AN OPTIMIZATION METHOD OF THERMAL PLANT UNIT COMMITMENT PROBLEM

In this paper, the Lagrangian method for thermal plant unit commitment and some properties of its Lagrangian multiplier are discussed. Besides, a polynomial algorithm for updating the Lagramgian multiplier is proposed, the critical point of operating unit commitment is derived, and the computational quatity is less than that of the method of subgradient. AS a result, the consumption of coal of the thermal plant is greatly reduced and greater economic benefits will gained.     

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.     

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。     

Analysis on the Optimal Operation in Hydroelectric Station of Gezhouba

The optimal operation in hydroelectric station unit is me opumum problem,which include unit commitment,load distribution and unit maintenance. The authors research the optimal operation in units in hydroelectric station of Gezhouba by the genetic algorithm and dynamic programming to satisfy the demand of economic operation of Gezhouba. They arrange and organize the program of the units in r6ason to go for the better benefit and satisfies security, credibihty and high quality of electric production . These methods can work out the most optimum result. When the head of water and the flux is confirmed they can distribute the flux to the units to realize the optimal operation of the scheduling in Gezhotlba.     

A new pattern recognition method of gas sensor array detection

BP neural network based gas sensor array detection pattern recognition has some disadvantages, such as slow convergence and local minimum problem. A modified immune neural network model which combines BP algorithm and immune algorithm is proposed to enhance global search capability and improve the performance of the neural network model. Orthogonal test is adopted to design the study samples of neural network. This ensures the accuracy of neural network while reducing the number of samples. The simulation results show that the proposed pattern recognition method solves the cross sensitivity of gas sensor effectively, overcomes the disadvantages of traditional BP neural network and improves the learning speed and detection accuracy.     

Improvement and Application on Back Propagation Network Based on Partial Least-squares Algorithm

This paper proposes a novel BP network model based on nonlinear iterative partial least-squares algorithm which can fit nonlinear data. The novel BP network model can reduce iterative step number and advance learning effieieney. This paper pretreats data by nonlinear iterative partial least-squares algorithms. The weights initialization of input floor and output floor are set by applying the loading weights of dependent variable and cause variable, the member of hidden nodes are set by applying factor numbers of nonlinear iterative partial leastsquares algorithm, the connection co- efficient is set by applying the connection matrix B. Performances of the BP, PLS, and PLS-BP are analyzed and compared. The results show that the PLS-BP has better fitting and forecasting than BP and PLS.     

THE APPLICATION OF NETWORK PROGRAMMING METHOD ON THE SHORT TERM OPTIMAL SCHEDULING FOR CASCADED HYDROPOWER STATIONS

The problem of the short term economic scheduling for cascaded hydropower stations is discussed with the network programming method. A new network model is built to dispatch the load to each unit when the unit commitment is determined. An algorithm of the minimal cost maximal flow is also discussed using the model and the algorithm to the Longxi River Cascaded Hydropower Stations, the computational result on the microcomputer VAX II shows that the model and the algorithm are correct and applicable.     

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。     

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.     

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.     

基于随机森林法的棉花叶片叶绿素含量估算

为了高效和无损地估算棉花叶片的叶绿素含量,本研究测定了棉花光谱反射率及叶绿素含量(soilandplant analyzerdevelopment,SPAD)值,对光谱数据进行包络线去除处理、立方根转换和倒数转换,以SPAD值与反射光谱之间的相关性为基础,通过随机森林法筛选出对棉花叶片SPAD值影响较大的特征波段,构建估算棉花叶片SPAD值的BP神经网络(back propagation artificial neural networks, BP ANN)、偏最小二乘回归(partial least squares regression,PLSR)两个模型。结果表明,在605～690nm范围内的反射率与SPAD值相关性达0.01显著水平,均呈负相关,相关系数最高值为-0.619。与原始光谱相比,经过变换后的棉花反射率与SPAD值相关性结果相差较大,其中去除包络线光谱在550～750 nm波段范围有效提高了相关性,相关性效果优于倒数转换数据和立方根转换数据。随机森林法能够有效评出对SPAD值影响较大的特征波段,进而提高模型估算精度。在两种模型中,基于去除包络线光谱建立的PLSR和BP神经网络模型的决定系数R~2分别为0.92、0.83,说明这两种模型的估算能力较好;两种模型RMSE分别为0.88、1.26, RE分别为1.30%、1.89%,表明PLSR模型的估算精度比BP神经网络模型高。从模型的验证效果来看,PLSR模型在估算棉花SPAD值方面有一定的优势和参考价值。     

GA Based on Structural Formula Binary-coding and Its Applications in BP Networks

According to the topological characteristics of BP networks, a Genetic Algorithm based on the structural formula binary-coding has been designed in this paper. By means of fractionalizing the large-scale solution-space and performing the GA operations to the fractionalized subspaces, the GA's global-convergence and parallelism can be utilized to search the subspace for the optimal solution in the whole solution-space, thus definitude the starting point and narrow the domain for the next BP's local-search. Testing shows that the two-step algorithm (GA-BP) can solve the existed problems in the NN's training such as local minimum, tardy convergence and so on.     

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。     

Fuzzy c-means Cluster Over Gentic Algorithm

Fuzzy c-means (FCM) algorithm is dynamic cluster algorithm whose result often is local optimal decision. There often exists insignificant clustering in the result of Fuzzy c-means algorithm when traditional union, Intersection and inclusion work in fuzzy set. Our research indicates there are no insignificant clustering in the result of Fuzzy c-means algorithm over genetic algorithm and partial optimal solution can be avoided with this algorithm to a certain extent. The coding, select, corresponding crossover and mutation operators are designed. Finally we compared the performance of GFCM and FCM with testing data. Results show that the performance of GFCM is far better than FCM.     

BP Neural Networks and Its Application in Lung Cancer Intelligent Diagnosis System

The learning algorithm of networks is discussed. The programming example of 3 layer BP networks is given with Visual C++6.0 program langue. Based on this model, a lung cancer intelligent diagnosis system is successfully implemented. Furthermore, the paper introduces network's structure design, preferencesand the source of stylebookdatum in factual applications. The ameliorative arithmetic is applied to the study of networks and BP dynamic evolving process is designed. The experiments indicatecell images are recognized and classified by the trained neural network. The study illustrates the system has feasibility and clinical value in lung cancer diagnosis.     

Stratification and Bearing Capacity Prediction Method Based on BP Neural Network for Foundation in Huaibei Plain

According to the features of stratification and obvious inhomogeneity in geological soil in Huaibei plain, BP neural network prediction method for stratification and bearing capacity calculation of multiple cross-bedded foundation was proposed. By comparing the results of drill sampling, static cone penetration tests and screw plate tests, plate loading tests, penetration resistance ps value was found as an evaluation index for stratification and bearing capacity prediction of cross-bedded foundation. Moreover, gradient descent algorithm and conjugate gradient algorithm BP neural network models were obtained, and the calculation results of the two algorithms were comparatively analyzed. The results show that penetration resistance value can be taken as an evaluation index for stratification and bearing capacity prediction of cross-bedded foundation in Huaibei plain. Gradient descent algorithm and conjugate gradient algorithm BP neural network models have good results for soil identification and bearing capacity determination, which can meet the accuracy requirements of actual engineering. However, the computational efficiency of gradient descent algorithm is significantly lower than that of conjugate gradient algorithm.     

Nonlinear Forecast System for River Water Pollution Based on GIS

Based on the analysis of the water pollution spatial distribution characters of Yangtze River in Chongqing,a new method based on the integration of BP neural network and genetic arithmetic(GA) is proposed.For some shortcomings existed in the standard BP neural network,this method has ultimately overcome these shortcomings by combining the GA with BP artificial neural network through altering stimulating function,adding momentum factor to power value for BP algorithm and introducing genetic arithmetic to searching for the knots of the hidden layer,momentum factor and learning level.Using this method can easily overcome the difficulty of measuring the water prediction model's parameters.GIS is used as a tool for data management and spatial analysis,and the prediction result of the model for the water pollution spatial distribution characters of Yangtze River in Chongqing is visualized and explored with the precision of more than 78%.     

artificial neural networks based prediction of electromagnetic compatibility problems

It is necessary to predict electromagnetic compatibility (EMC) for electronic equipment and systems. We proposed a fast EMC prediction approach via artificial neural networks (ANN). By choosing relevant electromagnetic interference parameters as the input prediction features, a back propagation (BP) neural network was used to construct the mapping between the input prediction features and the electromagnetic disturbance response of the sensitive system. The EMC fast prediction BP model was trained and tested by sample sets generated using an electromagnetic computational method. We used this method to predict the crosstalk coupling between two wires. The experimental results show the effectiveness of the proposed method.     

A Novel Method Based on Support Vector Machine for Defect Identification

The traditional fault detection suffers from complicated process, low accurate ratio and off-line implement. The improved methods of defect recognition by artificial neural networks (ANN) can lead to the problems of overfit and bad generalization because of finite samples. With a view of data mining and technique parameters directly, the new approach using support vector machine classification algorithm after removing redundant parameters by rough set theory and eliminating noise of data to identify the defects is discussed. The results of a experiment show that unlike conventional and ANN recognition methods the new technique performs better than conventional evaluation ones with advantages of high efficiency, lower cost, easy implement on-line, excellent generalization and so on. The approach provides a novel technique means for nondestructive defect identification of various products.     

Using BP Neural Network to Determine the Membership Function of the Input of Fuzzy Control System

While design the fuzzy controller, it is very important to determine the membership function of fuzzy variables.The data can be broadly classified as fuzzy sets by using the classification property of the BP neural network. The author selects a BP neural network with one hide layer and uses S function to the input and hide layer, and linear function to the output layer.Advanced BP algorithm isused to train the BP neural network in the environment of MATLAB . The nearer to the target values is the better the last output is.With the trained BP network , the membership values of the inputs can be got ten. This method has high rate and low error.