从受训神经-模糊模型中提取知识

本文对用于模式分类、函数逼近、参数估计的多层感知器 (MLPs)给出 1个清晰的关于内部行为的解释。作者以单隐层的 MLP为例 ,论述了关于 MLP的内部行为的半线性分析理论。对受训的MLP,将隐层单元的输出分别定义为网络输出的正、负“内部分量”;定义内部分量的连接权重集为给定问题的“内部判别模式”;建立了 MLP和模糊集相结合的新模型 ;分析了 MLP的结构为 N- 2 - 1和N- H- 1 ,给出权重初始化的方法 ;提出了 1种从受训神经 -模糊模型 (NFMs)中提取知识的全新的具有实用价值的方法。

遗传算法-模糊径向基神经网络模型预测自润滑镀层耐磨性

针对传统神经网络模型存在的缺陷,引入遗传算法和模糊运算建立遗传算法-模糊径向基神经网络模型(GA-FRBFNNM),介绍了模型结构和仿真思路。以自润滑镀层耐磨性为研究主题开展正交实验,在正交实验结果中任取10组数据作为训练样本用于模型训练,其余6组数据作为测试样本用于模型性能测试。结果表明:GAFRBFNNM的预测值更接近于真实值,其预测精度明显高于相同结构的径向基神经网络模型,验证了该模型是有效的,能够更准确预测自润滑镀层耐磨性。主要归因于引入模糊运算使得径向基神经网络全部节点都具备特定意义,另外引入遗传算法优化了训练算法,避免了模型陷入局部极小点等问题,使得模型性能得到有效提升。

Drought Forecasting in a Semi-arid Watershed Using Climate Signals:a Neuro-fuzzy Modeling Approach

Large-scale annual climate indices were used to forecast annual drought conditions in the Maharlu-Bakhtegan watershed,located in Iran,using a neuro-fuzzy model.The Standardized Precipitation Index(SPI) was used as a proxy for drought conditions.Among the 45 climate indices considered,eight identified as most relevant were the Atlantic Multidecadal Oscillation(AMO),Atlantic Meridional Mode(AMM),the Bivariate ENSO Time series(BEST),the East Central Tropical Pacific Surface Temperature(NINO 3.4),the Central Tropical Pacific Surface Temperature(NINO 4),the North Tropical Atlantic Index(NTA),the Southern Oscillation Index(SOI),and the Tropical Northern Atlantic Index(TNA).These indices accounted for 81% of the variance in the Principal Components Analysis(PCA) method.The Atlantic surface temperature(SST:Atlantic) had an inverse relationship with SPI,and the AMM index had the highest correlation.Drought forecasts of neuro-fuzzy model demonstrate better prediction at a two-year lag compared to a stepwise regression model.

Parametric optimization of friction stir welding process of age hardenable aluminum alloys-ANFIS modeling

A comparative approach was performed between the response surface method(RSM) and the adaptive neuro-fuzzy inference system(ANFIS) to enhance the tensile properties, including the ultimate tensile strength and the tensile elongation, of friction stir welded age hardenable AA6061 and AA2024 aluminum alloys. The effects of the welding parameters, namely the tool rotational speed, welding speed, axial load and pin profile, on the ultimate tensile strength and the tensile elongation were analyzed using a three-level, four-factor Box-Behnken experimental design. The developed design was utilized to train the ANFIS models. The predictive capabilities of RSM and ANFIS were compared based on the root mean square error, the mean absolute error, and the correlation coefficient based on the obtained data set. The results demonstrate that the developed ANFIS models are more effective than the RSM model.

Data-driven intelligent monitoring system for key variables in wastewater treatment process

In wastewater treatment process(WWTP), the accurate and real-time monitoring values of key variables are crucial for the operational strategies. However, most of the existing methods have difficulty in obtaining the real-time values of some key variables in the process. In order to handle this issue, a data-driven intelligent monitoring system, using the soft sensor technique and data distribution service, is developed to monitor the concentrations of effluent total phosphorous(TP) and ammonia nitrogen(NH_4-N). In this intelligent monitoring system, a fuzzy neural network(FNN) is applied for designing the soft sensor model, and a principal component analysis(PCA) method is used to select the input variables of the soft sensor model. Moreover, data transfer software is exploited to insert the soft sensor technique to the supervisory control and data acquisition(SCADA) system. Finally, this proposed intelligent monitoring system is tested in several real plants to demonstrate the reliability and effectiveness of the monitoring performance.

Auxiliary error and probability density function based neuro-fuzzy model and its application in batch processes

This paper focuses on resolving the identification problem of a neuro-fuzzy model(NFM) applied in batch processes. A hybrid learning algorithm is introduced to identify the proposed NFM with the idea of auxiliary error model and the identification principle based on the probability density function(PDF). The main contribution is that the NFM parameter updating approach is transformed into the shape control for the PDF of modeling error. More specifically, a virtual adaptive control system is constructed with the aid of the auxiliary error model and then the PDF shape control idea is used to tune NFM parameters so that the PDF of modeling error is controlled to follow a targeted PDF, which is in Gaussian or uniform distribution. Examples are used to validate the applicability of the proposed method and comparisons are made with the minimum mean square error based approaches.

Neuro-fuzzy predictive control for nonlinear application

Aiming at the unsatisfactory dynamic performances of conventional model predictive control (MPC) in a highly nonlinear process, a scheme employed the fuzzy neural network to realize the nonlinear process is proposed. The neuro-fuzzy predictor has the capability of achieving high predictive accuracy due to its nonlinear mapping and interpolation features, and adaptively updating network parameters by a learning procedure to reduce the model errors caused by changes of the process under control. To cope with the difficult problem of nonlinear optimization, Pepanaqi method was applied to search the optimal or suboptimal solution. Comparisons were made among the objective function values of alternatives in initial space. The search was then confined to shrink the smaller region according to results of comparisons. The convergent point was finally approached to be considered as the optimal or suboptimal solution. Experimental results of the neuro-fuzzy predictive control for drier application reveal that the proposed control scheme has less tracking errors and can smooth control actions, which is applicable to changes of drying condition.

Nonlinear Modeling and Neuro-Fuzzy Control of PEMFC

The proton exchange membrane generation technology is highly efficient, and clea n and is considered as the most hopeful “green” power technology. The operatin g principles of proton exchange membrane fuel cell (PEMFC) system involve thermody namics, electrochemistry, hydrodynamics and mass transfer theory, which comprise a complex nonlinear system, for which it is difficult to establish a mathematic al model and control online. This paper analyzed the characters of the PEMFC; an d used the approach and self-study ability of artificial neural networks to bui ld the model of nonlinear system, and adopted the adaptive neural-networks fuzz y infer system to build the temperature model of PEMFC which is used as the refe rence model of the control system, and adjusted the model parameters to control online. The model and control were implemented in SIMULINK environment. The resu lts of simulation show the test data and model have a good agreement. The model is useful for the optimal and real time control of PEMFC system.

Construction of Three-dimensional power security Defense System Based on T-S Model Fuzzy Neural Network

With the development of information networks, the problem of power security has increasingly caused many attention of people, but the simple power security defense system has been difficult to meet the current complex network environment. Aiming at this situation, by using the method of T-S fuzzy neural network model to analyze the characteristics of the data transmission in network, it has obtained corresponding threat information. By processing these threat information, it completes the construction of three-dimensional power security defense system. The paper carries on the corresponding data training methods by using T-shirt model fuzzy neural network, which has certain reference significance for the data analysis of other similar fields. At the same time, the study of building on the three-dimensional power security defense system aims to provide a theoretical reference for solving the security defense of the current complex network environment.

Hybrid denoising-jittering data processing approach to enhance sediment load prediction of muddy rivers

Successful modeling of hydroenvironmental processes widely relies on quantity and quality of accessible data,and noisy data can affect the modeling performance.On the other hand in training phase of any Artificial Intelligence(AI) based model,each training data set is usually a limited sample of possible patterns of the process and hence,might not show the behavior of whole population.Accordingly,in the present paper,wavelet-based denoising method was used to smooth hydrological time series.Thereafter,small normally distributed noises with the mean of zero and various standard deviations were generated and added to the smooth time series to form different denoised-jittered data sets.Finally,the obtained pre-processed data were imposed into Artificial Neural Network(ANN) and Adaptive Neuro-Fuzzy Inference System(ANFIS)models for daily runoff-sediment modeling of the Minnesota River.To evaluate the modeling performance,the outcomes were compared with results of multi linear regression(MLR) and Auto Regressive Integrated Moving Average(ARIMA)models.The comparison showed that the proposed data processing approach which serves both denoising and jittering techniques could enhance the performance of ANN and ANFIS based runoffsediment modeling of the case study up to 34%and 25%in the verification phase,respectively.

Damage Assessment of Existing Transmission Structures Using ANFIS （Adaptive Neuro-Fuzzy Inference） Model

This paper introduces a new methodology for the damage assessment of existing-transmission structures using six layers, zero order Sugeno model. The model is a hybrid fuzzy-neural system that combines the power of neural networks and fuzzy systems. It is a learning expert system that finds the parameters of the fuzzy sets and fuzzy rules by exploiting approximation techniques from neural networks. The condition ratings of the structural components are determined based on visually observed deterioration-symptoms and the severity of those symptoms. A supervised learning process using training data and expert opinions is used to develop the expert system rules and determine the ratings of the structural components. For the learning from training data, the model uses a combination of least-square estimator and gradient descent method. A sequential least square algorithm is used to determine the weighting factors that minimized the errors. A test case is given to illustrate the power of the proposed fuzzy-neural system. It is concluded that the Sugeno model＇s ability to tune the parameters based on the training data makes it superior to the rules produced by an expert in the conventional fuzzy logic systems.

智能车辆巡航建模与硬件在环试验

采用软计算方法设计了智能车辆巡航的神经网络-模糊控制模型。模型采用两输入、单输出方式,两输入分别为车间距离和两车相对车速,单输出为油门量或者刹车量。为了获取模型训练和仿真的数据,设计了车辆跟随试验,试验采用GPS和蓝牙无线通信模块来实时获得两车间的相对距离和相对车速。利用试验数据对CANFIS(collective adaptive neuro-fuzzy inference system)模型进行离线训练,建立了巡航车间距离模型。根据建立的模型设计了控制器,并基于实时仿真平台DSPACE1103进行了硬件在环试验。仿真结果和试验结果对比表明,建立的模型具有一定的合理性,较好地体现了车辆系统的非线性特性。

Application of BP neural network model with fuzzy optimization in retrieval of biomass parameters

The retrieval of the biomass parameters from active/passive microwave remote sensing data （10.2 GHz） is performed based on an iterative inversion of BP neural network model with fuzzy optimization. The BP neural network is trained by a set of the measurements of active and passive remote sensing and the ground truth data versus Day of Year during growth. Once the network training is complete, the model can be used to retrieve the temporal variations of the biomass parameters from another set of observation data. The model was used in weights and microware observation data of wheat growth in 1989 to retrieve biomass parameters change of wheat growth this year. The retrieved biomass parameters correspond well with the real data of the growth, which shows that the BP model is scientific and sound.

基于智能计算的广西冷湿极端天气定性和定量组合预报方法研究

论文以逐日气温和降水量数据、NCEP/NCAR再分析资料以及预报场资料为基础,将表征冬季低温冷害的冷湿指数作为预报量,先利用随机森林方法进行冬季逐日冷湿极端天气定性判别预报分析,再进一步以粒子群算法为基础的模糊神经网络集成个体生成技术方法,建立一种新的非线性智能计算定量集成预报模型(PSOFNN),进行了广西冷湿极端天气定量预报模型的预报建模研究。结果表明,论文提出的这种以不同的智能计算方法构建的定性、定量综合预报分析方法,比较符合极端天气小概率事件的预报特点,其中随机森林算法构建的定性预报模型,对广西冷湿极端天气事件的预报TS评分(Threat Score)为0.77,空报率为0.23,漏报率为0,ETS评分(Equitable Threat Score)为0.41,TSS评分(True Skill Statistic)为0.53。而采用粒子群-模糊神经网络方法构建的极端冷湿指数定量集成预报模型比其他线性和非线性预报模型具有更好的预报精度。其中PSOFNN集成预报模型在预报建模样本和独立预报样本个例相同的情况下,比回归方法的预报平均绝对误差下降了25%以上,比一般的普通模糊神经网络预报平均绝对误差下降了14.37%。主要原因是因为PSO-FNN集成预报模型通过改进集成个体的预报能力和增强集成个体的种群差异性,提高了集成预报模型的预报精度。因此,该智能计算集成预报模型的泛化能力显著提高,预报结果稳定可靠,为冷湿极端天气客观预报提供了新的预报工具和预报建模方法。

Calculation of maximum surface settlement induced by EPB shield tunnelling and introducing most effective parameter

This study aims to predict ground surface settlement due to shallow tunneling and introduce the most affecting parameters on this phenomenon.Based on data collected from Shanghai LRT Line 2 project undertaken by TBM-EPB method,this research has considered the tunnel's geometric,strength,and operational factors as the dependent variables.At first,multiple regression(MR) method was used to propose equations based on various parameters.The results indicated the dependency of surface settlement on many parameters so that the interactions among different parameters make it impossible to use MR method as it leads to equations of poor accuracy.As such,adaptive neuro-fuzzy inference system(ANFIS),was used to evaluate its capabilities in terms of predicting surface settlement.Among generated ANFIS models,the model with all input parameters considered produced the best prediction,so as its associated R^2 in the test phase was obtained to be 0.957.The equations and models in which operational factors were taken into consideration gave better prediction results indicating larger relative effect of such factors.For sensitivity analysis of ANFIS model,cosine amplitude method(CAM) was employed; among other dependent variables,fill factor of grouting(n) and grouting pressure(P) were identified as the most affecting parameters.

Prediction model for surface layer microhardness of processed TC17 via high energy shot peening

The bulk TC17was subjected to the high energy shot peening(HESP)at the air pressures ranging from0.35to0.55MPa and processing durations ranging from15to60min.The microhardness(HV0.02)from topmost surface to matrix of the HESP processed TC17was measured,which generally decreases with the increase of depth from topmost surface to matrix and presents different variation with air pressure and processing duration at different depths.A fuzzy neural network(FNN)model was established to predict the surface layer microhardness of the HESP processed TC17,where the maximum and average difference between the measured and the predicted microhardness were respectively8.5%and3.2%.Applying the FNN model,the effects of the air pressure and processing duration on the microhardness at different depths were analyzed,revealing the significant interaction between the refined layer shelling and the continuous grain refinement.

Settlement modeling in central core rockfill dams by new approaches

One of the most important reasons for the serious damage of embankment dams is their impermissible settlement.Therefore,it can be stated that the prediction of settlement of a dam is of paramount importance.This study aims to apply intelligent methods to predict settlement after constructing central core rockfill dams.Attempts were made in this research to prepare models for predicting settlement of these dams using the information of 35 different central core rockfill dams all over the world and Adaptive Neuro-Fuzzy Interface System(ANFIS) and Gene Expression Programming(GEP) methods.Parameters such as height of dam(H) and compressibility index(Ci) were considered as the input parameters.Finally,a form was designed using visual basic software for predicting dam settlement.With respect to the accuracy of the results obtained from the intelligent methods,they can be recommended for predicting settlement after constructing central core rockfill dams for the future plans.

Temperature modeling and control of Direct Methanol Fuel Cell based on adaptive neural fuzzy technology

Aiming at on-line controlling of Direct Methanol Fuel Cell （DMFC） stack, an adaptive neural fuzzy inference technology is adopted in the modeling and control of DMFC temperature system. In the modeling process, an Adaptive Neural Fuzzy Inference System （ANFIS） identification model of DMFC stack temperature is developed based on the input-output sampled data, which can avoid the internal complexity of DMFC stack. In the controlling process, with the network model trained well as the reference model of the DMFC control system, a novel fuzzy genetic algorithm is used to regulate the parameters and fuzzy rules of a neural fuzzy controller. In the simulation, compared with the nonlinear Proportional Integral Derivative （PID） and traditional fuzzy algorithm, the improved neural fuzzy controller designed in this paper gets better performance, as demonstrated by the simulation results.

Electric Energy Management Modeling for Kingdom of Bahrain

In the deregulated economy, the maximum load forecasting is important for the electric industry. Many applications are included such as the energy generation and purchasing. The aim of the present study is to find the most suitable models for the peak load of the Kingdom of Bahrain. Many mathematical methods have been developing for maximum load forecasting. In the present paper, the modeling of the maximum load, population and GDP （gross domestic product） versus years obtained. The curve fitting technique used to find that models, where Graph 4.4.2 as a tool used to find the models. As well, Neuro-Fuzzy used to find the three models. Therefore, three techniques are used. These three are exponential, linear modeling and Neuro-Fuzzy. It is found that, the Neuro-Fuzzy is the most suitable and realistic one. Then, the linear modeling is the next suitable one.

A SELF-SIMILAR LOCAL NEURO-FUZZY MODEL FOR SHORT-TERM DEMAND FORECASTING

This paper proposes a selfsimilar local neurofuzzy （SSLNF） model with mutual informati onbased input selection algorithm for the shortterm electricity demand forecasting. The proposed self similar model is composed of a number of local models, each being a local linear neurofuzzy （LLNF） model, and their associated validity functions and can be interpreted itself as an LLNF model. The proposed model is trained by a nested local liner model tree （NLOLIMOT） learning algorithm which partitions the input space into axisorthogonal subdomains and then fits an LLNF model and its associated validity function on each subdomain. Furthermore, the proposed approach allows different input spaces for rule premises （validity functions） and consequents （local models）. This appealing property is employed to assign the candidate input variables （i.e., previous load and temperature） which influence shortterm electricity demand in linear and nonlinear ways to local models and validity functions, respectively. Numerical results from shortterm load forecasting in the New England in 2002 demonstrated the accuracy of the SSLNF model for the STLF applications.