Zhao F.Q.,Lanzhou University of Technology |
Zhao F.Q.,Key Laboratory of Gansu Advanced Control for Industrial Process |
Tang J.X.,Lanzhou University of Technology |
Yang Y.H.,Lanzhou University of Technology
Journal of Computers | Year: 2012
Manufacturing supply chain(SC) faces changing business environment and various customer demands. Pareto Ant Colony Optimisation (P-ACO) in order to obtain the non-dominated set of different SC designs was utilized as the guidance for designing manufacturing SC. PACO explores the solution space on the basis of applying the Ant Colony Optimisation algorithm and implementing more than one pheromone matrix, one for every objective. The SC design problem has been addressed by using Pareto Ant Colony Optimisation in which two objectives are minimised simultaneously. There were tested two ways in which the quantity of pheromones in the PM is incremented. In the SPM, the pheromone increment is a function of the two objectives, cost and time, while in MPM the pheromone matrix is divided into two pheromones, one for the cost and another one for the time. It could be concluded that the number of solutions do not depend on if the pheromone is split or is a function of the two variables because both method explore the same solution space. Although both methods explore the same solution space, the POS generated by every one is different. The POS that is generated when the pheromone matrix is split got solutions with lower time and cost than SMP because in the probabilistic decision rule a value of λ = 0.2 is used. It means that the ants preferred solution with a low cost instead of solutions with low time. The strategy of letting the best-so-far ant deposit pheromone over the PM accelerates the algorithm to get the optimal POS although the number of ants in the colony is small. An experimental example is used to test the algorithm and show the benefits of utilising two pheromone matrices and multiple ant colonies in SC optimisation problem. © 2012 ACADEMY PUBLISHER.
Zhao F.,Lanzhou University of Technology |
Zhao F.,Northwestern Polytechnical University |
Zhao F.,Key Laboratory of Gansu Advanced Control for Industrial Process |
Yang Y.,Lanzhou University of Technology |
And 2 more authors.
Advanced Science Letters | Year: 2012
Dynamic rescheduling model and its solution method are of significant importance for the dynamic scheduling problem in manufacturing system. However, few attempts have been done on the universal communication and negotiation mechanism for the dynamic rescheduling problem and corresponding solution approach. A dynamic rescheduling model, which is based on Multi-Agent System (MAS), was proposed. A memetic algorithm with PSO (particle swarm optimization) and DE (differential evolution) was presented as the solution method to the rescheduling model. Furthermore, the simulation results in dynamic scheduling accompanying with its perturbation show that the proposed model and the algorithm are effective to the dynamic scheduling problem in manufacturing system. © 2012 American Scientific Publishers.
Zhang M.-G.,Lanzhou University of Technology |
Zhang M.-G.,Key Laboratory of Gansu Advanced Control for Industrial Process |
Li L.-R.,Lanzhou University of Technology
PEAM 2011 - Proceedings: 2011 IEEE Power Engineering and Automation Conference | Year: 2011
Load forecasting, especially short-term load forecasting is of great significance for the planning, scheduling, marketing of power system. In order to predict the daily load as accurate as possible,a combined prediction method based on Least Squares Support Vector Machine (LS-SVM) and BP Neural Network (BPNN) is proposed in this paper. The historical load of relational better six-day ahead and the day type are selected as input,and got 1-dimensional output variable. Two sets of different prediction results are obtained from LS-SVM method and BPNN method, which is combined by using the method of minimum variance to research the final prediction results. The load prediction results of northwest grid show that the combined forecasting method has better prediction accuracy than LS-SVM and BPNN method. Therefore, this method is efficient and practical for a short-term load forecasting of electric power system. © 2011 IEEE.
Zhao F.G.,Lanzhou University of Technology |
Zhao F.G.,Key Laboratory of Gansu Advanced Control for Industrial Process |
Li X.Y.,Lanzhou University of Technology |
Zhang Q.Y.,Lanzhou University of Technology |
Jonrinaldi,University of Exeter
Journal of Software | Year: 2012
For the whole matching cannot handle partial occlusion and lack of specificity, a new method using Polar- Radius-Invariant-Moment, which is based on Key-Points to extract features for target's shape recognition, is presented in this paper. Firstly, key-points of the hand shape are extracted through discrete curve evolution method. Secondly, Polar-Radius-Invariant-Moment based on Key- Points is used to describe the characteristics of the gesture shape. Finally, Euclidean distance is utilized in gesture recognition to verify the validity of this method. Hand objects are selected as the test case to testify the performance of the method. Simulation results prove that this method has a better classification character than that of obtained by the Polar-Radius-Invariant-Moment with recognizing the object shape rapidly and accurately, and that it also can keep highly stable even if the object contour was ill-segmented or noisy. © 2012 ACADEMY PUBLISHER.
Li W.,Lanzhou University of Technology |
Li W.,Key Laboratory of Gansu Advanced Control for Industrial Process |
Cheng Y.,Lanzhou University of Technology |
Xu D.,Lanzhou University of Technology
Huazhong Keji Daxue Xuebao (Ziran Kexue Ban)/Journal of Huazhong University of Science and Technology (Natural Science Edition) | Year: 2011
Method of inverse system for fault tolerance was presented, in which the method of nonlinear uncertain systems was the only part of the inverse reversible. After a robust sliding mode observer was introduced, the irreversible part and the actuator failures were estimated on-line. Thus modified inverse model was compensated and the faults were adjusted on-line, so that it accurately in numerical approximation on the non-linear uncertainty of the inverse model of the controlled object. Simulation results show that the proposed method not only for the non-explicit multivariable unknown nonlinear uncertain system has good control effect, but also in case of actuator failures have good fault-tolerance effect.