基于GRU神经网络的光伏电站数据预处理方法

光伏电站数据为时间序列数据,会受到通信传输、逆变器采集等因素的影响而包含大量异常数据,故该文研究一种基于深度学习的光伏电站数据预处理算法,进行数据清洗等预处理。一方面,根据组串逆变器的工作特性,对光伏电站数据的常见异常类型进行分析标记,结合滑动窗口法划分数据,构建用于深度学习训练的光伏电站数据集。另一方面,从激活函数、损失函数以及隐藏层等方面优化GRU神经网络模型,并利用自建数据集对该模型进行训练和测试。测试结果表明：该模型在实际光伏电站数据上的处理准确率达99.84%。     

基于小波BP神经网络的风电机组变桨系统故障预测

变桨故障是风电机组重要的停机故障之一,对变桨系统进行故障预测并提高预测精度,是风电开发的关键技术,不但保证电网安全运行而且减少运维成本。分析处理SCADA系统数据,提取相关联参数,即输出功率、风速、桨距角和转子转速。采用BP神经网络对系统进行模型训练,考虑到风电机组参数具有波动性、不确定性等,同时采用小波BP神经网络进行模型训练。建立变桨故障预测模型,预测未来15 d的变桨系统运行情况,用于制定合理的运维方案。通过MATLAB系统仿真研究,对比分析了预测模型性能指标、误差指标和输出数据图形,小波BP神经网络训练预测模型诊断精度比BP神经网络提高了17%,可信率提高了18%,诊断能力提高了15.4%,诊断误报率降低了17%。     

利用神经网络估算太阳辐射

太阳辐射是一项对太阳能利用，建筑能耗分析和农业等十分重要的气象数据，本文建立了日总太阳辐射月均值的神经网络估算模型，在此基础上利用北京市1971年至1995年的气象数据资料对神经网络进行了训练，用1996至2000年的数据对神经网络的估算进行了检验，并与其它经验模型的估算结果进行了对比，结果表明神经网络的估算结果与实测值吻合的较好，并且精度高于其它经验模型。因此利用神经网络来估算太阳辐射具有很好的应用前景。     

数据驱动的潮流能水轮机叶片翼型性能保真替代计算模型对比研究

在对翼型Bezier-PARSEC参数化软件FanOpt的应用方式改进后建立数据集,以翼型升阻比特性为目标,分别利用支持向量回归（SVR）、决策树、随机森林回归、全连接神经网络、一维卷积神经网络等机器学习模型进行拟合,训练模型比较拟合精度。结果表明,全连接神经网络、一维卷积神经网络作为替代计算模型在测试集上对升阻比的预测准确率可达97.86%,但相比于一维卷积神经网络,全连接神经网络在处理这种结构不复杂的数据集时更有优势。     

基于BP神经网络的GD-1高压共轨系统的建模研究

基于GD-1高压共轨燃油喷射系统,运用BP神经网络理论对GD-1系统高压油泵及共轨管进行建模,在Matlab平台上利用实际测得的数据对所建的神经模型进行训练,利用Simulink工具将训练好的高压油泵及共轨管模型与GD-1控制策略连接在一起进行闭环仿真,仿真结果表明设计的神经网络能很好地模拟共轨管内实际油压变化.     

PID补偿神经网络逆控制在超临界机组过热汽温控制中的应用

考虑超临界锅炉过热汽温系统的非线性、大惯性、大时延等特性,建立了过热汽温喷水减温系统的非线性动态神经网络逆模型,运用机组的历史运行数据对模型进行训练与校验。以训练好的模型为基础,构建了具有PID补偿环节的神经网络逆控制器,在MATLAB平台编制了实时控制程序。借助600 MW超临界机组全仿真系统,对过热汽温进行设定值扰动、大范围变工况扰动等仿真试验。结果表明:具有PID补偿的神经网络逆控制方案可有效降低动态变负荷过程中过热汽温的控制偏差,缩短汽温控制的稳定时间,与机组原控制方案相比具有更好的控制效果。     

基于灰色理论与神经网络的柴油机相继增压系统故障预测与诊断(英文)

增压柴油机是一个具有不确定性的复杂系统,灰色理论是一种处理不确定性的理论.为了保证柴油机的可靠运行,使用灰色预测理论与神经网络技术对柴油机相继增压系统进行了故障预测与诊断.采用MAT-LAB语言编制了灰色预测程序并训练了神经网络,训练好的神经网络具有良好的泛化性.采用GM(1,1)模型对相继增压柴油机运行参数进行预测,对于波动数据,使用改进GM(1,1)模型进行预测,预测参数与试验数据的相对误差均在5%以内.预测参数送入已培训好的神经网络中,对柴油机相继增压系统可能发生的故障进行预测与诊断.     

基于神经网络的锅炉对流受热面灰污监测研究

采用多层前向型神经网络,对电站锅炉对流受热面的实时污染状况建立了监测模型。模型选取合适的参数组成输入向量,利用电站数据采集系统下载的实时机组数据,经规格化处理后对神经网络进行训练。结果表明,训练后的神经网络可以较准确地实现锅炉对流受热面的积灰状态的在线监测,为吹灰方案的最优化打下了良好的基础。     

基于注意力机制的光伏热斑识别

为解决光伏的红外热图像含有大量噪声且不同状态红外图像分布不均衡导致的热斑难以识别的问题,以Vision Transformer(ViT)模型为基础,利用卷积神经网络改进模型特征提取,利用紧凑多头自注意力机制改进模型结构,提出一种光伏红外图像热斑识别模型ConCViT,利用CIFAR-10数据集对注意力权值进行预训练,以低信噪比小样本光伏红外图像为数据集,训练出高准确率的热斑检测模型。实验结果表明,ConCViT模型比传统卷积神经网络的识别准确率高12.02%,比深度卷积自编码网络的识别准确率高4.14%,并具有更快的收敛速度。     

基于调控一体自动化的变电站电气设备故障分析

电气设备是变电站发挥电力控制与管理功能的主要载体,电气设备故障监测是确保区域电力稳定运行的基础。文中构建了BP神经网络故障检测模型,利用调控一体自动化变电站的在线监测系统采集设备运行状态数据,将其作为故障检测样本;基于粒子群优化算法修正神经网络的权值与阈值,得到高精度的BP神经网络故障检测模型;将故障检测样本输入训练完成的模型中,输出电气设备故障检测结果即可判断何种电气设备发生故障,为巡检人员提供维修依据。     

楼宇型分布式能源系统设备容量和运行策略优化研究

设备容量优化和运行策略优化是分布式能源系统设计,运行的关键问题。为实现分布式能源系统的经济效益,能效水平和环境效益最大化,针对楼宇型分布式能源系统建立了相对普适化的物理模型和数学模型,以粒子群优化算法和线性规划相结合,采用两阶段优化方法计算系统的最优容量配置,并给出运行策略。以某写字楼的分布式能源系统为例,得到最优的系统设备容量和全年逐时运行策略,并采用遍历法验证计算结果的准确性。优化的分布式能源系统与传统供能系统相比,费用年值降低7.79%,年总能耗降低24.18%,污染物排放量减少了62.77 %。     

应用于分布式能源系统的设备选择和容量配置模型

提出了一种通用的分布式能源系统设备选择方法,建立了分布式能源系统中常用设备的数学模型,以年总费用最优为目标,利用GAMS软件对模型中的设备容量进行求解,对有无蓄能设备的分布式能源系统进行经济性分析。以天津市某办公楼为实例验证了模型及求解方法的有效性,结果表明:系统规划时考虑蓄能设备可以优化供能设备的容量配置,使系统的初投资降低9. 13%;在分时电价体系下,增加蓄能设备的系统可使年总运行费用降低4%。     

Characteristics of day‐ahead wind power forecast errors in Nordic countries and benefits of aggregation

The growing proportion of wind power in the Nordic power system increases day‐ahead forecasting errors, which have a link to the rising need for balancing power. However, having a large interconnected synchronous power system has its benefits, because it enables to aggregate imbalances from large geographical areas. In this paper, day‐ahead forecast errors from four Nordic countries and the impacts of wind power plant dispersion on forecast errors in areas of different sizes are studied. The forecast accuracy in different regions depends on the amount of the total wind power capacity in the region, how dispersed the capacity is and the forecast model applied. Further, there is a saturation effect involved, after which the reduction in the relative forecast error is not very large anymore. The correlations of day‐ahead forecast errors between areas decline rapidly when the distance increases. All error statistics show a strong decreasing trend up to the area sizes of 50,000 km². The average mean absolute error (MAE) in different regions is 5.7% of installed capacity. However, MAE of a smaller area can be over 8% of the capacity, but when all the Nordic regions are aggregated together, the capacity‐normalized MAE decreases to 2.5%. The average of the largest errors for different regions is 39.8% and when looking at the largest forecast errors for smaller areas, the largest errors can exceed 80% of the installed capacity, whereas at the Nordic level, the maximum forecast error is only 13.5% of the installed capacity. Copyright © 2016 John Wiley & Sons, Ltd.     

基于状态转移矩阵的小水电机组发电系统可靠性评估

作为我国重要的电力资源,小水电机组大规模并网运行给发电系统乃至整个电网可靠性与安全运行带来新的挑战。由于小水电数量众多,容量不等,采用现有递推公式建立停运容量模型来进行小水电机组发电系统可靠性评估时步骤繁琐,中间数据多,工作量大,对此提出了一种基于状态转移矩阵的小水电发电系统可靠性算法,用状态空间图表示小水电机组停运容量模型,基于状态空间算法直接计算小水电机组各个停运容量状态的确切概率,由转移率矩阵适当变换求取增量频率,用分块矩阵组合表示小水电机组停运容量模型和负荷容量模型的组合,合并相同容量状态,增加等步长而未出现的容量状态,即得小水电发电系统的裕度容量模型,进而计算累积概率、累积频率等可靠性指标。算例分析验证了此算法的可行性,可在实际操作中推广应用。     

Optimal asset allocation of wind energy units in conjunction with demand response for a large‐scale electric grid

Demand response is considered to be a realistic and comparatively inexpensive solution aimed at increasing the penetration of renewable generations into the bulk electricity systems. The work in this paper highlights the demand response in conjunction with the optimal capacity of installed wind energy resources allocation. Authors proposed a total annual system cost model to minimize the cost of allocating wind power generating assets. This model contains capacity expansion, production, uncertainty, wind variability, emissions, and elasticity in demand to find out cost per hour to deliver electricity. A large‐scale electric grid (25 GW) is used to apply this model. Authors discovered that demand response based on interhourly system is not as much helpful as demand response grounded on intrahourly system. According to results, 32% wind generation share will provide the least cost. It is also worth noting that optimal amount of wind generation is much sensitive to installation cost as well as carbon tax.     

提高油田机采系统用能效率的措施

胜利油田1616口游梁式抽油机井的机采系统测试表明，游梁式抽油机井平均功率因数为0．461，平均系统效率只有28．59％．泵充满系统小于0．4的抽油机井占测试总井数的33％。分析显示．造成抽油机系统效率低的原因主要是：变压器与异步电动机之间匹配方式不合理，设备容量大，自身损耗大；占油田电机总量31．4％的Y系列电机的平均空载损耗大，为2.86kw：低产液井电机的冲次可调范围有限，受电机最低转度和皮带轮最小包角限制，其最小冲次仅可调到4次／min，但1～1．5次，min的冲次即可满足生产需要，造成能耗浪费。为此，开发应用新型抽油机脱动系统，抽油机电机负载率从15％提高到35％，平均有功节电率和无功节电率分别达到15％和80％；开发应用抽油机平衡度实时在线检测技术，实现节能降耗；开发新型永磁电机，以增大启动力矩，合理调节冲次。     

基于直流潮流计算的风电接入能力分析

依据电力系统分析中采用的直流潮流算法,研究了风电功率在静态安全约束下的传输功率能力,建立了传输功率约束的详细数学模型以及基于直流潮流的简化模型并给出了求解方法,对大规模互联电网建立了交流潮流与有功功率约束相结合的实用模型。通过系统中的算例表明,该方法简捷、可靠,能够满足实际分析计算需要。     

Dispatch modeling of a regional power generation system – Integrating wind power

A modeling tool has been developed which can be used to analyze interaction between intermittent wind power generation and thermal power plant generation in a regional electricity grid system. The model uses a mixed integer programming (MIP) approach to determine the power plant dispatch strategy which yields the lowest systems costs. In the model, each large thermal plant is described separately, including properties such as start-up time, start-up cost and minimum load level. The model is evaluated using western Denmark as a case study.For western Denmark, it is found that the inclusion of start-up performance (i.e. start-up time and related costs) and minimum load level of the power generating units have a significant impact on the results. It is shown that the inclusion of these aspects influences the analysis of the effect of wind power variations on the production patterns of thermal units in the system. The model demonstrates how the introduction of wind power production and associated variations change the dispatch order of the large thermal power plants in the western Denmark system so that the unit with the lowest running costs no longer has the highest capacity factor. It is shown that this effect only is detected if start-up performance and minimum load level limitations are included in the optimization. It can also be concluded that start-up performance and minimum load level must be taken into account if the total system costs and emissions are not to be underestimated. The simulations show that if these aspects are disregarded, both total costs and total emissions of the power system are underestimated, with 5% in the case of western Denmark. Models such as the one developed in this work can be efficient tools to understand the effects of large-scale wind power integration in a power generation system with base load plants.     

Simulating subhourly variability of wind power output

As installed wind power capacity grows, subhourly variability in wind power output becomes increasingly important for determining the system flexibility needs, operating reserve requirements, and cost associated with wind integration. This paper presents a new methodology for simulating subhourly wind power output based on hourly average time series, which are often produced for system planning analyses, for both existing wind plants and expanded, hypothetical portfolios of wind plants. The subhourly model has an AR(p)‐ARCH(q) structure with exogenous input in the heteroskedasticity term. Model coefficients may be fit directly to high‐pass filtered historical data if it exists; for sets of wind plants containing hypothetical plants for which there are no historical data, this paper presents a method to determine model coefficients based on wind plant capacities, capacity factors, and pairwise distances. Unlike predecessors, the model presented in this paper is independent of wind speed data, captures explicitly the high variability associated with intermediate levels of power output, and captures distance‐dependent correlation between the power output of wind plants across subhourly frequencies. The model is parameterized with 1‐minute 2014 plant‐level wind power data from Electric Reliability Council of Texas (ERCOT) and validated out‐of‐sample against analogous 2015 data. The expanded‐capacity model, fit to 2014 data, produces accurate subhourly time series for the 2015 wind fleet (a 49% capacity expansion) based only on the 2015 system's wind plant capacities, capacity factors, and pairwise distances. This supports its use in simulating subhourly fleet aggregate wind power variability for future high‐wind scenarios.     

Energy and cost savings in household refrigerating appliances: A simulation-based design approach

In this study a novel design methodology for household refrigeration systems focused on both energy savings and cost reduction is presented and evaluated. Mathematical models were put forward for each of the system components and used to simulate the energy performance of the entire refrigeration system. The system simulation model was validated against experimental data obtained for a single-door 300-l vertical freezer. It was found that the model predictions for the energy consumption, cooling capacity and runtime ratio deviated from the experimental data within an error band of ±10%. An optimization algorithm was built upon the simulation model to size the condenser and evaporator heat transfer areas, and also the cabinet insulation thickness aiming at minimizing the total cost of the refrigeration system for a target energy consumption. A trade-off relation between the minimum cost and the minimum energy consumption was achieved, bringing about a system configuration that consumes 14% less energy than the baseline system if the total cost remains unchanged. The effect of the compressor stroke volume and efficiency on the minimum cost was also taken into account. It was demonstrated that the refrigerator/freezer becomes less costly in cases where highly efficient compressors are used in low energy consumption refrigerating appliances.