基于BP神经网络的天线识别

本文比较研究了BP神经网络中的几种常用算法,针对这些不同算法下的BP神经网络进行训练,并得出了各自网络的性能。在此基础上,针对经典BP算法和LM算法进行对比研究,找到LM算法的改进之处。此外,在实际的应用中表明,不仅不同的BP算法对网络的运算速度、泛化能力等有较大的影响,而且BP神经网络对隐含层神经元数目也很敏感。我们希望在BP神经网络的基础上,构建一种合适的天线模型,来应用于天线的分类识别,这将具有很大的现实意义。     

基于Adaboost集成BP网络原煤生产成本预测研究

为了提高传统BP神经网络预测模型精度,避免BP网络容易陷入局部极值、收敛速度慢等问题,将BP神经网络与Ada-boost算法相结合,提出了一种Adaboost集成BP神经网络模型.结合磁县观台煤矿原煤生产成本相关数据,建立了原煤生产成本预测的Adaboost集成BP神经网络模型,将该模型用于实际的原煤成本预测.结果表明：该模型预测精度高于传统的BP神经网络,收敛速度快,具有较强的鲁棒性,预测精度能满足实际预测需要,为原煤生产成本预测提供了一种新的途径,也为原煤生产成本控制提供了重要依据.     

基于BP-Adaboost强预测器的瓦斯涌出量预测

为了提高传统BP神经网络瓦斯涌出量预测模型精度,避免BP网络容易陷入局部极值、收敛速度慢等问题,将BP神经网络和Adaboost算法相结合,提出了一种BP-Adaboost强预测器模型.将该模型用于实际瓦斯涌出量预测,并进行了40次仿真实验.结果表明:该模型预测精度高于传统的BP神经网络,且收敛速度快,具有较强的鲁棒性,预测精度能满足实际工程需要,为瓦斯涌出量预测提供了一种新的途径.     

修正的BP神经网络森林火灾成灾面积预测研究

以1997-2012年《中国林业统计年鉴》的全国森林火灾成灾面积为依据,应用BP神经网络模型对成灾面积进行了预测,对预测方法进行了检验.在此基础上,利用残差提出了修正的BP神经网络模型,并对预测方法进行了改进.研究结果表明,修正的BP神经网络预测精度高于BP神经网络,预测相对误差平均为7.2％,可应用于森林火灾成灾面积的预测.     

基于Matlab的神经网络在绵阳空气质量预测中的应用

以绵阳市2014～2016年空气污染指数(API)以及SO_2、NO_2、PM_(10)等污染物为研究对象,探讨了绵阳市空气污染的变化规律,并分析它们与常规观测的地面气象资料之间的关系。尝试采用多元线性回归方法及BP神经网络方法建立污染预报模型,并检验分析两种模型的可行性。结果表明基于BP神经网络的预报模型在污染预报中可行,并建立基于BP神经网络进行空气质量预测的预测模型,利用历史资料进行验证。     

基于BP神经网络的空调性能预测研究

针对空调系统传统设计所需的匹配试验量大,严重浪费公司水电以及人力资源的问题,本文提出BP神经网络对空调系统的性能进行预测的方法,以设计参数压缩机冷量、冷凝器和蒸发器面积和实验参数室内外环境温度、负荷量等为输入,以空调的能力和能效比为输出,对BP网络进行训练、测试。结果表明:BP神经网络算法的均方误差为0.005 8,预测结果的总相关系数为0.974 47,可用于空调系统的设计选型,减少实验成本。     

基于BP模型对城市交通噪声的数据处理和预测

城市交通噪声的预测和评价技术是城市交通可持续发展的重要研究内容，直接为城市交通规划中环境容量分析和环境影响评价服务。本文通过实测的大量数据，运用神经网络中的BP模型及其算法建立车辆数、道路宽度和交通噪声之间的关系，对城市道路交通噪声数据进行处理和预测。     

基于BP神经网络模型在珊溪水库水华预测中的应用

在珊溪水库藻类暴发期间应急监测数据的基础上,建立pH值、高锰酸盐指数、总氮、总磷、叶绿素a数据矩阵。运用MATLAB R2015b GUI可视化界面模块,将应急监测数据样本空间分为训练样本、验证样本、测试样本,建立珊溪水库BP神经网络模型,预测了珊溪水库藻类暴发期间叶绿素a浓度。BP神经网络建模结果显示:输出数据与实测数据相关系数0.978,平均相对误差-0.19%,标准方差18.54%,模型稳定性较好,叶绿素a预测结果符合预期。BP神经网络预测模型为珊溪水库饮用水水源地环境保护提供了科学依据。     

PCA-ANN在典型航空材料铝合金腐蚀数据预测中的应用

针对航空装备在热带海洋大气环境下服役时对金属材料大气腐蚀预测的需求，提出了主成分分析法（PCA）优化的BP神经网络（BPNN）和广义回归神经网络（GRNN）模拟热带海洋大气腐蚀预测模型。研究结果表明，PCA可以很好地对原始数据的进行特征提取，降低样本集的维度。PCA-BPNN和PCA-GRNN模型的拟合优度与预测精度没有明显的相关关系。与PCA-BPNN相比，PCA-GRNN的预测精度高、稳定性更好，这为模拟热带海洋大气腐蚀研究提供了新思路，具有较好的借鉴意义。     

BAF处理系统的BP神经网络模型

本文以某污水处理厂曝气生物滤池（Biologlcal aerated filter，BAF）的实际运行数据为基础，采用人工神经网络（Artificial neural network，ANN）方法，建立起BAF处理系统的BP神经网络预测模型。模型运算结果表明，预测值和实测值能较好地吻合，起到了模拟预测的效果，同时能优化运行状态。该模型的建立为BAF处理系统的预测及运行管理供了一条简便实用的途径，具有良好的研究和工程实用价值。     

模拟退火算法在大气环境质量综合评价中的应用

为了建立一种对多项指标都能普遍使用的大气环境质量综合评价模式,利用模拟退火算法对幂函数加和型指数评价中参数进行优化,并用优化后的综合评价指数公式对大气环境进行综合评价。该公式应用于大量实例的分析评价结果与其他多种评价方法评价结果比较表明:该公式不受污染物种类和数目多少的限制,计算简便,具有可行性、实用性和通用性。     

Influences of Spatial Scale and Soil Permeability on Relationships Between Land Cover and Baseflow Stream Nutrient Concentrations

The Little Miami River (LMR) basin, dominated by agriculture, contains two geologically-distinct regions; a glaciated northern till plain with soils three times more permeable than a southern, pre-Wisconsinan drift plain. The influences of two landscape measures, percent row crop cover (%RCC, computed at three spatial scales), and soil permeability (PERM), on baseflow nutrient concentrations were modeled using linear regressions. Quarterly water samples collected for four years were analyzed for nitrate-N (NN), Kjeldahl-N (KN), total-N (TN), and total-P (TP). In till plain streams (n = 17), NN concentrations were 8.5-times greater than drift plain streams (n = 18), but KN and TP were 20–40% lower at comparable %RCC. These differences resulted in TN/TP molar ratios >80 in till plain streams, but <6 in drift plain streams. For till plain steams regression models based on %RCC accounted for 79% of the variance in NN concentrations but only 27% in drift plain streams. However, regressions on %RCC accounted for 68–75% of the KN and TP concentration variance in the drift plain streams but essentially none in the till plain. Catchment PERM influenced the regional NN/KN ratios which were 10-fold higher in the drift plain streams. For both till and drift streams the catchment scale %RCC gave the best predictions of NN, a water soluble anion, but the smaller spatial scales produced better models for insoluble nutrient species (e.g., KN and TP). Published literature on Ohio streams indicates that these inter-regional differences in nutrient ratios have potential implications for aquatic biota in the receiving streams.     

A novel framework-based cuckoo search algorithm for sizing and optimization of grid-independent hybrid renewable energy systems

Hybrid renewable energy systems (HRES) turned into an appealing choice for supplying loads in remote areas. The application of smart grid principals in HRES provides a communication between the load and generation from the HRES. Using smart grid in the HRES will optimally utilize the generating resources to reschedule the loads depending on its importance. This paper presents a new proposed design and optimization simulation program for techno-economic sizing of grid-independent hybrid PV/wind/diesel/battery energy system using Cuckoo search (CS) optimization algorithm. Using of CS will help to get the global minimum cost condition and prevent the simulation to be stuck around local minimum. A new proposed simulation program (NPSP) is acquainted using CS to determine the optimum size of each component of the HRES for the lowest cost of generated energy and the lowest value of dummy energy, at highest reliability. A detailed economic methodology to obtain the price of the generated energy has been introduced. Results showed that using CS reduced the time required to obtain the optimal size with higher accuracy than other techniques used iterative techniques, Genetic Algorithm (GA), and Particle Swarm Optimization (PSO). Numerous significant outcomes can be extracted from the proposed program that could help scientists and decision makers.     

Short-Term Hydroscheduling with Discrepant Objectives Using Multi-Step Progressive Optimality Algorithm1

Cheng, Chuntian, Jianjian Shen, Xinyu Wu, and Kwok-wing Chau, 2012. Short-Term Hydroscheduling with Discrepant Objectives Using Multi-step Progressive Optimality Algorithm. Journal of the American Water Resources Association (JAWRA) 48(3): 464-479. DOI: 10.1111/j.1752-1688.2011.00628.x Abstract: With increase in the number and total capacity of hydropower plants in power systems, optimality algorithms with a single objective are not suitable for optimizing the operation of complex hydropower systems to meet complex demands. Hydropower plants should prioritize discrepant objectives, such as peak regulation and maximizing generation during solving of optimal operation problems of hydropower systems. In this article, we present a multi-step progressive optimality algorithm (MSPOA) for the short-term hydroscheduling (STHS) problem to improve the quality of optimal solutions and enhance the convergence speed of progressive optimality algorithm (POA). In MSPOA, the original problem is first decomposed into a sequence of problems with the longer time steps. Next, the problem with the longest time step is solved, and the optimal solution is used as the initial solution for the problem with the second longest time step. This process proceeds until the original problem with the shortest time step is solved. The proposed discrepant-objective method and solution technique are tested for two types of hydroelectric systems. The results show that MSPOA can give better solutions and cost less time than POA due to enlarging feasible range of decision variables and reducing the number of computational stages. Discrepant objectives among hydropower plants can express the operation characteristics of complex hydropower systems more accurately than unique objective or multiple objectives.     

DERIVING THE NONLINEAR RISK-BENEFIT ALGORITHM FOR RESERVOIRS1

ABSTRACT: The Nonlinear Risk-Benefit (NRB) Algorithm includes risk as one of the objectives in a multiple-objective optimization problem. The NRB Algorithm is derived by extending the Surrogate Worth Trade-Off method to quadratic programming. This category of problem is common in water resources planning and design, especially multipurpose reservoir systems. Consequently, an example is given using the algorithm for optimally operating a multipurpose reservoir.     

Modeling Hydraulic and Thermal Electricity Production Based on Genetic Algorithm-Time Series (GATS)

Abstract

This study deals with the estimation of electricity production from hydraulic and thermal sources using the Genetic Algorithm (GA) with time series (TS) approach. Two forms of the mathematical models are developed, of which one is exponential and the second is polynomial. The power form of the Genetic Algorithm-Time Series (GATS) model is used for the thermal electricity production. The polynomial form of the GATS is used for the electricity production from the hydraulic sources. The GATS weighting parameters are obtained by minimizing the Sum of Squared Error (SSE) between observed and estimated electricity production from both sources. Therefore, the fitness function adapted is the minimization of the SSE for use in the GA process. The application of the GATS model is correspondingly presented. Some future scenarios are made to increase the electricity production from hydraulic sources. Variations of the electricity production from thermal and hydraulic energy sources are analyzed. Future prospects of electricity production are dealt with in terms of policy changes. The GATS models are used for making scenarios for future electricity planning policy. Results also show if current trend continues, the thermal electricity production amounts to 75% of the total electricity production, which is undesirable for environmental concerns. Results also shows that if new policy is to move from the thermal to hydraulic electricity production, the hydraulic sources will meet the demand until 2020.     

Neural Network Input Selection for Hydrological Forecasting Affected by Snowmelt1

Abstract: Snowmelt largely affects runoff in watersheds in Nordic countries. Neural networks (NN) are particularly attractive for streamflow forecasting whereas they rely at least on daily streamflow and precipitation observations. The selection of pertinent model inputs is a major concern in NNs implementation. This study investigates performance of auxiliary NN inputs that allow short‐term streamflow forecasting without resorting to a deterministic snowmelt routine. A case study is presented for the Rivière des Anglais watershed (700 km²) located in Southern Québec, Canada. Streamflow (Q), precipitations (rain R and snow S, or total P), temperature (T) and snow lying (A) observations, combined with climatic and snowmelt proxy data, including snowmelt flow (Q_SM) obtained from a deterministic model, were tested. NN implemented with antecedent Q and R produced the largest gains in performance. Introducing increments of A and T to the NNs further improved the performance. Long‐term averages, seasonal data, and Q_SM failed to improve the networks.     

Optimizing mixture properties of biodiesel production using genetic algorithm-based evolutionary support vector machine

Nowadays, biodiesel is used as one of the alternative renewable energy due to the increasing energy demand. However, optimum production of biodiesel still requires a huge number of expensive and time-consuming laboratory tests. To address the problem, this research develops a novel Genetic Algorithm-based Evolutionary Support Vector Machine (GA-ESIM). The GA-ESIM is an Artificial Intelligence (AI)-based tool that combines K-means Chaotic Genetic Algorithm (KCGA) and Evolutionary Support Vector Machine Inference Model (ESIM). The ESIM is utilized as a supervised learning technique to establish a highly accurate prediction model between the input--output of biodiesel mixture properties; and the KCGA is used to perform the simulation to obtain the optimum mixture properties based on the prediction model. A real biodiesel experimental data is provided to validate the GA-ESIM performance. Our simulation results demonstrate that the GA-ESIM establishes a prediction model with better accuracy than other AI-based tool and thus obtains the mixture properties with the biodiesel yield of 99.9%, higher than the best experimental data record, 97.4%.     

A numerical model for drying of spherical object in an indirect type solar dryer and estimating the drying time at different moisture level and air temperature

A numerical model for simultaneous heat and mass transfer was developed for solar drying of spherical objects and the object considered is green peas. Solar collector outlet temperature is assumed as drying chamber temperature and justified through energy balance equations. Assumptions are imposed on heat and mass transfer governing equations without losing the physics of the problem. Discretization is performed by finite difference method with implicit scheme. To generalize, the governing equation and boundary conditions are non-dimensionalized. The set of finite difference equations was solved by Tridiagonal Matrix Algorithm and a computer code in MATLAB was developed to solve them. The drying curves showed two stages of drying, initial, and secondary drying stage. At all drying temperatures and drying time, the center moisture was maximum and it was minimum at the boundary. A percentage of 85.67 surface moisture content and 25.33% center moisture was eliminated in the first 1 hr at 348 K. The product should be dried up to 7.45, 4.74, and 3.74 hr at air drying temperatures of 318, 333, and 348 K respectively, to maintain 10% of the product’s initial moisture content. The result is compared with the experimental result from literature and they are found to be in good agreement.     

GIS‐Based Spatial Precipitation Estimation: A Comparison of Geostatistical Approaches1

Abstract: As one of the primary inputs that drive watershed dynamics, the estimation of spatial variability of precipitation has been shown to be crucial for accurate distributed hydrologic modeling. In this study, a Geographic Information System program, which incorporates Nearest Neighborhood (NN), Inverse Distance Weighted (IDW), Simple Kriging (SK), Ordinary Kriging (OK), Simple Kriging with Local Means (SKlm), and Kriging with External Drift (KED), was developed to facilitate automatic spatial precipitation estimation. Elevation and spatial coordinate information were used as auxiliary variables in SKlm and KED methods. The above spatial interpolation methods were applied in the Luohe watershed with an area of 5,239 km², which is located downstream of the Yellow River basin, for estimating 10 years’ (1991‐2000) daily spatial precipitation using 41 rain gauges. The results obtained in this study show that the spatial precipitation maps estimated by different interpolation methods have similar areal mean precipitation depth, but significantly different values of maximum precipitation, minimum precipitation, and coefficient of variation. The accuracy of the spatial precipitation estimated by different interpolation methods was evaluated using a correlation coefficient, Nash‐Sutcliffe efficiency, and relative mean absolute error. Compared with NN and IDW methods that are widely used in distributed hydrologic modeling systems, the geostatistical methods incorporated in this GIS program can provide more accurate spatial precipitation estimation. Overall, the SKlm_EL_X and KED_EL_X, which incorporate both elevation and spatial coordinate as auxiliary into SKlm and KED, respectively, obtained higher correlation coefficient and Nash‐Sutcliffe efficiency, and lower relative mean absolute error than other methods tested. The GIS program developed in this study can serve as an effective and efficient tool to implement advanced geostatistics methods that incorporate auxiliary information to improve spatial precipitation estimation for hydrologic models.