Search

The prediction of port throughput is very complicate, and its accuracy is affected by many socio-economic factors, particularly affected by their embedded distributed randomness of these factors and mixed noises produced in the processes of data collection, transformation, and calculation. Firstly, in view of the v-support vector regression hybridized with Gauss function (briefed as Gauss- vSVR model), to well solve the nonlinear and mixed noises, this paper uses this model to simulate the nonlinear evolving system of port throughput series. Then, to look for more suitable parameter combination of this model and take into account that GA still suffers from the problems of trapped into local optima and time-consuming, this study integrates the global chaotic perturbation algorithm by using Cat mapping function and local acceleration search algorithm by employing cloud theory, i.e., abbreviated as chaotic cloud genetic algorithm (CCGA), to well determine the parameter values for an Gauss- vSVR model. Additionally, based on the principal component analysis and correlation analysis method, an input vector decision method (namely IVD) is proposed to identify the final input variables for Gauss- vSVR model. Finally, hybridization of IVD and CCGA with Gauss- vSVR model, namely IG vSVR-CCGA, is proposed for port throughput forecasting. Subsequently, the port throughput data and its associate socio-economic factors of two largest Chinese ports, Shanghai Port and Tianjin Port, are employed as practical examples to test forecast performance. The numerical results indicate that the proposed hybrid forecasting model receives more satisfied forecasting performance than other classical prediction models; in the meanwhile, the CCGA algorithm also obtains higher optimal efficiency than other alternative algorithms. [ABSTRACT FROM AUTHOR]

The prediction of vessel traffic flow is complicated, its accuracy is influenced by uncertain socio-economic factors, especially by the singular points existed in the statistical data. Recently, the robust v -support vector regression model ( R SVR) has been successfully employed to solve non-linear regression and time-series problems with the singular points. This paper will firstly propose a novel hybrid algorithm, namely chaotic cloud simulated annealing genetic algorithm (C cat CSAGA) for optimizing the parameters of R SVR, to improve the performance of vessel traffic flow prediction. In which, the proposed C cat CSAGA employs cat mapping to carefully expand variable searching space, to overcome premature local optimum, and uses cloud model efficiently to search a better solution in a small neighborhood of the current optimal solution, to improve the search efficiency. Secondly, the kernel principal component analysis (KPCA) algorithm is adopted to determine the final input vectors from the candidate input variables. Finally, a numerical example of vessel traffic flow and its influence factors data from Tianjin are employed to test the forecasting performance of the proposed K R SVR-C cat CSAGA model. [ABSTRACT FROM AUTHOR]

To improve urban passenger prediction accuracy with parametric projection pursuit regression, acceleration genetic algorithm was improved with cat map, Gaussian distribution and local searching. A new chaos accelerating genetic algorithm (NCAGA) was presented, used to optimize the best projection direction a of PPPR model. A hybrid algorithm of NCAGA-PPPR urban passenger forecasting model was proposed, in which the best projection direction was hybrid optimized inner by the NCAGA at the time of optimizing outer the number of ridge functions M. The simulation prediction was made with observed data, compared with BP neural network model, traditional PPR model and PPPR model optimized by acceleration genetic algorithm. The urban passenger forecasting accuracy is higher than the others, which the mean absolute relative error is less than 3.1%. The new hybrid algorithm can improve prediction accuracy of urban passenger and can be used efficaciously to forecast the urban passenger. [ABSTRACT FROM AUTHOR]

Aiming at the prediction of passenger traffic volume with small samples, multi-dimension and nonlinearity, v-support vector regression (v-SVR) is introduced to forecast passenger traffic volume. To seek the optimal forecast accuracy and generalization performance of v-SVR, chaos adaptive genetic algorithm (CAGA) is used to optimize the parameter, which is based on chaos mapping and adaptive mechanism. Then, a new passenger traffic volume forecasting model of v-SVR named by CAGA-v--SVR is proposed. The model is applied to forecasting passenger traffic volume with data of 1978-2008. Compared with RBF neural network model, GA-SVR model and GA-v-SVR model, it is concluded that CAGA-v-SVR prediction model has higher prediction precision, and can effectively predict passenger traffic volume with less than 2. 3% of mean absolute relative error. [ABSTRACT FROM AUTHOR]

Compared with a large power grid, a microgrid electric load (MEL) has the characteristics of strong nonlinearity, multiple factors, and large fluctuation, which lead to it being difficult to receive more accurate forecasting performances. To solve the abovementioned characteristics of a MEL time series, the least squares support vector machine (LS-SVR) hybridizing with meta-heuristic algorithms is applied to simulate the nonlinear system of a MEL time series. As it is known that the fruit fly optimization algorithm (FOA) has several embedded drawbacks that lead to problems, this paper applies a quantum computing mechanism (QCM) to empower each fruit fly to possess quantum behavior during the searching processes, i.e., a QFOA algorithm. Eventually, the cat chaotic mapping function is introduced into the QFOA algorithm, namely CQFOA, to implement the chaotic global perturbation strategy to help fruit flies to escape from the local optima while the population's diversity is poor. Finally, a new MEL forecasting method, namely the LS-SVR-CQFOA model, is established by hybridizing the LS-SVR model with CQFOA. The experimental results illustrate that, in three datasets, the proposed LS-SVR-CQFOA model is superior to other alternative models, including BPNN (back-propagation neural networks), LS-SVR-CQPSO (LS-SVR with chaotic quantum particle swarm optimization algorithm), LS-SVR-CQTS (LS-SVR with chaotic quantum tabu search algorithm), LS-SVR-CQGA (LS-SVR with chaotic quantum genetic algorithm), LS-SVR-CQBA (LS-SVR with chaotic quantum bat algorithm), LS-SVR-FOA, and LS-SVR-QFOA models, in terms of forecasting accuracy indexes. In addition, it passes the significance test at a 97.5% confidence level. [ABSTRACT FROM AUTHOR]