基于灰色预测分析的船用低速柴油机轴承磨损研究

基于某型船用低速柴油机安装的XTS-W轴承监控系统的监测数据,建立了船用低速柴油机轴承磨损灰色系统预测分析模型.采用该模型对轴承磨损量进行预测,并将预测值与系统监测值对比.结果 显示:在不充分满足灰色预测分析模型要求的情况下,部分监测点的预测精度仍较高,表明所建立的预测分析模型可用于预测船用低速柴油机轴承的磨损状况.     

基于BP神经网络的输油管道能耗预测方法

为准确预测输油管道运行过程中的能源消耗,通过分析输油管道运行报表,确定评价输油生产过程中的主要能耗指标,利用BP神经网络,建立了输油管道能耗预测模型。用某输油管道运行能耗数据做样本,对所建立的能耗预测模型进行训练与测试。预测结果表明,该模型能准确地预测输油管道的能耗,预测平均相对偏差不超过4％。该方法为输油管道能耗预测提供了一种新思路,有较高的使用价值。     

基于灰色缓冲算子-卡尔曼滤波双修正的风电功率实时预测研究

文章针对风电功率预测的精度问题,提出了一种基于灰色缓冲算子-卡尔曼滤波双修正的神经网络预测模型。通过建立灰色缓冲算子,消除数据中的冲击扰动,还原其本身的特性;对缓冲之后的数据进行训练,建立人工神经网络进行滚动预测;利用卡尔曼滤波器对预测结果进行二次修正,消除预测结果中的奇异点。仿真结果表明,与传统的预测方法相比,该模型可有效地提高预测精度。     

基于ARMA模型的风电场风速短期预测

通过分析达坂城风电场风速数据并建立ARMA模型,基于时间序列分析法实现了提前1h风速预测,分析预测结果证明预测时间和风速震荡性是影响风速预测精度的主要因素,为更长时间的风速预测提供理论基础。     

固体废弃物气化处理半经验模型研究

对固体废弃物的气化处理半经验模型进行了研究.根据固体废弃物气化处理的特点,提出了以质量方程和能量方程为基础的气化处理模型.运用相似理论和因次分析方法建立了固体废弃物气化产气成分的浓度预测公式,建立了气化过程的准则数半经验预测模型.对不同实验工况下的固体废弃物的实验数据进行回归分析,拟合得到了较好的预测效果,回归误差和预测误差均较小.为建立固体废弃物气化模型进行了积极的探索.     

输变电工程造价智能评估研究

面对输变电工程越来越高的成本,有必要加强对输变电工程成本在合理范围内的管控。为了达到节约建筑资源、降低成本、提高效率的目的,已成为现阶段必须解决的关键问题。目前,输变电工程的工程设计和估算过程缺乏预测功能,为了解决这一问题,本文建立了基于神经网络的输变电工程成本智能预测智能模型,建立了成本体系,并运用主成分分析方法提取了输变电工程成本的主要影响因素。通过测试验证,建立的智能智能模型预测结果更加准确,可以合理预测输变电工程的成本。     

基于方法组合创新思想的电力系统长期负荷预测

针对建立准确、有效的负荷预测模型是长期电力系统负荷预测的关键,基于灰色预测、趋势预测和二次指数平滑法等传统的定量负荷预测方法,引入方法组合创新思想,根据最小二乘原理采用Eviews软件进行变量分析及变量类比,剔除异常值并建立拟合曲线以预测电力负荷,并以供电企业的实际售电量为算例,验证了该方法的可行性和有效性.     

支持向量机回归模型在径流预测中的应用

为了探索提高径流中长期预测精度的有效途径,尝试建立了基于支持向量机的径流预测模型,并应用于西江流域梧州站的年、月径流预测中,取得了很好的效果。并与神经网络预测进行对比,结果表明该模型的预测精度要高于人工神经网络模型。     

中国小水电开发利用Logistic预测模型

基于中国大陆1981～2010年小水电装机容量的变化特点,采用加速遗传算法建立了中国小水电开发利用Logistic预测模型,运用自助法对Logistic预测模型的误差进行重抽样,建立了基于自助法的中国小水电开发利用Logistic预测模型(LP-SHD),并预测了2015、2020年中国小水电装机容量95%的置信区间.实例结果表明,LP-SHD模型物理概念清晰、计算简便、预测精度高,在其他可再生能源资源开发利用预测中具有推广应用价值.     

基于改进递归深度信念网络的CSP电站短期出力预测

以预测CSP电站短期出力为目的,首先引入自适应思想对递归深度信念网络的训练算法进行改进,并建立直接法向辐射的短期预测模型。随后提出一种结合静态模型的CSP电站短期出力预测方法。最后进行性能检验,验证了改进递归深度信念网络的可行性,以及CSP电站短期出力预测方法的有效性。研究结果表明：建立的改进递归深度信念网络可提升预测准确性和收敛速度;提出的CSP电站短期出力预测方法可较为准确地预测其短期出力情况。     

数字式异步电机转矩转速测量装置的研制

在异步电机高性能矢量控制或电机测试系统中,为更好地研究异步电机控制方案对电机转矩、转速等参数的采集、存储、显示,研究了一种用于实验室基于软件技术的改进型数字式异步电机转矩转速测量装置,通过对异步电机基本参数测量即可实现转矩转速的测量.实验结果表明,该转矩转速测试装置精度高、抗干扰能力强、操作简单,大大提高了测量装置的可靠性和寿命,降低了测量装置的成本.     

A new energy recovery double winding cage-rotor induction machine

This paper proposes a new double winding induction machine and its speed control methods. The machine consists of two stator windings and one cage rotor. One stator acts as a motor and the other as a generator. By controlling the voltage supplied to the secondary or the generator winding, the rotor speed can be adjusted. The machine has a similar speed control characteristic to that of a slip-ring induction motor equipped with the rotor energy recovery scheme. The construction, principle, equivalent circuit, and speed control schemes of the new machine are presented. The performance characteristics of the machine are analyzed using the equivalent circuit and verified by experimental results.     

Modeling and control of a wind turbine driven doubly fed induction generator

This paper presents the simulation results of a grid-connected wind driven doubly fed induction machine (DFIM) together with some real machine performance results. The modeling of the machine considers operating conditions below and above synchronous speed, which are actually achieved by means of a double-sided PWM converter joining the machine rotor to the grid. In order to decouple the active and reactive powers generated by the machine, stator-flux-oriented vector control is applied. The wind generator mathematical model developed in this paper is used to show how such a control strategy offers the possibility of controlling the power factor of the energy to be generated.     

Micromachine for real-time generator simulation design limits and anovel solution with adjustable parameters

The authors give limits on dimensional scaling in assessing the same transient and steady-state performance of the down-scaled micromachine as that of the full-size machine. This system has the ability to model various types of synchronous and asynchronous machines in real time by means of one machine and to simulate the effects of various types of field and damper windings in the rotor without having a damper cage in the physical model of the machine. By means of this system, new types of machine control such as two-axes control and supplementary damping control can be studied. Test and identification results are given which compare favorably with a chosen set of characteristic time constants     

系统弯管机电气系统开发和应用

这台系统弯管机是我厂具有国际先进水平的数控大型成套设备,该设备为数控机床,通过对设备的多年使用,发现电液比例控制阀效率高,抗污染能力强,在结构与成本上具有明显优势,能够满足我们的使用条件,因此在此次改造过程中,把原有的电液伺服阀改为电液比例控制阀,降低设备的故障率,同时为了能更充分地发挥机床的巨大威力,选择西门子公司新...     

感应淬火技术在汽车零件制造中的应用

传统的低层次的PLC控制淬火机床机械动作和单一时间控制模式难以保证结构复杂的汽车零件淬火质量，而计算机软硬件智能化思想改变了传统的工业控制面貌，使我国淬火装置水平达到了一个全新的境界。     

基于支持向量机的过热汽温内模控制

提出一种基于支持向量机回归的内模控制算法并应用于过热气温控制。首先,简介支持向量机回归算法和内模控制原理;其次,提出基于支持向量机回归算法构造的被控过程的逆模型,将该逆模型作为控制器,并在主控系统中增加鲁棒控制器以增加系统的鲁棒性及跟踪性能。仿真结果表明提出的控制算法的控制品质优于传统的PID控制算法。     

PV-battery energy storage system operating of asynchronous motor driven by using fuzzy sliding mode control

This article tends on the designing of renewable-storage electrical system for asynchronous machine application. The machine requires a photovoltaic energy as a primary source of renewable energy. Such as the major drawback of solar sources are neither continuous nor regular in the time. To overcome this problem, a battery energy storage system will be added to make sure the continuity of the operation machine. Generally, electrical machine is driven by using sliding mode control where the major problem in this control technique design is chattering phenomenon which can be defeated using a sliding surface based on fuzzy logic. The proposed technique is performed with the proposed control system. A simulation test of the system for varied load motor conditions proved that the system is preferment. When, the generated power from PV is superior that the demands, the rest is stored in the battery. In the opposite condition, PV and battery required load. Moreover, the system which consists of machine and its control is enabling to produce electrical energy when it is functioning in the other sense of rotation. Four quadrant machine operations guarantee many possibilities of charging and discharging of the battery that assure the continuity of operating system.     

Implementation of indirect vector control on an integrated digitalsignal processor-based system

This paper presents an implementation of indirect vector control of an induction machine on an integrated DSP (digital signal processor) system manufactured by dSPACE GmbH. The system integrates into a single board the computational power of a TMS320C31 DSP with extra peripherals needed in vector control application, and therefore requires minimal hardware development. The induction machine parameters required for the vector control operation are obtained through an offline parameters identification using the maximum likelihood estimation technique with DC voltage source excitation. It is shown through extensive experimental study that the offline identified parameters yield in a reliable field orientation of the induction machine     

Effectiveness of active noise and vibration cancellation for switched reluctance machines operating under alternative control strategies

The effectiveness of active vibration cancellation by inducing antiphase vibration of the stator to reduce the acoustic noise emitted from a switched reluctance machine is evaluated under typical operating modes, viz., single pulse excitation, pulsewidth modulation (PWM) voltage control, and PWM current control. Measurements in both the frequency and time domains are complemented by sound pressure level measurements. It is shown that active vibration cancellation is most effective in reducing a single mode of vibration when a switched reluctance (SR) machine is operated under single-pulse excitation, and becomes less effective when more than one dominant vibration mode exists within the audible frequency range. In general, it also works relatively well when the machine is operated under fixed frequency PWM voltage control, although less effectively than for single pulse control. Further, it is shown that the technique is ineffective when the machine is operated under PWM current control since the duration of the zero voltage period varies significantly from the optimal value due to the random nature of the PWM.