用多变量模糊神经网络实现双抽汽轮机热电负荷的解耦控制

由于经典控制理论和现代控制理论分别在双抽汽轮机电、热负荷协调控制应用中的不足，提出了用模糊神经网络控制来实现解耦控制。以15MW双抽汽轮机为例，建立系统动态数学模型，并进行了模糊神经网络解耦控制的数字仿真。     

低真空抽凝式汽轮机排汽供热的研究及其改造

抽凝式汽轮机在无抽汽热负荷或抽汽热负荷很少时进行低真空供热,经改变压力级和速度级的喷嘴数后,在同样进汽条件下．出力可大幅度提高。     

汽轮机压缩机组的解耦控制

该文介绍采用电液比例控制技术的汽轮压缩机组调节系统，讨论了汽轮机蒸汽阀门电液控制单元及汽轮机抽汽压力与压缩机出口压力的解耦控制方法，并证明了采用前馈解控制的汽轮压缩机组调节系统对传感器失效具有一定的容错能力。     

350MW容量等级双抽汽轮机的技术创新

为满足节能环保,减少CO2排放,发展热电联供发电的要求,上海汽轮机厂最新推出了350MW容量等级的双抽汽汽轮机系列。本文论述了调整抽汽设计的关键,以及在开发双抽汽模块设计中,在通流部分、结构、配汽机构等方面所采取的优化技术和创新。结果表明,该系列产品是安全、可靠、高效的,具有广阔的市场前景。     

基于模糊神经网络的发电用单轴燃气轮机的解耦控制

针对燃气轮机这样大型复杂且具有强非线性多变量耦合的动力系统，提出用模糊隶属函数型神经网络与模糊控制融合的解耦方法(DMFFCNN)进行解耦控制，从而克服了现代控制理论对解耦系统类型的限制。试验表明，这种方法是有效的，为机组运行时的静态优化设计的实现奠定了基础。     

抽凝式和凝汽式汽轮机改为背压机的优势

目前我国部分热电厂拆除原有的凝汽式或抽凝式汽轮机,采用背压式汽轮机,以期提高机组运行效率及经济性。背压机组因没有低压回热加热系统,只能采用高压排汽在除氧器中加热补水,造成的热损失较大。因而考虑设计一种新型背压机,它是利用在机内做完功后的低压回热抽汽及排汽来加热补水。方案是把原有的抽凝式、凝汽式机组改造为既可向外部供气,又可用低压回热抽汽及排汽加热补水的机组。因凝汽器被改造为第一级补水加热器,没有冷端放热损失,成为了一种新型背压机,它不仅能节省大量投资,而且能降低热耗,提高机组能效,是一种既可行又经济的改造方法。     

双抽汽汽轮机数字电液控制系统的开发及其应用

介绍了双抽汽汽轮机数字电液控制系统的开发及其应用。它将为哈尔滨汽轮机有限责任公司发展双轴（也包括单抽）汽汽轮机起到了推波助澜的作用。汽轮机及其控制系统必须由汽轮机制造厂家一起向用户配套提供，以保证汽轮机顺利地起动和安全可靠地运行。     

单元机组协调系统的神经网络解耦控制仿真研究

神经网络控制是一种新颖的智能控制方法。应用神经网络技术,对难以精确建模的复杂非线性对象进行神经网络模型辨识作为控制器,提高了响应的快速性和准确性,可满足工业过程所提出的安全性、可靠性与易实现性的要求。大型火力发电单元机组协调控制系统是一个相对复杂的多变量控制系统,控制对象具有大时滞、时变、非线性、强耦合的特点,传统的PID控制算法很难实现对过程参数的良好跟踪和理想的控制效果。针对单元机组协调控制系统的特点将神经网络解耦控制应用于单元机组协调控制系统中,仿真实验表明,神经网络解耦控制具有较强的适应性和较高的控制精度,提高了负荷的响应速率,控制效果优于传统的PID控制算法。     

双抽汽式汽轮机的动态数学模型和仿真研究

本文推导了双抽汽式汽轮机的线性和非线性动态数学模型.肯定了目前国内汽轮机厂在设计双抽式汽轮机控制系统时所用的线性动态数学模型的可用性,同时分析了液压控制系统中非线性因素对机组控制系统品质的不良影响.     

基于对角递归神经网络整定的PID解耦单元机组负荷控制系统

针对火电厂单元机组具有多变量强耦合、非线性及参数时变的受控对象，提出了基于对角递归神经网络整定的PID解耦控制方法，其主要特点是能够提供一个对角递归神经网络来辩识系统模型，进而对PID控制器参数进行整定，实现多变量解耦控制。通过对火电机组负荷控制系统的设计和仿真研究，表明系统达到了动态近似解耦、静态完全解耦和无静差跟踪，并具有响应速度快，鲁棒性好等特点。图5参6     

Development of a need-oriented steam turbine cycle simulation toolbox

We call an electric generation mechanism, whose working fluid is water, as a steam turbine cycle or Rankine cycle. A steam turbine cycle has been adopted in fossil-fuel or nuclear power plants because it is suitable for large-scale and continuous operation. To check the performance level of a steam turbine cycle, performance tests are carried out according to the performance test codes (PTCs) provided by the authorized institute. However, authors found that it was too difficult to follow the PTCs in actual plants. The reason is signal reliability and the prerequisites of the PTCs. The effort to overcome these shortcomings already started and some commercial solutions were also developed. These solutions include a general-purpose simulation code with signal validation techniques. Despite these achievements, performance engineers are suffering from the inconsistency between their previous performance analysis work and new performance analysis tools. This study proposed the solution, a need-oriented turbine cycle simulation toolbox, which is another general-purpose simulation code with the signal validation method based on multivariate statistics. This toolbox removed the inconsistency by customizing the actual needs of performance engineers. The accuracy of the developed simulation code was validated with other commercial turbine cycle simulation codes.     

汽轮机功能可靠性仿真研究的新方法

以电站汽轮机为例，研究探讨了设备的功能可靠性仿真方法，根据该设备的性能退化规律，建立了其性能退化模型，开发了功能可靠性仿真算法，并对其进行了功能可靠性仿真。计算得到的设备可靠性数据符合同类设备的规律，算法可用于同类设备及复杂热力系统的功能可靠性仿真计算。     

基于改进共轭梯度优化BP神经网络的风电机组变桨距控制

根据共轭梯度算法和传统BP神经网络的变桨距控制器的原理,针对兆瓦级风电机组变桨距控制设计了一种改进共轭梯度优化BP神经网络的变桨距PID参数自整定控制器,此控制器采用改进共轭梯度法修正BP神经网络的权值和阈值,实现BP神经网络变桨距PID控制器的在线整定。在Matlab/Simulink中仿真,仿真结果表明,采用此变桨距控制器可以在额定风速之上快速响应,在相同风速状况下使发电机桨距角调节命令更加准确,风轮转速更加平稳,输出功率维持在额定功率附近,取得了很好的变桨距控制效果。     

Impact of shaft torsionals in steam turbine control

Shaft torsional vibrations are discussed along with their effect on steam turbine-generator control following severe supply-network disturbances and the destabilization of shaft torsional modes through the action of high-speed electrohydraulic controllers. It is shown that shaft flexibility can exert a significant influence on steam turbine-generator response following a severe supply-system disturbance, particularly when the turbine has nonlinear valve stroking and fast valving. The effect can be minimized by the careful location of a speed sensor along the turbine shaft and by filtering speed or acceleration signals to reduce the speed input of troublesome low-frequency torsional vibrations to an insignificant level. The effect of shaft torsionals on control system response to an islanding situation and on the above mentioned destabilization is evaluated. Digital implementation of steam-turbine control systems in relation to quantization, sampling, response time in protection systems, and software integrity is examined     

Nonlinear model identification of wind turbine with a neural network

A nonlinear model of wind turbine based on a neural network (NN) is described for the estimation of wind turbine output power. The proposed nonlinear model uses the wind speed average, the standard deviation and the past output power as input data. An anemometer with a sampling rate of one second provides the wind speed data. The NN identification process uses a 10-min average speed with its standard deviation. The typical local data collected in September 2000 is used for the training, while those of October 2000 are used to validate the model. The optimal NN configuration is found to be 8-5-1 (8 inputs, 5 neurons on the hidden layer, one neuron on the output layer). The estimated mean square errors for the wind turbine output power are less than 1%. A comparison between the NN model and the stochastic model mostly used in the wind power prediction is done. This work is a basic tool to estimate wind turbine energy production from the average wind speed.     

Numerical simulation of a steam pipeline network

The present paper describes a computer program that can simulate the behaviour of a network of steam pipelines, such as those used in Larderello geothermal field, fed by superheated steam. Also given are some applications of this program to a network in operation.     

A decoupling control strategy for wind turbine blades equipped with active flow controllers

The use of active controls has shown to be of substantial help in supporting the increasing size of wind turbines by reducing peak stresses and fatigue loads. In this respect, this paper proposes the use of intuitive frequency‐based control strategies for reducing loads in wind turbine blades equipped with multi‐input multi‐output (MIMO) active flow controllers. For that purpose, a loop‐shaping approach is considered for analysing the dynamic of actively controlled wind turbine blades. Preliminary aeroelastic simulations are carried out to validate the results. It is shown that the MIMO vibration control problem can effectively be decomposed into a number of decoupled single‐input single‐output control problems because of the strong correlation between the dominant aeroelastic blade dynamics and actuator deployments. As a result, it is demonstrated that classical single‐input single‐output control systems can perform as efficiently as MIMO controllers for damping the aeroelastic dynamics of wind turbine blades. Copyright © 2016 John Wiley & Sons, Ltd.     

基于BP神经网络的温度控制系统

文中介绍了基于BP(Back Pmpagation)的神经网络气化炉温度控制系统。对BP神经网络控制算法作了详细的介绍，运用模糊逻辑控制概念赋予隐层含义，并决定其节点数，同时用高斯核函数作为节点激励函数，并做了仿真研究，叙述了系统的硬件与软件构成，试验表明所设计的系统操作方便、安全可靠，所选择的控制算法适应性强，控制效果良好。     

永磁同步风电机组神经网络滑模多目标优化控制

将永磁同步风力发电机组中的变流器和电网用等效负载代替并对控制回路进行简化,得到非线性仿射形式的机组模型,利用反馈线性化方法对系统进行精确线性化。固定参数离散指数趋近律滑模控制算法主要缺陷是如两个参数匹配不当,可能会使求得的控制量过大,同时系统在滑模面附近剧烈的高频抖振会导致机组所要承受的机械应力增加,动态性能变差,利用神经网络的自适应学习能力对这两个控制参数进行实时优化,根据机组控制目标定义一个综合性能指标,通过优化该指标得到网络权值修正算法。仿真结果表明,该方法可以使系统快速到达滑模面,实现了机组对最优转速的快速跟踪;同时有效抑制了系统的抖振,减小了额外的疲劳载荷,实现了多目标优化控制。     

50MW抽汽冷凝式汽轮机叶片断裂原因分析

阐明了50MW抽汽冷凝式汽轮机转子动叶片对汽轮机组安全稳定运行的重要性、分析了引起汽轮机叶片断裂的多种原因，根据上海石化热电一站50MW抽汽冷凝式汽轮机在运行工况下叶片断裂的症状，以及在汽轮机检修过程中对叶片断裂截面的剖析，着重分析了热电一站汽轮机历年叶片断裂的原因，认为提高汽轮机叶片设计可靠性，加强汽轮机叶片的特性监控，以及控制运行中汽轮机中压缸的蒸汽流量，是防止50MW抽汽冷凝式汽轮机叶片断裂的主要手段。