Clustering Analysis for Time Feature of User Power Consumption Based on Structural Self-adaptation ANN

In view of the important effect of clustering analysis in data mining, the clustering rules and its curve are studied to solve the problem of determining clustering number. A kind of self-adaptation clustering ANN is presented based on SOFM ANN, which can automatically determine the clustering number. Based on practical sales data, the time feature analysis of power user consumption are carried out by using the self-adaptation clustering ANN, whose conclusion has the important referenced values for adjusting power price correspondingly and arranging power producing reasonably.     

利用神经网络提取棉花叶片数字图像氮素含量的初步研究

选取6种输入向量组合，利用线性网络、BP网络以及径向基网络等3种神经网络模型进行比较研究，筛选最适宜网络模型和最佳输入组合，建立叶片数字图像彩色信息和叶片氮含量的关系模型，探索利用神经网络技术获取叶片数字图像信息的方法。结果表明，径向基网络在利用数字图像(B，H，G-R，G/R)指标作为网络输入向量时，能够实现获取棉花叶片数字图像氮含量的目标。径向基网络训练的180组样本的训练精度均达到极显著水平（r = 0.9022**），30组测试样本的预测值与实测值也达到极显著相关（r = 0.8674**），径向基网络和(B，H，G-R，G/R)向量是一种适合本研究的数学模型。对利用神经网络提取棉花叶片数字图像氮含量技术的初步探索，拓展了神经网络和数字图像技术在农业生产中的应用。     

用模糊神经网络提高洪峰预报精度的研究

在大量研究的基础上,提出了基于模糊理论的神经网络改进算法,用来提高对洪峰的预报精度。该方法在网络训练时引入模糊理论来确定网络误差修改的程度。引入的算法增大了大值输出样本和期望输出的误差,使得网络向着提高洪峰拟合精度的方向修改权重。应用表明,改进的模糊BP神经网络能够较好的反映洪水演进机理,提高了神经网络洪水预报模型对洪峰的预报精度,保证了洪峰预报的可靠性。     

带锚FRP受剪加固梁非剥离破坏模式BP网络预测

对带可靠锚固FRP受剪加固混凝土梁的非剥离剪切破坏模式做了细化分类,即包括FRP断裂控制的破坏、受压区混凝土（达到极限应力状态）压碎控制的破坏、FRP断裂与混凝土压碎同步发生的界限破坏等3种模式;利用BP神经网络建立了带锚纤维受剪加固梁破坏模式的智能预测模型,与31根非剥离破坏加固梁试验的对比结果显示：模型总体精度达到90%,说明建立的破坏模式网络预测模型适用于带锚纤维受剪加固梁非剥离剪切破坏模式的判别。     

Combination of GPS monitoring and ANN prediction for slope deformation

Based on the evolution pattern of slope deformation and failure, stability status of a slope is evaluated through analyzing the data of GPS monitoring. A neural network model of slope displacement time is developed based on the GPS monitoring data. It can be used to forecast slope deformation trends. A neural network displacement prediction model of slope deformation is proposed with Matlab ANN toolbox. Upon a case study, the ANN prediction results based on GPS monitoring data are analyzed.     

Vacuum Circuit Breaker''s Condition Simulation Based on Artificial Neural Network and Electrical Endurance Quality Criterion

How to select the factors that have more great affection on the condition based maintenance and how to make sure the relation among them are two key problems. A new method's proposed for vacuum circuit breaker's condition based maintenance, which combines electrical endurance quality criterions and condition recognition arithmetic based on artificial neural network. The paper proposes a model based on electrical-endurance quality criterions by applying ANN in vacuum circuit breaker's condition based maintenance. The simulation indicates that the criterion selected in this way are reasonable and the network with new learning error function has a much better generalization ability.     

人工神经网络与近红外光谱分析

介绍了人工神经网络的原理、结构、算法和研究进展,以及该方法在近红外光谱分析中的重要地位和应用。     

A Kind of Fuzzy System Modelling and Control Based on Neural Networks

Combining artificial neural networks(ANN) with fuzzy system theory,a kind of modelling & control method of fuzzy system based on ANN is presented.The simulation researches have verified that the proposed approach can be applied effectively to a number of control systems which are defficult to build strict mathematical model.     

Application of Artificial Neural Network in Elastic Internal Force Analysis of Two Way Slab

In this paper, the Artificial Neural Network (ANN) is applied to the internal force analysis of structure. After introducing the model and algorithm of back-propagation network, a trained BP network of three layers is used to calculate the maximum of elastic moment of middle span of two-way slab. In order to enhance the generalized capability of network, the modified value of weight should be regarded as convergence standard and in order to accelerate the learning process without vibration, the method of adding momentum coefficient is adopted. The analysis program of BP network is carried out with the software of Matlab. The result demonstrates that application of ANN in structure analysis is feasible.     

Active Control of Building Structure Using Artificial Neural Networks

A study on application of artificial neural networks(ANN) to active structural control is presented. A feedforward neural network with an adaptive back propagation training method is used in this paper. In the back propagation training, the learning rate is determined by ensuring the decrease of the total error function of the output training patterns at each training cycle. Four-layer-BP networks is constructed to produce the active force for the building structure subjected to earthquake excitation. Numerical examples of single-degree-of freedom systems under earthquake excitations are given to illustrate the effectiveness of the proposed control strategy.     

Prediction Model of Exploitation Sink Using Ann and GIS

A new idea is put forward on researches of prediction method of coal mining subsidence supported by GIS, and a new method is given about quantitative prediction of mining subsidence by means of GIS and ANN(Artificial Neural Network).This paper has completed a lot of work concerning choosing factor,data processing, establishing and validating preliminary ANN prediction model. Further more,data processing is carried out by GIS software,and the BP training method is used for modeling the exploitation sink system. Subsequently,the error is qualitatively analyzed with considering the result of verification. Researches show that the ANN prediction model supported by GIS has theoretical feasibility and realistic significance in predicting complex exploitation sink system,and GIS and ANN possess wide application prospects in the prediction of exploitation sink.     

Artificial neural network approach for prediction of ammonia emission from field-applied manure and relative significance assessment of ammonia emission factors

This article presents a systematic method for enhancing the estimation accuracy of ammonia emission from field-applied manure and for assessing the relative significance of ammonia emission factors, using the feedforward-backpropagation artificial neural network (ANN) approach.

The multivariate linear regression (MLR) method well describes the ammonia emission tendency with the emission factor variation. However, ammonia emission from manure slurry is too complex to be captured in a linear regression model. This necessitates a model which can describe complex nonlinear effects between the ammonia emission variables such as soil and manure states, climate and agronomic factors. In the present study, a principle component analysis (PCA) based preprocessing and weight partitioning method (WPM) based postprocessing ANN approach (called the PWA approach) is proposed to account for the complex nonlinear effects.

The ammonia emission is predicted with precision by the 11 emission factors, using the nonlinear ANN approach. The relative importance among the 11 emission factors is identified using the elasticity analysis in the MLR method and using the WPM in the ANN approach. The relative significance obtained quantitatively by the PWA approach in the present study gives an excellent explanation of the most important processes controlling NH₃ emission.     

Load Prediction with Neural Network Method and Operation Optimization of Ice Storage Air Conditioning System

This paper presents some common control strategy of ice storage air conditioning system. It is suggested that the optimization on ice storage air conditioning system should be based on accurate load prediction and the artificial neural network (ANN) modeling for load prediction was presented.     

ANN Disaster Identifying of Wujiawan Landslide in Wanzhou Urban

The paper aims at ANN disaster-possibility identifying of Wujiawan Landslide. ANN construction and parameter setup are analyzed for landslide disaster identifying by ANN, based on a typical landslide-Wujiawan landslide in Wanzhou urban, by confirming evaluation factor and establishing sample data. The ANN model is trained by the similar landslide sample in Wanzhou urban, then the disaster is identified in several different conditions of Wujiawan landslide. Finally, the same conclusion are found by analyzing combined ANN Disaster-Identifying and limit-equilibrium-method calculation. The results show that AAN is accurate and satisfied to be used landslide disaster-possibility identifying.     

Application of Artificial Neural Network in Modeling Paint Waste Water Treated by Coagulation Oxidation Process

An artificial neural network (ANN) model was established based on data of paint waste water treated by coagulation oxidation process, using the improved back propagation algorithm. The model was then used to fit and predict some experimental data. The results indicated that the errors between computed data and experimental data were much small. Furthermore, the ANN model could correctly reflect the mechanism of some factors which affected the efficiency of paint waste water treatment.     

Fisher discriminant analysis model for selecting the retaining structure type of deep foundation pit

Aiming at the problem of traditional evaluation methods of deep foundation pit for selecting the retaining structure type, based on the statistical theory and following the principle of security, economic and reasonable, a Fisher discriminant analysis(FDA) model for selecting the retaining structure type for deep foundation pit is established. 10 selected indicators which influence selection of deep excavation program are taken into account as discriminant factors, and the supporting schemes for deep foundation pit are classified into 5 groups, viz. gravity of the cement-soil type, soil nailing wall, pile anchors, pile supports and underground continuous wall. After training and testing 64 sets of measured data, the discriminant functions of FDA are solved, the re-substitution method is introduced to verify the stability of FDA model and the ratio of mis-discrimination is 14.1%. Another 10 groups of measured data are tested as forecast samples by the proposed model, and the correct rate is equal to 100%. Therefore, the feasibility of the proposed model is validated. Moreover, the proposed model is adopted for the New World Center Project in China, and the prediction results are in line with the artificial neural network(ANN) and the actual situation. The result shows that the deep foundation pit supporting structure lectotype decision of FDA model has excellent discriminant performance and the resubstitution error rate is low. It is easy and efficient to make discriminant analysis using this model and it provides efficient method to select deep excavation retaining structure and a practical new approach to choose the structural type of deep foundation pit optimization.     

热力管道小漏温度场特征标定研究

泄漏特别是小漏预警对热力管道的安全维护具有重要意义。受空间分辨率的影响,分布式光纤传感器对小漏引起的局部温度变化测试精度较低,测量温度与实际温度差异较大。以布里渊光时域反射仪(BOTDR)作为测量手段,提出了一种建立分布式光纤测量温度与实际温度之间对应关系的方法。设计完成了小漏温度场模拟测量实验,通过高斯拟合对测量数据进行特征提取,再用人工神经网络建立测量温度与实际温度的映射模型。结果表明:设计的实验方案可获得代表管道小漏温度分布的先验数据,基于此训练的人工神经网络可确立实际温度场与BOTDR测量温度场的对应关系,提高了光纤测试精度并为泄漏预警策略的制定提供了依据。     

Information Fusion Algorithm Based Evidence Combination

This paper applies the evidence combination theory to fuse the information of multi-neural network classifier. In order to make each classifier approach the ideal state, the heredity algorithm is applied to train it. The different capacity of each classifier is caused by different classified feature. Input feature can't be identified by one classifier and may be identified by another, Model identification can be performed by multi-classifier, output result can be thought of evidence, further more,the BPA of each classifier is determined, then the procession of the model identification must be improved.     

Application of soft sensor in the on line measurement of theiron ore’s surface temperature

It is difficult to measure the surface temperature of iron ore directly. A method is put forward to handle this problem by using soft sensing technique. This on line measurement method is used to replace the Lagrange interpolation off line method to estimate surface temperature. The method used L M optimum algorithm to build up ANN soft sensor model combined with off line learning neural network to establish the correlation between input variables and target variables, to achieve the surface temperature on line detection. The results of simulation and experimentation indicate that the method is reasonable and feasible.