基于灰色+神经网络复合预测模型的研究

本文对单一的预测方式不能综合考虑复合因素对预测造成的影响,容易给预测带来偏差过大的后果的基础上,首先对数据进行描述性分析和相关分析,再使用Lasso参数估计方法,对某市财政收入数据进行相关因素分析,选取对财政收入影响较大的8个因素,使用灰色模型对这8个影响因素进行接下来2年值的预测,并使用后验差检验判别对预测结果进行评价,得出预测结果好;最后使用预测因素值建立神经网络预测模型,得到对接下来2年的财政收入预测结果。

划分时段的公交车辆到站时间复合预测模型

在总结国内外现有的公交车辆到站时间预测模型的基础上,选取常用的三种经典预测模型,即Kalman滤波模型、BP神经网络模型、时间序列模型。分析并针对各个模型的特点及适用条件,对影响公交车辆到站时间的要素进行筛选,明确各模型的输入及输出条件,划分公交运行的高峰、平峰、低峰时段,并验证三种预测模型在各个时段的预测精度,进一步确定复合预测模型中各阶段预测方法 ,以实现公交车辆的到站时间的精确预测。

基于五桥臂MMC-UPQC的复合模型预测控制研究

针对模块化多电平统一电能质量调节器(modular multilevel unified power quality conditioner, MMC-UPQC)六桥臂结构下的单相桥臂故障问题,提出了一种五桥臂拓扑,这种新型拓扑可实现故障情况下的电能质量补偿。首先,对MMC-UPQC串并联侧的数学模型进行分析,提出了一种复合模型预测控制(hybrid model predictive control,H-MPC),所提控制方法结合了有限集模型预测控制(finite-control-set model predictive control, FCS-MPC)以及快速模型预测控制(fast model predictive control, F-MPC)。然后,通过构建两侧独立的价值函数减少了控制方法的计算量,同时也实现了五桥臂解耦控制。最后,相比传统线性(例如PI)和非线性(例如无源控制passivity-based control,PBC)的控制策略,所提复合模型预测控制在电压补偿、负序电压抑制以及谐波电流补偿等方面具有一定优势,并在一定程度上避免了复杂的参数整定及坐标变化环节。仿真实验结果证明了所提控制方法的可行性和优越性。

永磁伺服电机复合模型预测控制

永磁电机效率和功率密度高、力矩和惯量比大,是工业伺服领域的主流电机。伺服控制技术是充分发挥永磁电机优势、提升伺服系统运行性能的关键。目前,永磁伺服系统多采用多环级联的比例-积分(PI)控制器,但由于积分器的滞后效应,PI动态响应速度较慢,抗干扰能力较差,难以满足机械臂、精密加工等高性能伺服控制的动、静态性能要求。因此,提出一种广义模型预测控制与有限集模型预测控制相结合的复合模型预测控制策略。此外,还提出一种广义模型预测控制的低运算量实现方法及一种机械参数估计方法。试验结果表明,所提复合模型预测控制可提高永磁伺服电机的动态响应速度和抗负载扰动能力。

短期负荷预测分析及一种新模型构想

在分析、比较目前较为成熟的短期负荷预测模型的优、缺点的基础上,提出一种基于支持向量机理论的复合预测模型的构想,该模型通过对2个训练集样本的选择,弥补因单一训练集过大(包含冗余信息)或过小(必要信息遗失)造成的预测精度下降。为提高训练的收敛速度,预测时,可通过将负荷分类的方式简化模型。

一种风电功率的复合预测

本文以东南沿海地区某风力发电场数据为背景,在分析原始数据特点后,确定了相应的缺失数据的填补方法以及数据的预分解方法。之后针对数据预处理结果建立了基于时间序列和优化的BP神经网络复合预测模型,并给出风电功率预测结果。最后比较了复合模型与其它模型预测的均方误差以说明复合预测模型在提高预测精度方面的优势。

复合预测在马尾松林分蓄积量生长过程中的应用

用林分林龄——公顷蓄积量序列,先用两种方法建立回归曲线模型,进行年龄——公顷蓄积量预测,然后根据两种回归曲线模型预测值,用复合预测方法进行预测,从而减小预测误差,提高预测精度,使预测结果更符合客观实际。

基于TCN-DenseNet的烧结矿FeO含量预测

烧结矿FeO含量是烧结工序的一项重要质量和能耗指标,也对高炉冶炼有直接影响。针对目前化学检测法检测烧结矿FeO含量时存在较长时间滞后的现状,本文提出了一种时域卷积网络(Temporal Convolutional Network,TCN)与密集连接卷积神经网络(Densely Connected Convolutional Network,DenseNet)混合的烧结矿FeO含量预测方法。首先采用TCN建立烧结矿FeO含量的时间序列预测模型,同时采集烧结机尾断面红外图像,采用DenseNet建立烧结矿FeO预测模型,通过自适应加权平均方法将两者的输出结果进行整合,获得最终的烧结矿FeO含量预测值。针对烧结矿层断面红外图像的特征,对DenseNet进行了添加注意力层、修改卷积块结构,并修改了浅层卷积层大小和步长等改进措施。在国内某钢铁公司的大型烧结机的实际生产数据上对模型进行了验证,经过数据处理、模型参数优化等操作后,本文所提的TCN-DenseNet混合模型的烧结矿FeO含量预测在测试集绝对误差±0.4%以内命中率可达94.34%,均方根误差为0.21,优于单独使用TCN或者DenseNet进行建模时的预测效果。该方法对提高烧结矿FeO含量预测的准确性和稳定性效果显著,可以为烧结现场的生产操作提供数据支撑。

A hybrid model for short-term rainstorm forecasting based on a back-propagation neural network and synoptic diagnosis

Rainstorms are one of the most important types of natural disaster in China.In order to enhance the ability to forecast rainstorms in the short term,this paper explores how to combine a back-propagation neural network(BPNN)with synoptic diagnosis for predicting rainstorms,and analyzes the hit rates of rainstorms for the above two methods using the county of Tianquan as a case study.Results showed that the traditional synoptic diagnosis method still has an important referential meaning for most rainstorm types through synoptic typing and statistics of physical quantities based on historical cases,and the threat score(TS)of rainstorms was more than 0.75.However,the accuracy for two rainstorm types influenced by low-level easterly inverted troughs was less than 40%.The BPNN method efficiently forecasted these two rainstorm types;the TS and equitable threat score(ETS)of rainstorms were 0.80 and 0.79,respectively.The TS and ETS of the hybrid model that combined the BPNN and synoptic diagnosis methods exceeded the forecast score of multi-numerical simulations over the Sichuan Basin without exception.This kind of hybrid model enhanced the forecasting accuracy of rainstorms.The findings of this study provide certain reference value for the future development of refined forecast models with local features.

Immobilization of Lactobacillus rhamnosus TISTR108 on Crude Pectin of Krung Kha Mao Leaves （Cissampe/os pareira L.） to Produce Lactic Acid in Longan Juice

L-（＋）-lactic acid production was studied by immobilized Lactobacillus rhamnosus T1STR108 on crude pectin from Krung Kha Mao Leaves. Central composite design was employed to determine the maximum lactic acid production of 42.88 g L-1 in predicted model with the factors at 4.11 g L1 of pectin, 6.05 mLLl inoculum and 1.09 mm of bead diameter. Statistical analyses demonstrated very high significance for the regression model, since the F-value computed 116.09 was much higher than the tabulated F-value 2.08 for the lactic acid production at 5% level for linear and quadratic polynomial regression models. The highest experimental lactic acid production was 43.57 g L^-1 at 96 h of fermentation, 1.58% higher than the predicted value.

Critical weight on bit of double-driven bottomhole assembly during vertical and fast drilling

It is difficult to determine the optimal weight on bit (WOB) of the double-driven bottomhole assembly (DD-BHA, with double stabilizers and a bent housing positive displacement motor (PDM)) which is employed during vertical and fast drilling. High WOB leads to well deviation out of control, and low WOB leads to low rate of penetration (ROP). So considering the rock physical properties, the anisotropy index function of polycrystalline diamond compact (PDC) bit was derived with the structure and cutting performance parameters of the bit, and the effect of natural hole deviation tendencies on the performance of DD-BHA resisting deviation was represented. The concept of elliptic deformation ratio was used to characterize the performance of DD-BHA resisting deviation. Eventually, a model calculating the critical WOB was established. By comparing the model predictions with the measured hole angle changes in the field, the results show that the model predictions are accurate with error less than 5.8%, which can meet the operational requirements in the projects. Furthermore, the model was adopted to justify and guide the operating conditions and parameters during drilling, which shows that the optimum WOB predicted by the model can not only control deviation but also improve ROP effectively. The model is independent on the formation characteristics of blocks, so it can be expanded widely to other oilfields.

Predicting Complex Word Emotions and Topics through a Hierarchical Bayesian Network

In this paper, we provide a Word Emotion Topic （WET） model to predict the complex word e- motion information from text, and discover the dis- trbution of emotions among different topics. A complex emotion is defined as the combination of one or more singular emotions from following 8 basic emotion categories： joy, love, expectation, sur- prise, anxiety, sorrow, anger and hate. We use a hi- erarchical Bayesian network to model the emotions and topics in the text. Both the complex emotions and topics are drawn from raw texts, without con- sidering any complicated language features. Our ex- periment shows promising results of word emotion prediction, which outperforms the traditional parsing methods such as the Hidden Markov Model and the Conditional Random Fields（CRFs） on raw text. We also explore the topic distribution by examining the emotion topic variation in an emotion topic diagram.

Numerical analysis of temperature rise within 70MPa composite hydrogen vehicle cylinder during fast refueling

The numerical simulation model for predicting fast filling process of 70 MPa type Ⅲ(with metal liner) hydrogen vehicle cylinder was presented,which has considered turbulence,real gas effect and solid heat transfer issues.Through the numerical analysis method,the temperature distributions of the gas within the solid walls were revealed; adiabatic filling was studied to evaluate the heat dissipation during the filling; the influences of various filling conditions on temperature rise were analyzed in detail.Finally,cold filling was proposed to evaluate the effect on temperature rise and SoC(state of charge) within the cylinder.The hydrogen pre-cooling was proved to be an effective solution to reduce maximum temperature and acquire higher SoC during the filling process.

Simplification and improvement of prediction model for elastic modulus of particulate reinforced metal matrix composite

In this paper, we proposed a five-zone model to predict the elastic modulus of particulate reinforced metal matrix composite. We simplified the calculation by ignoring structural parameters including particulate shape, arrangement pattern and dimensional variance mode which have no obvious influence on the elastic modulus of a composite, and improved the precision of the method by stressing the interaction of interfaces with pariculates and maxtrix of the composite. The five- zone model can reflect effects of interface modulus on elastic modulus of composite. It overcomes limitations of expressions of rigidity mixed law and flexibility mixed law. The original idea of five zone model is to put forward the particulate/interface interactive zone and matrix/interface interactive zone. By organically integrating the rigidity mixed law and flexibility mixed law, the model can predict the engineering elastic constant of a composite effectively.

A Composite Model Predictive Control Strategy for Furnaces

Tube furnaces are essential and primary energy intensive facilities in petrochemical plants. Operational optimization of furnaces could not only help to improve product quality but also benefit to reduce energy consumption and exhaust emission. Inspired by this idea, this paper presents a composite model predictive control(CMPC)strategy, which, taking advantage of distributed model predictive control architectures, combines tracking nonlinear model predictive control and economic nonlinear model predictive control metrics to keep process running smoothly and optimize operational conditions. The controllers connected with two kinds of communication networks are easy to organize and maintain, and stable to process interferences. A fast solution algorithm combining interior point solvers and Newton's method is accommodated to the CMPC realization, with reasonable CPU computing time and suitable online applications. Simulation for industrial case demonstrates that the proposed approach can ensure stable operations of furnaces, improve heat efficiency, and reduce the emission effectively.

A Simulation Software for the Prediction of Thermal and Mechanical Properties of Wood Plastic Composites

Modelling and simulation has become an important tool in research and development. Simulation models are used to develop better understanding of the internal properties and impact of various parameters on the final quality of the product or process. Simulation model reduces the number of experiments and saves the wastage of material, time and money and are widely used in automobile industry, aircrafts manufacturing, process engineering, training for military, health care sector and many more. Wood Plastic Composite （WPC） is a bio-composite made by mixing wood fibers and plastic granules together at high temperature by compression molding or injection molding. A large quantity of WPC is rejected due to poor quality and low mechanical strength. There is a need to improve the understanding of the wood plastic composites, with both theoretical and experimental analysis. The impact of various parameters and processing conditions on the final product is not known to the industry people, due to less simulation models in this field. A new simulation software WPC Soft is developed to predict the mechanical and thermal properties of WPC. The software can predict the mechanical and thermal properties of WPC. The simulation results were validated with the experimental results and it was observed that the predicted values are quite close to the experimental values and with the further refining of the model, prediction can be further improved. The present simulation software can be easily used by the industry people and it requires very little knowledge of computers or modeling for its operation.

A model for parameter estimation of multistage centrifugal compressor and compressor performance analysis using genetic algorithm

A model for performance prediction of multistage centrifugal compressor is proposed. The model allows the users to predict the compressor performance, e.g. pressure ratio, efficiency and losses using the compressor geometric information and speed by a stage stacking calculation based on the characteristics of each stage. To develop the compressor elemental stage charac- teristics, the compressor losses, such as incidence losses and friction losses, are mathematically modeled. For a composite sys- tems, for instance a gas turbine power plant, the performance of the multistage centrifugal compressor can be evaluated. Since some important parameters of the compressor model, e.g., the slip factor or, shock loss coefficient （and reference diameter DI, are hard to be determined by empirical laws, a genetic algorithm （GA） is used to solve the parameter estimation problem of the proposed model, and in turn the compressor performance analysis and parameters study are performed. The surge line for the multistage centrifugal compressor can also be determined from the simulation results. Furthermore, the model presented here provides a valuable tool for evaluating the multistage centrifugal compressor performance as a function of various operation parameters.