The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The paper combines rough sets and ANN to analyze inventory early-warning in supply chains. The introduction of Rough sets cuts down the input dimensions of ANN, and the ANN algorithm is improved by adding the momentum factor mc and applying adaptive learning rate. Lastly, according to the inventory data of a manufacturing enterprise in Handan City, the paper proves the validity of the proposed model.
In safety-critical systems, components must be highly reliable. Once a component's reliability is lower than the criterion, it will be replaced and transferred to be used in another system which does not need such high reliability. The safety virtual age criterion not only relates with the cost of repair actions, but also with the repair degrees and the systempsilas wearing degrees. In this paper,...
Construction projects are one-off endeavors with many unique features such as long period, complicated processes, abominable environment, financial intensity and dynamic organization structures and such organizational and technological complexity generates enormous risks. And environmental and economical dimensions throughout the life cycle of the construction project, so the construction project...
By rough programming, we mean the optimization theory dealing with rough decision problems. This paper constructs a general framework of rough chance-constrained programming. We also design a spectrum of rough simulations for computing uncertain functions arising in the area of rough programming. To speed up the process of handling uncertain functions, we train a neural network to approximate uncertain...
This paper deals with load optimal dispatching among power plant units in the context of electric power market competition environment. The multi-agent evolutionary algorithm (MAEA) is specially useful in load optimal dispatching, the model of which is set up based on characteristic curve of the generator coal consumption approached by quadratic function. In particular, the following restraints are...
A comprehensive and objective post-evaluation of rural electric network reformation (RENR) will bring a scientific suggestion for decision-making with the financing is shortage. In this paper, on the base of analyzing methods used in post-evaluation before, a new model, which integrated with fuzzy theory, interval number and analytic hierarchy process (AHP), is proposed. Multi-factors and information...
Because power loads are influenced by various factors, and the changes of power load presents are complicate, the traditional forecasting methods are always not satisfied. According to the random-increase and non-linearity fluctuation of residual series, gray neural network forecasting can reflect the increase character and non-linearity relationship. This paper using the improved ACO method as the...
According to the low sample and multifactor impact for long-medium term power load forecasting, the grey relational grade was used in screening factors, the combined model in BP neural network and SVM was established, and the multivariate variables and history load variables were used to roll prediction. The combined predictive values are obviously better than single method. Empirical study showed...
With the continuous deepening of the power system reform and the blackouts of someplace on the world, the safety of the power grid has received high attention from all sections of the society. The former researches on the power grid safety are mostly about special parts, the method to estimate the whole power grid safety should be improved in the future. In this paper, according to the characters...
The paper adopts rough set reduction algorithm to reduce the influence factors of power plant selection and eliminate the uncorrelated attribution, through which we can obtain typical samples. After this, adopting fuzzy method to calculate the membership degree of the typical samples, which are looked on as the input of BP Neural Network and the expert values are as the expected output to train the...
Through collecting official statistics as much as possible, 2005 social accounting matrix (SAM) of China is built up. Based on the 2005SAM, the method of matrix multiplier is adopted in the analysis of the impact of electricity price adjustment on the departments of national economy and the quantification of the impact. It is revealed that the impact of multiplier on capital account (the expenditure...
The paper proposes short-term power load forecasting model based on fuzzy RBF neural network, it has overcome the BP algorithm's disadvantage of slow convergence rate and it fall into partially the smallest insufficiency easily. RBF network model in the use of the latest neighborhood clustering algorithm, and the network structure and the parameters are double-adjusted and the training speed and forecast...
This paper optimizes the wavelet neural networks with genetic algorithms which has the optimization of the overall search capabilities, and establishes the model of wavelet neural networks based on genetic algorithms. It overcomes the shortcomings of BP neural network for their own, and it can get higher accuracy and faster convergence. The examples also show that the model can improve forecast accuracy...
An improved BP Neural Network with additional momentum and adaptive learning is proposed in the paper to predict the growth rate of electricity consumption in China. Matlab7 is used as modeling tool to design the model. Current year GDP growth, electric power consumption growth and growth rate of secondary industry are taken as input variables while next year electric power consumption growth is predicted...
It is proved that the individual risk models for a single period are equivalent to the collective risk models for a single period when the number-of-claim random variable has 0-1 distribution. Furthermore, the individual risk model is close to a compound Poisson distribution when the number-of-claim random variable has a Poisson distribution with small parameter. Applying the Panjer iteration as calculation...
We have developed a model to investigate the technology innovation effects of increased outsourcing of production on a low cost region. Increase in profits and reduction in the resource requirement by adapting related innovation could spur the increase of outsourcing. Firms in lower cost region can develop the capacity of innovation through dasialearning by doingpsila and even investment in R&D...
The paper proposes credit risk assessment model of commercial banks based on fuzzy probabilistic neural network model (FPNN) which combines the relative membership degree in fuzzy mathematics with Probabilistic Neural Network (PNN). The model makes up for a deficiency of ANN and BP arithmetic. Finally, an example is used to prove the calculation of this method is succinct rapid, and the evaluative...
A new triangular fuzzy number critical path method was proposed based on linear programming for solving the uncertain project problem. alpha -cut set was used to transfer the fuzzy number to interval number, and a credibility degree was defined so that the total time range and the different critical paths were given by adjusting the credibility degree. Meanwhile, by analyzing the sensitivity coefficient,...
The use of neural networks (NNs) for financial applications is quite common because of their excellent performances of treating non-linear data with self-learning capability. Often arises the problem of a black-box approach,i.e. after having trained neural networks for a particular problem, it is almost impossible to analyse them for how they work. The Fuzzy Neural Networks(FNN) allow to add rules...
Although some of these methods lead to models with a satisfactory ability to discriminate between healthy and bankrupt firms, they suffer from some limitations, often due to only give an alarm, but cannot forecast. This is why we have undertaken a research aiming at weakening these limitations. We propose an Exponential Smoothing Forecasting and Pattern Recognition (ESFPR) approach and illustrate...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.