Improved unequal interval grey model and its applications

A new method to improve prediction precision of GM(1,1) model with unequal time interval is presented.The grey derivative is multiplied by a parameter to guarantee the time response function satisfying approximately exponential function distribution.To simplify the process of parametric estimation,an approximate value is taken for the multiplied parameter.Then the estimators of coefficient of development and grey action quantity can be derived.At the same time,the principle of the new information priority is also considered.We take the last item of the first-order accumulated generation operator(1-AGO) on raw data sequence as the initial condition in the time response function.Then the new information can be taken full advantage of through the improved initial condition.Some properties of this new model are also discussed.The presented method is actually a combination of improvement of grey derivative and improvement of the initial condition.The results of an example indicate that the proposed method can improve prediction precision prominently.     

Reliability analysis of torpedo loading based on fractional-order optimization model

A fractional-order cumulative optimization GM(1,2)model based on grey theory is proposed to study the relationship between torpedo loading and working reliabilities. In this model, the average relative error function related to order and background value is established. Taking the average relative error function as the objective function, the optimal value of the two parameters is obtained through the optimization method,and the minimum value of the average relative error is determined. The calc...     

Comprehensive multivariate grey incidence degree based on principal component analysis

To overcome the too fine-grained granularity problem of multivariate grey incidence analysis and to explore the comprehensive incidence analysis model, three multivariate grey incidences degree models based on principal component analysis （PCA） are proposed. Firstly, the PCA method is introduced to extract the feature sequences of a behavioral matrix. Then, the grey incidence analysis between two behavioral matrices is transformed into the similarity and nearness measure between their feature sequences. Based on the classic grey incidence analysis theory, absolute and relative incidence degree models for feature sequences are constructed, and a comprehensive grey incidence model is proposed. Furthermore, the properties of models are researched. It proves that the proposed models satisfy the properties of translation invariance, multiple transformation invariance, and axioms of the grey incidence analysis, respectively. Finally, a case is studied. The results illustrate that the model is effective than other multivariate grey incidence analysis models.     

Interval grey number sequence prediction by using non-homogenous exponential discrete grey forecasting model

This paper aims to study a new grey prediction approach and its solution for forecasting the main system variable whose accurate value could not be collected while the potential value set could be defined. Based on the traditional nonhomogenous discrete grey forecasting model(NDGM), the interval grey number and its algebra operations are redefined and combined with the NDGM model to construct a new interval grey number sequence prediction approach. The solving principle of the model is analyzed, the new accuracy evaluation indices, i.e. mean absolute percentage error of mean value sequence(MAPEM) and mean percent of interval sequence simulating value set covered(MPSVSC), are defined and, the procedure of the interval grey number sequence based the NDGM(IG-NDGM) is given out. Finally, a numerical case is used to test the modelling accuracy of the proposed model. Results show that the proposed approach could solve the interval grey number sequence prediction problem and it is much better than the traditional DGM(1,1) model and GM(1,1) model.     

Modeling mechanism and extension of GM （1, 1）

Firstly,the research progress of grey model GM(1,1) is summarized,which is divided into three development stages: assimilation,alienation and melting stages.Then,the matrix analysis theory is used to study the modeling mechanism of GM(1,1),which decomposes the modeling data matrix into raw data transformation matrix,accumulated generating operation matrix and background value selection matrix.The changes of these three matrices are the essential reasons affecting the modeling and the accuracy of GM(1,1).Finally,the paper proposes a generalization grey model GGM(1,1),which is a extended form of GM(1,1) and also a unified form of model GM(1,1),model GM(1,1,α),stage grey model,hopping grey model,generalized accumulated model,strengthening operator model,weakening operator model and unequal interval model.And the theory and practical significance of the extended model is analyzed.     

Data processing of small samples based on grey distance information approach

Data processing of small samples is an important and valuable research problem in the electronic equipment test. Because it is difficult and complex to determine the probability distribution of small samples, it is difficult to use the traditional probability theory to process the samples and assess the degree of uncertainty. Using the grey relational theory and the norm theory, the grey distance information approach, which is based on the grey distance information quantity of a sample and the average grey distance information quantity of the samples, is proposed in this article. The definitions of the grey distance information quantity of a sample and the average grey distance information quantity of the samples, with their characteristics and algorithms, are introduced. The correlative problems, including the algorithm of estimated value, the standard deviation, and the acceptance and rejection criteria of the samples and estimated results, are also proposed. Moreover, the information whitening ratio is introduced to select the weight algorithm and to compare the different samples. Several examples are given to demonstrate the application of the proposed approach. The examples show that the proposed approach, which has no demand for the probability distribution of small samples, is feasible and effective.     

Collaborative optimization of maintenance and spare ordering of continuously degrading systems

A collaborative optimization model for maintenance and spare ordering of a single-unit degrading system is proposed in this paper based on the continuous detection. A gamma distribution is used to model the material degradation. The degrading decrement after the imperfect maintenance action is assumed as a random variable normal distribution. This model aims to obtain the optimal maintenance policy and spare ordering point with the expected cost rate within system lifecycle as the optimization objective. The rationality and feasibility of the model are proved through a numerical example.     

Fusion Method for Interval Number Multi-sensor Information

The interval numbers are used to types and observation of sensors, a new fusion represent the characteristic values of object method for multi-sensor object recognition is proposed from the viewpoint of decision making theory. The method defines the distance matrix and grey association matrix between all object types and unknown object. After solving the optimization problem of maximizing the standard deviations for all attributes, the weights of the attributes are obtained. Thus, the result of recognition for the unknown object is given by the grey association degree. This method avoids the subjectivity of selecting attributes weights. It is straightforward and can be performed on computer easily. The simulated example demonstrates the feasibility and effectiveness of the proposed method.     

Expansion modelling of discrete grey model based on multi-factor information aggregation

This paper aims to study a novel expansion discrete grey forecasting model, which could aggregate input information more effectively. In general, existing multi-factor grey forecasting models, such as one order and h variables grey forecasting model （GM （1, h））, always aggregate the main system variable and independent variables in a linear form rather than a nonlinear form, while a nonlinear form could be used in more cases than the linear form. And the nonlinear form could aggregate collinear independent factors, which widely lie in many multi-factor forecasting problems. To overcome this problem, a new approach, named as the Solow residual method, is proposed to aggregate independent factors. And a new expansion model, feedback multi-factor discrete grey forecasting model based on the Solow residual method （abbreviated as FDGM （1, h））, is proposed accordingly. Then the feedback control equation and the parameters＇ solution of the FDGM （1, h） model are given. Finally, a real application is used to test the modelling accuracy of the FDGM （1, h） model. Results show that the FDGM （1, h） model is much better than the nonhomogeneous discrete grey forecasting model （NDGM） and the GM （1, h） model.     

Combined model based on optimized multi-variable grey model and multiple linear regression

The construction method of background value is improved in the original multi-variable grey model (MGM(1,m)) from its source of construction errors.The MGM(1,m) with optimized background value is used to eliminate the random fluctuations or errors of the observational data of all variables,and the combined prediction model together with the multiple linear regression is established in order to improve the simulation and prediction accuracy of the combined model.Finally,a combined model of the MGM(1,2) with optimized background value and the binary linear regression is constructed by an example.The results show that the model has good effects for simulation and prediction.     

Mitigating end effects of EMD using non-equidistance grey model

Aiming at mitigating end effects of empirical mode decomposition (EMD), a new approach motivated by the nonequidistance grey model (NGM) termed as NGM(1,1) is proposed. Other than trapezoid formulas, the cubic Hermite spline is put forward to improve the accuracy of derivative to the accumulated generating operation (AGO) series. Hopefully, it is worth stressing that the proposed NGM(1,1) model is particularly useful for predicting uncertainty data. Qualitative and quantitative comparisons between the proposed approach and other well-known algorithms are carried out through computer simulations on synthetic as well as natural signals. Simulation results demonstrate the proposed method can reduce end effects and improve the decomposition results of EMD.     

Ranking grey numbers based on dominance grey degrees

With respect to the decision making problems where a lot of fuzzy and grey information always exists in the real-life decision making information system methods as fuzzy mathematics, it is difficult for such uncertainty probability, and interval numbers to deal with. To this end, based on the thought and method of grey numbers, grey degrees and interval numbers, the concept of dominance grey degree is defined. And then a method of ranking interval grey numbers based on the dominance grey degree is proposed. After discussing the relevant properties, the paper finally uses an example to demonstrate the effectiveness and applicability of the model. The result shows that the proposed model can more accurately describe uncertainty decision making problems, and realize the total ordering process for multiple-attribute decision-making problems.     

Improved Oustaloup approximation of fractional-order operators using adaptive chaotic particle swarm optimization

A rational approximation method of the fractional-order derivative and integral operators is proposed. The turning frequency points are fixed in each frequency interval in the standard Oustaloup approximation. In the improved Oustaloup method, the turning frequency points are determined by the adaptive chaotic particle swarm optimization (PSO). The average velocity is proposed to reduce the iterations of the PSO. The chaotic search scheme is combined to reduce the opportunity of the premature phenomenon. Two fitness functions are given to minimize the zero-pole and amplitude-phase frequency errors for the underlying optimization problems. Some numerical examples are compared to demonstrate the effectiveness and accuracy of this proposed rational approximation method.     

Equivalency and unbiasedness of grey prediction models

In order to deeply research the structure discrepancy and modeling mechanism among different grey prediction models, the equivalence and unbiasedness of grey prediction models are analyzed and verified. The results show that all the grey prediction models that are strictly derived from x(0)(k) +az(1)(k) = b have the identical model structure and simulation precision. Moreover, the unbiased simulation for the homogeneous exponential sequence can be accomplished. However, the models derived from dx(1)/dt + ax(1)= b are only close to those derived from x(0)(k) + az(1)(k) = b provided that |a| has to satisfy|a| < 0.1; neither could the unbiased simulation for the homogeneous exponential sequence be achieved. The above conclusions are proved and verified through some theorems and examples.     

Image segmentation algorithm based on high-dimension fuzzy character and restrained clustering network

An image segmentation algorithm of the restrained fuzzy Kohonen clustering network （RFKCN） based on high- dimension fuzzy character is proposed. The algorithm includes two steps. The first step is the fuzzification of pixels in which two redundant images are built by fuzzy mean value and fuzzy median value. The second step is to construct a three-dimensional （3-D） feature vector of redundant images and their original images and cluster the feature vector through RFKCN, to realize image seg- mentation. The proposed algorithm fully takes into account not only gray distribution information of pixels, but also relevant information and fuzzy information among neighboring pixels in constructing 3- D character space. Based on the combination of competitiveness, redundancy and complementary of the information, the proposed algorithm improves the accuracy of clustering. Theoretical anal- yses and experimental results demonstrate that the proposed algorithm has a good segmentation performance.     

Shape control on probability density function in stochastic systems

A novel strategy of probability density function （PDF） shape control is proposed in stochastic systems. The control er is designed whose parameters are optimal y obtained through the improved particle swarm optimization algorithm. The parameters of the control er are viewed as the space position of a particle in particle swarm optimization algorithm and updated continual y until the control er makes the PDF of the state variable as close as possible to the expected PDF. The proposed PDF shape control technique is compared with the equivalent linearization technique through simulation experiments. The results show the superiority and the effectiveness of the proposed method. The control er is excellent in making the state PDF fol ow the expected PDF and has the very smal error between the state PDF and the expected PDF, solving the control problem of the PDF shape in stochastic systems effectively.     

Adjustable entropy function method for support vector machine

Based on KKT complementary condition in optimization theory, an unconstrained non-differential optimization model for support vector machine is proposed. An adjustable entropy function method is given to deal with the proposed optimization problem and the Newton algorithm is used to figure out the optimal solution. The proposed method can find an optimal solution with a relatively small parameter p, which avoids the numerical overflow in the traditional entropy function methods. It is a new approach to solve support vector machine. The theoretical analysis and experimental results illustrate the feasibility and efficiency of the proposed algorithm.     

Inspection interval optimization for aircraft composite structures with dent and delamination damage

The optimization of inspection intervals for composite structures has been proposed,but only one damage type,dent damage,has been addressed so far.The present study focuses on the two main damage types of dent and delamination,and a model for optimizing the inspection interval of composite structures is proposed to minimize the total maintenance cost on the premise that the probability of structure failure will not exceed the acceptable level.In order to analyze the damage characteristics and the residual strength of the composite structure,the frequency,energy,size,and depth of the damage are studied,and the situation of missing detection during the inspection is considered.The structural residual strength and total maintenance cost are quantified corresponding to different inspection intervals.The proposed optimization method relieves the constraints in previous simulation methods,and is more consistent with the actual situation.Finally,the outer wing of aircraft is taken as an example,and with the historical cases and experimental data,the optimization method is verified.The optimal inspection interval is shorter than the actually implemented inspection interval,and the corresponding maintenance cost is reduced by 23.3%.The result shows the feasibility and effectiveness of the proposed optimization method.     

Multi-parameters uncertainty analysis of logistic support process based on GERT

The uncertainty analysis is an effective sensitivity analysis method for system model analysis and optimization. However,the existing single-factor uncertainty analysis methods are not well used in the logistic support systems with multiple decision-making factors. The multiple transfer parameters graphical evaluation and review technique（MTP-GERT） is used to model the logistic support process in consideration of two important factors, support activity time and support activity resources, which are two primary causes for the logistic support process uncertainty. On this basis,a global sensitivity analysis（GSA） method based on covariance is designed to analyze the logistic support process uncertainty. The aircraft support process is selected as a case application which illustrates the validity of the proposed method to analyze the support process uncertainty, and some feasible recommendations are proposed for aircraft support decision making on carrier.     

Consistency analysis of accelerated degradation mechanism based on gray theory

A fundamental premise of an accelerated testing is that the failure mechanism under elevated and normal stress levels should remain the same. Thus, verification of the consistency of failure mechanisms is essential during an accelerated testing. A new consistency analysis method based on the gray theory is pro- posed for complex products. First of all, existing consistency ana- lysis methods are reviewed with a focus on the comparison of the differences among them. Then, the proposed consistency ana- lysis method is introduced. Two effective gray prediction models, gray dynamic model and new information and equal dimensional （NIED） model, are adapted in the proposed method. The process to determine the dimension of NIED model is also discussed, and a decision rule is expanded. Based on that, the procedure of ap- plying the new consistent analysis method is developed. Finally, a case study of the consistency analysis of a reliability enhancement testing is conducted to demonstrate and validate the proposed method.