一种热线式空气流量传感器非线性静态模型辨识

根据在流量计标定试验台上对一种热线式空气流量传感器静态标定的数据,用解析建模法和数值建模法建立了热线式传感器的非线性静态模型.文中对比分析了几种模型的辨识精度.结果表明,基于多项式最小二乘和FLANN网络的模型基本相同,能够满足测量的要求,而基于BP网络的模型有着更高的辨识精度.     

汽车发动机热式空气质量传感器的动态特性

采用非线性建模方法研究了发动机热式空气质量流量（MAF）传感器的动态特性。基于静、动态标定数据,用两步辨识法建立了传感器的Hammerstein模型。建立多幅值激励下的模型时,充分考虑了激励信号源的延迟特性,并采用3种不同的方法对传感器的实际输入进行了估计。利用建立的模型分别计算了热线式和热膜式MAF传感器的时域和频域性能指标,并进行了分析和比较。结果表明,MAF传感器的动态响应特性与激励信号的大小有关,在发动机瞬态空燃比控制策略中应该采取措施补偿由于传感器变延迟特性引起的动态测量误差。     

汽油机进气道油膜模型参数辨识算法的研究

为寻求较为理想的油膜模型参数辨识算法,从随机误差相关性出发,分别推导最小二乘辨识和基于夏氏偏差修正法辨识的计算过程;并以燃油阶跃补偿标定试验为参考,比较了最小二乘辨识和基于夏氏偏差修正法辨识效果.结果表明考虑相关随机误差的夏氏辨识在辨识效果上明显优于最小二乘辨识,且比起受工况及温度影响较大的燃油阶跃补偿标定试验在数据离散程度方面有一定的优势.基于辨识算法对油膜模型参数的辨识结果相对而言更贴近真实模型参数值,为油膜模型参数辨识提供了理论依据.     

基于PMU实测小干扰数据的负荷闭环辨识方法

负荷建模是电力系统建模的重要部分,但是负荷模型的复杂性、非线性以及随机性增加了负荷建模的难度。提出了一种基于PMU实测小干扰数据进行负荷模型闭环辨识的方法。首先,建立了传递函数表达式下的负荷模型,并构建了应用于负荷模型辨识的闭环系统。而后提出基于预报误差法辨识负荷BJ模型的方法,通过数据预处理、模型阶次选择以及模型转换详细介绍该方法。最后,通过浙江电网220 kV华金变电站、太真变电站和清漾变电站的PMU实测数据的算例,验证了该方法在只有随机小干扰的情况下仍可以较为精确地辨识出负荷模型。     

模糊辨识方法在蓄电池建模中的应用

针对光伏系统中蓄电池建模十分困难的问题,基于T—S模糊模型,提出一种模糊建模方法。该方法采用三角形隶属函数计算给定样本的隶属度,利用稳态卡尔曼滤波器辨识模糊模型的结论参数,并把辨识模型的仿真结果与系统实验测量数据相对比,以检验模糊模型的可靠性。实验结果表明,这种新的蓄电池建模方法具有很高的精度,为高度复杂的非线性电化学过程的模型化提供了一条新途经。     

输电线路微风振动传感器设计

微风振动是架空线路时常出现的现象,长期振动会造成导线断股断线、金具脱落等危害。针对现有微风振动在线监测传感器测量误差较大的问题,设计了一种微风振动在线监测数字传感器,传感器在采用悬臂梁式位移计的基础上,将线性回归的方法和BP神经网络算法应用到传感器的非线性标定当中,并进行了传感器标定实验。研究表明,线性回归的标定算法比神经网络的标定算法精度更高,标定后最大相对误差为1.93%。根据研究结果可知,微风振动传感器由频率不同带来的非线性误差,可以采用线性回归的方法进行补偿,能够提高传感器的测量精度,从而能为导线状态检修提供更加可靠的参考依据     

热线式空气流量计在便携式尾气分析仪中的应用研究

介绍了一种广泛应用于桑塔纳发动机的热线式空气流量计在摩托车工况排放分析仪中的应用.介绍了热线式空气流量计的理论依据、基本原理和使用方面的问题,通过最小二乘法将标准数据进行拟合,作出拟合曲线从而对流量计进行精确标定,再通过标准测试,将流量计的输出结果与标准流量进行比较,从而表明热线式空气流量计在尾气分析仪中能够进行空气流量的准确测量.     

架空输电线路三维建模方法现状及展望

介绍了基于二维图纸建模,激光扫描建模和航测建模架空输电线路模型构建的几种方法,并做了详细的对比分析,结果表明:利用设计图纸建模法效率和模型精细度较高,但是数据量较大,不利于系统的管理,适用于新建线路;利用施竣工图纸建模法效率低下,资料较难获取,精度无法保证,适用于早期修建且资料完整的线路;基于激光雷达建模法精度和模型精细度较高,但是作业强度大、成本高,适用于小面积、精度要求较高的平原地区线路及变电站;基于航测建模法可以制作已有线路走廊通道和线路本体三维模型数据,效率和精度较高,可以满足运维管理的需求,同时也可作为线路净空距离分析等各种深化应用分析的第一手数据,值得推广应用。     

基于ROLS算法的RBF神经网络燃料电池电特性建模

提出了一种基于ROLS算法的RBF神经网络辨识建立直接甲醇燃料电池(DMFC)电特性模型的新方法。以电池的工作温度为输入量,电池的电压／电流密度为输出量,利用1200组实验数据作为训练和测试样本,建立了在不同工作温度下,电池的电压／电流密度动态响应模型,仿真结果表明采用RBF神经网络辨识建模的方法是有效的,建立的模型精度较高。     

基于T-S模型的质子交换膜燃料电池控制建模

对PEMFC非线性复杂被控对象,提出了一种在线辨识模糊预测算法,用模糊聚类和线性辨识方法在线建立PEMFC控制系统的T—S模糊预测模型,仿真实验结果表明了该模糊辨识建模方法具有建模简单、模型精度高等优点,亦证明了该算法的有效性和优越性。研究结果对质子交换膜燃料电池控制系统的建模和控制具有一定的实用价值。     

Artificial neural network modelling and experimental verification of the operating current of mono-crystalline photovoltaic modules

This article presents the artificial neural network modelling of the operating current of a 120 Wp of mono-crystalline photovoltaic module. As an alternative method to analytical modelling approaches, this study uses the advantages of neural networks such as no required knowledge of internal system parameters, less computational effort and a compact solution for multivariable problems. Generalised regression neural network model is used in the present article to predict the operating current of the photovoltaic module. To show its merit, the current predicted from the artificial neural network modelling is compared to that from the analytical model. The five-parameter analytical model is drawn from the equivalent electrical circuit that includes light-generated current, diode reverse saturation current, and series and shunt resistances. The operating current predicted from both the neural and analytical models are compared to the measured current. Results have shown that the artificial neural network modelling provides a better prediction of the current than the five-parameter analytical model.     

Bayesian inference and uncertainty quantification for hydrogen-enriched and lean-premixed combustion systems

Development of probabilistic modelling tools to perform Bayesian inference and uncertainty quantification (UQ) is a challenging task for practical hydrogen-enriched and low-emission combustion systems due to the need to take into account simultaneously simulated fluid dynamics and detailed combustion chemistry. A large number of evaluations is required to calibrate models and estimate parameters using experimental data within the framework of Bayesian inference. This task is computationally prohibitive in high-fidelity and deterministic approaches such as large eddy simulation (LES) to design and optimize combustion systems. Therefore, there is a need to develop methods that: (a) are suitable for Bayesian inference studies and (b) characterize a range of solutions based on the uncertainty of modelling parameters and input conditions. This paper aims to develop a computationally-efficient toolchain to address these issues for probabilistic modelling of NOx emission in hydrogen-enriched and lean-premixed combustion systems. A novel method is implemented into the toolchain using a chemical reactor network (CRN) model, non-intrusive polynomial chaos expansion based on the point collocation method (NIPCE-PCM), and the Markov Chain Monte Carlo (MCMC) method. First, a CRN model is generated for a combustion system burning hydrogen-enriched methane/air mixtures at high-pressure lean-premixed conditions to compute NOx emission. A set of metamodels is then developed using NIPCE-PCM as a computationally efficient alternative to the physics-based CRN model. These surrogate models and experimental data are then implemented in the MCMC method to perform a two-step Bayesian calibration to maximize the agreement between model predictions and measurements. The average standard deviations for the prediction of exit temperature and NOx emission are reduced by almost 90% using this method. The calibrated model then used with confidence for global sensitivity and reliability analysis studies, which show that the volume of the main-flame zone is the most important parameter for NOx emission. The results show satisfactory performance for the developed toolchain to perform Bayesian inference and UQ studies, enabling a robust and consistent process for designing and optimising low-emission combustion systems.     

基于混沌海鸥优化算法的含光伏发电系统负荷模型参数辨识

针对分布式光伏发电系统广泛接入配电网,导致电力系统潮流计算速度和精度较低的问题,提出一种基于混沌海鸥优化算法的含光伏发电系统负荷模型参数辨识模型。首先,在综合负荷模型的虚拟母线上接入等效光伏发电系统的负荷模型,从而建立配电网广义负荷模型;之后,提出一种将混沌优化与海鸥优化相结合的优化算法,基于该算法完成配电网的等值,并在此基础上进行含光伏发电的综合负荷模型参数辨识。最后,通过仿真表明该文提出的算法,相比于传统的粒子群算法和单一海鸥优化算法,在计算精度和收敛速度等方面具有优越性,并可应用于负荷模型的参数优化。     

面向次同步振荡分析的直驱风电机群建模

针对低运行工况下直驱风电机群并网出现的次同步振荡(SSO)问题,基于图形化分块建模,在建立风力机、集电线路及电网等独立元件的模型基础上,构造不同集电线路拓扑结构下直驱风电机群的线性化模型.利用特征值分析法和阻抗分析法识别系统中存在的振荡模式,并分析运行工况、电流内环控制参数和风力机数量等因素对SSO频率和阻尼的影响.研...     

基于区域同步特性的小水电出力随机建模及其准入容量评估

为准确方便地计算小水电的准入容量,首先考虑负荷及小水电出力的随机特性,建立了相应的随机模型;然后针对不同地区气象条件的差异性,提出区域同步特性概念,利用拟合精度较高的非参数核密度估计法作为参考,对水电出力进行常规分布类型建模,并以节点电压不越限为约束条件,得到不同负荷分布下的准入容量表达式,该表达式考虑了模型的差异性,引入修正系数;最后以实际运行配电网为算例,利用所提方法并结合随机潮流进行仿真计算,从而验证了算法的合理性。     

A Review on solid oxide fuel cell models

Since the model plays an important role in diagnosing solid oxide fuel cell (SOFC) system, this paper proposes a review of existing SOFC models for model-based diagnosis of SOFC stack and system. Three categories of modelling based on the white-, the black- and the grey-box approaches are introduced. The white-box model includes two types, i.e. physical model and equivalent circuit model based on EIS technique. The black-box model is based on artificial intelligence and its realisation relies mainly on experimental data. The grey-box model is more flexible: it is a physical representation but with some parts being modelled empirically. Validation of models is discussed and a hierarchical modelling approach involving all of three modelling methods is briefly mentioned, which gives an overview of the design for implementing a generic diagnostic tool on SOFC system.     

基于ARMAX-GARCH模型的微电网风功率预测

目前风功率预测多为风功率期望的点预测,且以采样间隔较大的功率序列作为建模序列,这样会降低预测模型对风功率时序特征模拟的准确度和可信度。文中基于小采样间隔风功率序列,提出ARMAX-GARCH风功率预测模型。通过构造风功率新息序列,结合小时平均风功率序列,建立ARMAX点预测模型,采用BIC最小信息准则和相关性分析实现模型定阶和外生变量选择;采用GARCH模型模拟残差的波动特性实现区间预测。以海岛微电网实测风功率数据为例,进行提前1 h风功率预测。结果表明,与持续法、ARMA和RBF神经网络相比,该预测模型能显著提高风功率期望的点预测精度并具有较好的区间预测效果。     

Intelligent soft-computing based modelling of naturally ventilated buildings

The paper presents recent results on the application of the soft computing methodology for modelling of the internal climate in office buildings. More specifically, a part of a recently completed naturally ventilated building is considered which comprises three neighbouring offices and one corridor within the Portland Building at the University of Portsmouth. The approach adopted uses fuzzy logic for modelling, neural networks for adaptation and genetic algorithms for optimisation of the fuzzy model. The fuzzy models are of the Takagi-Sugeno type and are built by subtractive clustering. As a result of the latter, the initial values of the antecedent non-linear membership functions and the consequent linear algebraic equations parameters are determined. A method of extensive search of fuzzy model structures is presented which fully explores the dynamics of the plant. The model parameters are further adjusted by a back-propagation training neural network and a real-valued genetic algorithm in order to obtain a better fit to the measured data. Results with real data are presented for two types of models, namely Regression Delay and Proportional Difference. These models are applied for predicting internal air temperatures.     

Torsional system parameter identification of turbine-generator sets

Accurate low order linear models that represent the torsional motion of turbine-generator sets are needed for determining shaft torsional responses resulting from subsynchronous resonance conditions, electric system faults and planned/unplanned switching actions in the electric network. This paper outlines the theoretical background and the methodology used for identification of linear state-space models of turbine-generator systems. These analytic mass-spring-damper models are lumped-parameter approximations, which in reality represent a continuous nonlinear system. For transient torque studies these models are adequate representations of the torsional dynamics of interest. Reduced analytic models of any particular turbine-generator unit, however, usually do not match precisely the behavior of the real machine. The paper describes an optimization method that can give a more precise representation of a particular turbine-generator based on actual plant tests and an assumed model of that unit. The parameter identification process is illustrated using plant test data from a 618 MVA turbine-generator unit     

A comparison between state-of-the-art and neural network modelling of solar collectors

The state-of-the-art modelling of solar collectors as described in the European Standard EN 12975-2 is based on equations describing the thermal behaviour of the collectors by characterising the physical phenomena, e.g. transmission of irradiance through transparent covers, absorption of irradiance by the absorber, temperature dependent heat losses and others. This approach leads to so called collector parameters that describe these phenomena, e.g. the zero-loss collector efficiency η₀ or the heat loss coefficients a₁ and a₂.Although the state-of-the-art approach in collector modelling and testing fits most of the collector types very well there are some collector designs (e.g. “Sydney” tubes using heat pipes and “water-in-glass” collectors) which cannot be modelled with the same accuracy than conventional collectors like flat plate or standard evacuated tubular collectors. The artificial neural network (ANN) approach could be an appropriate alternative to overcome this drawback.To compare the different approaches of modelling investigations for a conventional flat plate collector and an evacuated “Sydney” tubular collector have been carried out based on performance measurements according to the European Standard EN 12975-2. The investigations include the parameter identification (training), the comparisons between measured and modelled collector output and the simulated yearly collector yield for a solar domestic hot water system for both models.The obtained results show better agreement between measured and calculated collector output for the artificial neural network approach compared with the state-of-the-art modelling. The investigations also show that for the ANN approach special test sequences have to be designed and that the determination of the ANN that fits the thermal performance of the collector in the best way depends significantly on the expertise of the user.Nevertheless artificial neural networks have the potential to become an interesting alternative to the state-of-the-art collector models used today.