pricing problem of fractional order Black Scholes equations combined with European options under the generalized CEV model is introduced.Based on the generalized CEV volatility model with dividends proposed by Cox and Jumarie,the pricing formula satisfied by European options is derived.Secondly,the difference is dispersed in space and time,and the effectiveness of the model is verified through numerical simulation.Finally,the pricing and inverse problem of fractional order Black Scholes equation European options under the generalized CEV model were discussed,and numerical experiments were conducted.Empirical analysis was conducted in conjunction with the Chinese options market,and the elasticity coefficient was inverted.
Joint optimization of runway and taxiway will help increase the usage rate of airport’s existing hardware and software resources and ease flight delays. Firstly, the paper considers the relevant regulations of taxiing and the constraints of runway release interval, and takes the shortest total taxi time of each flight as the objective function to construct a joint optimization model based on the basic element layout of the airport. Secondly, the A* algorithm is improved for its characteristics and the actual situation of the airport surface, and the optimization model is solved by this improved method. Finally, taking Nanjing Lukou International Airport as an example, the optimal solution obtained by improved A* algorithm is compared with the actual running data to verify the applicability of the model.
Considering the environmental awareness of contractors, this paper constructs a construction material supply chain consisting of the government, a manufacturer, and a contractor, analyzes the decision-making and influencing factors of various stakeholders under no subsidy, contractor subsidy, and manufacturer subsidy, and studies the formulation of subsidy policies for construction waste recycling from the perspective of the government. The results show that: compared with subsidizing the manufacturer, when the government subsidizes the contractor, the recycling rate is lower, and the manufacturer’s profit, consumer surplus, and social welfare are higher. The manufacturer will use the price mechanism to grab government subsidies at this time, which will increase governmental subsidy expenditure. We also show that the recycling cost and the environmental awareness of the contractor affect the decision-making of the government and the manufacturer in different ways. Therefore, the government should analyze the contractor’s environmental awareness and evaluate the manufacturer’s recycling cost before formulating subsidy policies to promote construction waste recycling.
In this note we give the definition of generalized Kato type and establish for a bounded linear operator defined on a Hilbert space the sufficient and necessary conditions for which property (ω) holds by means of the induced spectrum of generalized Kato property. In addition, the stability of property (ω) is discussed.
Recently, a new treatment based on Taylor's expansion to give the estimate of the convergence radius of iterative method for multiple roots has been presented. It has been successfully applied to enlarge the convergence radius of the modified Newton's method and Osada's method for multiple roots. This paper re-investigates the convergence radius of Halley's method under the condition that the derivative f ( m + 1 ) of function f satisfies the center-Hölder continuous condition. We show that our result can be obtained under much weaker condition and has a wider range of application than that given by Bi et. al.(2011) 21.
Basing on the Wiener amalgam spaces, the purpose of this paper is to introduce the weighted Hardy–Orlicz-amalgam spaces of martingales and to establish the atomic decomposition theorem. As an application, we obtain a duality theorem, namely, the dual of martingale weighted Hardy–Orlicz-amalgam spaces are generalized martingale weighted amalgam Campanato spaces.
This paper combines prospect theory with real option pricing model to construct a new value assessment model for big data assets. For the high uncertainty of future earnings and risk of big data assets, this paper analyzes that their value characteristics are in line with the American call option firstly. On this basis, we use the value function in the prospect theory to calculate decision makers' subjective judgments on the value of underlying big data assets under each state, and use the weighting function to calculate the decision makers' subjective judgment on the weight of expansion right, downsize right and abandon right. In example, we use least-squares Monte Carlo simulation method to perform a simulation which verifies the real option pricing method based on perspective of prospect theory can obtain more reasonable assessment result for big data assets.
提取了川南经济区2015年NPP-VIIRS夜间灯光数据灯光计数、灯光总量、灯光均值、灯光标准差、灯光最小值、灯光最大值等6个指标,并收集整理了该地区2015年与城市化密切相关的地区生产总值、第一产业增加值、地方公共财政支出、城镇化率、年末常住人口等5个社会统计数据指标,发现不同尺度下各指标的相关性有较大差异.在此基础上,利用偏最小二乘法对地市级尺度上夜间灯光数据的6个指标和社会统计数据的5个指标进行了回归,比较了县级尺度上偏最小二乘回归与最小二乘回归的效果.结果表明:利用夜间灯光数据回归地区生产总值、第一产业增加值、地方公共财政支出、城镇化率、年末常住人口时均能通过检验,但回归效果上有一定的差异;用夜间灯光数据回归以上社会统计数据时不能只考虑灯光总量,在对年末常住人口等指标进行回归时灯光最大值等也起着及其重要的作用.
针对固定权系数组合预测模型在滑坡监测数据多期预测预报中存在精度下降过快的问题,以最小二乘线性组合预测模型为基础,结合生物体新陈代谢过程,提出了一种基于最小二乘准则的滑坡位移动态权系数线性组合预测模型的构建方法,并引入SSE、MSE、MAE、MAPE和MSPE5种误差指标对该模型的预测精度进行评价.通过黄茨滑坡B2监测点和重庆市巫溪县物流园区长宁路边坡工程DW4监测点的监测数据进行算例验证,实验结果表明,基于最小二乘准则的滑坡位移动态权系数线性组合预测模型的预测评价指标均优于选用的单项预测模型以及普通线性组合预测模型.
模糊投资组合选择问题是在基本投资组合模型中引入模糊集理论,使所建立的模型与实际市场更加吻合,但同时也增加了模型求解难度.因此,本文针对两种不同的模糊投资组合模型,提出一种改进帝企鹅优化算法.算法首先引入可行性准则,处理模糊投资组合模型中的约束.其次,算法中加入变异机制,平衡算法的开发和探索能力,引导种群向最优个体收敛.通过对CEC 2006中的13个标准测试问题及两个模糊投资组合问题实例进行数值实验,并与其他群智能优化算法进行结果比较,发现本文所提出的算法具有较好的优化性能,并且对于求解模糊投资组合选择问题是有效的.
证明了鞅的q阶均方算子S(q)(.)在两类BMO空间BMOφq(X)和wBMOφq(X)上有界的充要条件是X同构于q 一致凸Banach空间,所得结果给出了 Banach空间的q 一致凸性新的等价刻画形式,并且将已有文献中的相关结论进行了推广.
在集合Ω中,把犹豫模糊集、Ω模糊集相结合来研究BCI-代数.首先在BCI代数中引入闭Ω犹豫模糊理想的概念,讨论它的一些性质和等价刻画;其次,在闭犹豫模糊理想概念的基础上研究了如何构造闭Ω犹豫模糊理想,讨论了闭Ω犹豫模糊理想的同态像与同态原像的性质;最后,给出了闭Ω犹豫模糊理想与乘积型BCI代数的闭Ω犹豫模糊理想的关系.
大类资产配置策略为资金相对庞大的机构投资者提供了一个有效获取稳健收益的手段,通过因子收益分布及相关性的预测能更好地进行大类资产配置.将大类资产因子配置的思想与机器学习算法预测有机结合,首先筛选宏观因子及风格因子,利用长短记忆神经网络(LSTM)方法预测组合收益,得到最优因子组合;然后结合最优因子组合中蕴含的信息,修正对资产预期收益率的估计并提出了大类资产的权重配置方案.通过在全球18种大类资产上进行的算例分析表明,采用本文模型得到的配置策略相较于其它模型有更高的收益风险比、较低的年化波动以及较小的最大回撤.研究结论可为机构投资者的大类资产配置提供理论借鉴.
针对地铁深基坑的难度大、风险高等特点,提出采用熵值法和突变理论相结合的方法进行地铁基坑施工的安全性评估方法.选取人员、机械、材料、技术、环境5个方面的因素建立3级20个评估指标的安全评估指标体系,在此基础上通过熵值权重和突变级数构建了地铁深基坑施工安全性评估数学模型.依托福州轨道交通6号线工程第3标段明挖深基坑工程,对评估模型进行验证,其结果符合实际工程情况,说明模型可行,结果可靠,能够指导实际工程.
对广义Rosenau-KdV方程提出一种在时间层和空间层上分别具有二阶和四阶精度的三层线性差分格式,所建格式是离散质量守恒和离散能量守恒的,利用离散能量法证明了差分格式的可解性、收敛性和稳定性.数值实验验证了该格式的精度和守恒性.
本文研究了一类捕鱼期、休渔期交替变换,带有Holling Ⅱ型功能反应与Beddington-DeAngelis型功能反应的两种群的捕食食饵模型.主要研究系统有界性、持久性、灭绝性等动力学行为,通过构造合适的Lyapunov函数来研究平衡点的全局渐近稳定性,并建立了相应的判定准则,最后通过数值模拟验证了理论结果的有效性.
变窗宽局部多项式模型是探索被估计曲线复杂变化结构的有力工具,其关键思想在于针对不同变化模式的数据区间选择不同窗宽参数进行拟合.基于回归树模型提出一种变窗宽多项式拟合方法,利用回归树的分类功能识别具有不同变化结构的数据区间,在每个子区间上独立选取最优窗宽,通过局部多项式回归实现复杂结构曲线的变窗宽拟合.随机模拟的结果表明,回归树能够有效识别具有不同变化模式的数据区间,变窗宽局部多项式拟合均方误差小,具有计算高效、结果易解释的特点.
研究考虑政府对绿色产品的补贴,对制造商和零售商组成的二级绿色闭环供应链构建集中决策和以制造商为主导的Stackelberg博弈模型.研究发现在以制造商为主导决策下,企业生产的绿色度、相关定价以及供应链利润均没有达到最优.在此基础上,以集中决策下的整体利润为目标,提出了收入共享成本共担契约协调,得到最优的协调比例.并通过算例和参数敏感性分析,验证了相关推论的正确性和使用契约的有效性.研究表明,当政府单位补贴、绿色度敏感系数和回收价敏感系数越大,绿色产品研发投入系数和价格敏感系数越小时,各企业利润越大.
