A generalized 3-D four-wing chaotic system
- Authors: Wang, Zenghui
- Date: 2009
- Subjects: Chaos , Four-wing chaos , Lyapunov exponents , Bifurcation
- Language: English
- Type: Article
- Identifier: http://hdl.handle.net/10210/17750 , uj:15920 , Citation: Wang, Z. et al. 2009. A generalized 3-D four-wing chaotic system. International journal of bifurcation and chaos, 19(11): 3841-3853. , Abstract: In this paper, several three-dimentional (3-D) four-wing smooth quadratic autonomous chaotic systems are analyzed. It is shown that these systems have similar features.
- Full Text: false
- Authors: Wang, Zenghui
- Date: 2009
- Subjects: Chaos , Four-wing chaos , Lyapunov exponents , Bifurcation
- Language: English
- Type: Article
- Identifier: http://hdl.handle.net/10210/17750 , uj:15920 , Citation: Wang, Z. et al. 2009. A generalized 3-D four-wing chaotic system. International journal of bifurcation and chaos, 19(11): 3841-3853. , Abstract: In this paper, several three-dimentional (3-D) four-wing smooth quadratic autonomous chaotic systems are analyzed. It is shown that these systems have similar features.
- Full Text: false
A hyperchaotic system without equilibrium
- Wang, Zenghui, Cang, Shijian, Ochola, Elisha Oketch, Sun, Yanxia
- Authors: Wang, Zenghui , Cang, Shijian , Ochola, Elisha Oketch , Sun, Yanxia
- Date: 2012
- Subjects: Hyperchaos , Chaos , Lyapunov exponents , Poincare map , Equilibrium
- Type: Journal article
- Identifier: http://hdl.handle.net/10210/20529 , uj:16109 , Citation: Wang, Z. et al. 2012. A hyperchaotic system without equilibrium. Nonlinear Dynamics 69(2012):531–537. DOI: 10.1007/s11071-011-0284-z
- Description: Abstract: This article introduces a new chaotic system of 4-D autonomous ordinary differential equations, which has no equilibrium. This system shows a hyper-chaotic attractor. There is no sink in this system as there is no equilibrium. The proposed system is investigated through numerical simulations and analyses including time phase portraits, Lyapunov exponents, and Poincaré maps. There is little difference between this chaotic system and other chaotic systems with one or several equilibria shown by phase portraits, Lyapunov exponents and time series methods, but the Poincaré maps show this system is a chaotic system with more complicated dynamics. Moreover, the circuit realization is also presented.
- Full Text:
- Authors: Wang, Zenghui , Cang, Shijian , Ochola, Elisha Oketch , Sun, Yanxia
- Date: 2012
- Subjects: Hyperchaos , Chaos , Lyapunov exponents , Poincare map , Equilibrium
- Type: Journal article
- Identifier: http://hdl.handle.net/10210/20529 , uj:16109 , Citation: Wang, Z. et al. 2012. A hyperchaotic system without equilibrium. Nonlinear Dynamics 69(2012):531–537. DOI: 10.1007/s11071-011-0284-z
- Description: Abstract: This article introduces a new chaotic system of 4-D autonomous ordinary differential equations, which has no equilibrium. This system shows a hyper-chaotic attractor. There is no sink in this system as there is no equilibrium. The proposed system is investigated through numerical simulations and analyses including time phase portraits, Lyapunov exponents, and Poincaré maps. There is little difference between this chaotic system and other chaotic systems with one or several equilibria shown by phase portraits, Lyapunov exponents and time series methods, but the Poincaré maps show this system is a chaotic system with more complicated dynamics. Moreover, the circuit realization is also presented.
- Full Text:
A machine-learning based nonintrusive smart home appliance status recognition
- Matindife, Liston, Sun, Yanxia, Wang, Zenghui
- Authors: Matindife, Liston , Sun, Yanxia , Wang, Zenghui
- Date: 2020
- Language: English
- Type: Article
- Identifier: http://hdl.handle.net/10210/450398 , uj:39589 , Citation: Matindife, L., Sun, Y. & Wang, Z. 2020. A machine-learning based nonintrusive smart home appliance status recognition. , DOI: https://doi.org/10.1155/2020/9356165
- Description: Abstract: In a smart home, the nonintrusive load monitoring recognition scheme normally achieves high appliance recognition performance in the case where the appliance signals have widely varying power levels and signature characteristics. However, it becomes more difficult to recognize appliances with equal or very close power specifications, often with almost identical signature characteristics. In literature, complex methods based on transient event detection and multiple classifiers that operate on different hand crafted features of the signal have been proposed to tackle this issue. In this paper, we propose a deep learning approach that dispenses with the complex transient event detection and hand crafting of signal features to provide high performance recognition of close tolerance appliances. ,e appliance classification is premised on the deep multilayer perceptron having three appliance signal parameters as input to increase the number of trainable samples and hence accuracy. In the case where we have limited data, we implement a transfer learning-based appliance classification strategy. With the view of obtaining an appropriate high performing disaggregation deep learning network for the said problem, we explore individually three deep learning disaggregation algorithms based on the multiple parallel structure convolutional neural networks, the recurrent neural network with parallel dense layers for a shared input, and the hybrid convolutional recurrent neural network. We disaggregate a total of three signal parameters per appliance in each case. To evaluate the performance of the proposed method, some simulations and comparisons have been carried out, and the results show that the proposed method can achieve promising performance.
- Full Text:
- Authors: Matindife, Liston , Sun, Yanxia , Wang, Zenghui
- Date: 2020
- Language: English
- Type: Article
- Identifier: http://hdl.handle.net/10210/450398 , uj:39589 , Citation: Matindife, L., Sun, Y. & Wang, Z. 2020. A machine-learning based nonintrusive smart home appliance status recognition. , DOI: https://doi.org/10.1155/2020/9356165
- Description: Abstract: In a smart home, the nonintrusive load monitoring recognition scheme normally achieves high appliance recognition performance in the case where the appliance signals have widely varying power levels and signature characteristics. However, it becomes more difficult to recognize appliances with equal or very close power specifications, often with almost identical signature characteristics. In literature, complex methods based on transient event detection and multiple classifiers that operate on different hand crafted features of the signal have been proposed to tackle this issue. In this paper, we propose a deep learning approach that dispenses with the complex transient event detection and hand crafting of signal features to provide high performance recognition of close tolerance appliances. ,e appliance classification is premised on the deep multilayer perceptron having three appliance signal parameters as input to increase the number of trainable samples and hence accuracy. In the case where we have limited data, we implement a transfer learning-based appliance classification strategy. With the view of obtaining an appropriate high performing disaggregation deep learning network for the said problem, we explore individually three deep learning disaggregation algorithms based on the multiple parallel structure convolutional neural networks, the recurrent neural network with parallel dense layers for a shared input, and the hybrid convolutional recurrent neural network. We disaggregate a total of three signal parameters per appliance in each case. To evaluate the performance of the proposed method, some simulations and comparisons have been carried out, and the results show that the proposed method can achieve promising performance.
- Full Text:
A new golden ratio local search based particle swarm optimization
- Sun, Yanxia, Van Wyk, Barend Jacobus, Wang, Zenghui
- Authors: Sun, Yanxia , Van Wyk, Barend Jacobus , Wang, Zenghui
- Date: 2012
- Subjects: Particle swarm optimization , Golden ratio , Local search
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/21329 , uj:16141 , Citation: Sun, Y., Van Wyk, B.J. & Wang, Z. 2012. A new golden ratio local search based particle swarm optimization. 2012 International Conference on Systems and Informatics (ICSAI 2012), Yantai, China 19-20 May, 2012. p. 754-757. ISBN: 978-1-4673-0198-5
- Description: Abstract: At beginning of the search process of particle swarm optimization, one of the disadvantages is that PSO focuses on the global search while the local search is weaken. However, at the end of the search procedure, the PSO focuses on the local search as all most all the particles converge to small areas which may make the particle swarm trapped in the local minima if no particle find position near the minima at the beginning of search procedure. To improve the optimization performance, the local search is necessary for particle swarm optimization. In this paper, golden ratio is used to determine the size of the search area. Only two positions need to be checked to find whether there are local positions with lower fitness value around a certain particle position. This method is easy to use. It is also tested using several famous benchmarks with high dimensions and big search space to the efficiency of the proposed method.
- Full Text:
- Authors: Sun, Yanxia , Van Wyk, Barend Jacobus , Wang, Zenghui
- Date: 2012
- Subjects: Particle swarm optimization , Golden ratio , Local search
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/21329 , uj:16141 , Citation: Sun, Y., Van Wyk, B.J. & Wang, Z. 2012. A new golden ratio local search based particle swarm optimization. 2012 International Conference on Systems and Informatics (ICSAI 2012), Yantai, China 19-20 May, 2012. p. 754-757. ISBN: 978-1-4673-0198-5
- Description: Abstract: At beginning of the search process of particle swarm optimization, one of the disadvantages is that PSO focuses on the global search while the local search is weaken. However, at the end of the search procedure, the PSO focuses on the local search as all most all the particles converge to small areas which may make the particle swarm trapped in the local minima if no particle find position near the minima at the beginning of search procedure. To improve the optimization performance, the local search is necessary for particle swarm optimization. In this paper, golden ratio is used to determine the size of the search area. Only two positions need to be checked to find whether there are local positions with lower fitness value around a certain particle position. This method is easy to use. It is also tested using several famous benchmarks with high dimensions and big search space to the efficiency of the proposed method.
- Full Text:
A new multi-swarm multi-objective particle swarm optimization based on pareto front set
- Sun, Yanxia, Van Wyk, Barend Jacobus, Wang, Zenghui
- Authors: Sun, Yanxia , Van Wyk, Barend Jacobus , Wang, Zenghui
- Date: 2011
- Subjects: Multi-objective optimization , Particle swarm optimization , Multiple swarms , Pareto front
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/22263 , uj:16181 , Citation: Sun, Y., Van Wyk, B.J. & Wang, Z. 2011. A new multi-swarm multi-objective particle swarm optimization based on pareto front set. Lecture Notes in Artificial Intelligence (LNAI) 6839:203-210. ISBN:978-3-642-25944-9
- Description: Abstract: In this paper, a new multi-swarm method is proposed for multiobjective particle swarm optimization. To enhance the Pareto front searching ability of PSO, the particles are divided into many swarms. Several swarms are dynamically searching the objective space around some points of the Pareto front set. The rest of particles are searching the space keeping away from the Pareto front to improve the global search ability. Simulation results and comparisons with existing Multi-objective Particle Swarm Optimization methods demonstrate that the proposed method effectively enhances the search efficiency and improves the search quality. , Originally presented at 2011 International Conference on Intelligent Computing, Zhengzhou, China 11-14 August, 2011.
- Full Text:
- Authors: Sun, Yanxia , Van Wyk, Barend Jacobus , Wang, Zenghui
- Date: 2011
- Subjects: Multi-objective optimization , Particle swarm optimization , Multiple swarms , Pareto front
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/22263 , uj:16181 , Citation: Sun, Y., Van Wyk, B.J. & Wang, Z. 2011. A new multi-swarm multi-objective particle swarm optimization based on pareto front set. Lecture Notes in Artificial Intelligence (LNAI) 6839:203-210. ISBN:978-3-642-25944-9
- Description: Abstract: In this paper, a new multi-swarm method is proposed for multiobjective particle swarm optimization. To enhance the Pareto front searching ability of PSO, the particles are divided into many swarms. Several swarms are dynamically searching the objective space around some points of the Pareto front set. The rest of particles are searching the space keeping away from the Pareto front to improve the global search ability. Simulation results and comparisons with existing Multi-objective Particle Swarm Optimization methods demonstrate that the proposed method effectively enhances the search efficiency and improves the search quality. , Originally presented at 2011 International Conference on Intelligent Computing, Zhengzhou, China 11-14 August, 2011.
- Full Text:
A new multi-swarm multi-objective particle swarm optimization based power and supply voltage unbalance optimization of three-phase submerged arc furnace
- Authors: Sun, Yanxia , Wang, Zenghui
- Date: 2015
- Subjects: Multi-objective optimization , Particle swarm optimization , Submerged arc furnace , Power optimization , Supply voltage unbalances
- Language: English
- Type: Conference proceedings
- Identifier: http://ujcontent.uj.ac.za8080/10210/390077 , http://hdl.handle.net/10210/20559 , uj:16112 , Citation: Sun, Y. & Wang, Z. 2015. A new multi-swarm multi-objective particle swarm optimization based power and supply voltage unbalance optimization of three-phase submerged arc furnace. Proceedings of the Sixth International Conference on Swarm Intelligence (ICSI 2015), Beijing, China, 25-29, June 2015.
- Description: Abstract: To improve the production ability of a three-phase submerged arc furnace (SAF), it is necessary to maximize the power input; minimize the supply voltage unbalances to reduce the side effect of the power grids. In this paper, maximizing the power input and minimum the supply voltage unbalances based on a proposed multi-swarm multi-objective particle swarm optimization algorithm are focused on. It is necessary to have objective functions when an optimization algorithm is applied. However, it is difficult to get the mathematic model of a three-phase submerged arc furnace according to its mechanisms because the system is complex and there are many disturbances. The neural networks (NN) have been applied since its ability can be used as an arbitrary function approximation mechanism based on the observed data. Based on the Pareto front, a multi-swarm multi-objective particle swarm optimization is pro-posed, which can be used to optimize the NN model of the three-phase SAF. The optimization results showed the efficiency of the proposed method.
- Full Text:
- Authors: Sun, Yanxia , Wang, Zenghui
- Date: 2015
- Subjects: Multi-objective optimization , Particle swarm optimization , Submerged arc furnace , Power optimization , Supply voltage unbalances
- Language: English
- Type: Conference proceedings
- Identifier: http://ujcontent.uj.ac.za8080/10210/390077 , http://hdl.handle.net/10210/20559 , uj:16112 , Citation: Sun, Y. & Wang, Z. 2015. A new multi-swarm multi-objective particle swarm optimization based power and supply voltage unbalance optimization of three-phase submerged arc furnace. Proceedings of the Sixth International Conference on Swarm Intelligence (ICSI 2015), Beijing, China, 25-29, June 2015.
- Description: Abstract: To improve the production ability of a three-phase submerged arc furnace (SAF), it is necessary to maximize the power input; minimize the supply voltage unbalances to reduce the side effect of the power grids. In this paper, maximizing the power input and minimum the supply voltage unbalances based on a proposed multi-swarm multi-objective particle swarm optimization algorithm are focused on. It is necessary to have objective functions when an optimization algorithm is applied. However, it is difficult to get the mathematic model of a three-phase submerged arc furnace according to its mechanisms because the system is complex and there are many disturbances. The neural networks (NN) have been applied since its ability can be used as an arbitrary function approximation mechanism based on the observed data. Based on the Pareto front, a multi-swarm multi-objective particle swarm optimization is pro-posed, which can be used to optimize the NN model of the three-phase SAF. The optimization results showed the efficiency of the proposed method.
- Full Text:
A novel algorithm for optimizing the Pareto set in dynamic problem spaces
- Essiet, Ima O., Sun, Yanxia, Wang, Zenghui
- Authors: Essiet, Ima O. , Sun, Yanxia , Wang, Zenghui
- Date: 2018
- Subjects: Component , Formatting , Style
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/274916 , uj:29352 , Citation: Essiet, I.O., Sun, Y. & Wang, Z. 2018. A novel algorithm for optimizing the Pareto set in dynamic problem spaces.
- Description: Abstract: This paper presents an algorithm based on dynamic multiobjective optimization (DMO) which employs a single randomly mutating time-variant archive to balance convergence and diversity in order to efficiently select the final, non-dominated Pareto set. The algorithm is tested on selected dynamic optimization benchmark functions and improvement in the performance of the single archive approach is demonstrated by improved performance metrics and overall computational time.
- Full Text:
- Authors: Essiet, Ima O. , Sun, Yanxia , Wang, Zenghui
- Date: 2018
- Subjects: Component , Formatting , Style
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/274916 , uj:29352 , Citation: Essiet, I.O., Sun, Y. & Wang, Z. 2018. A novel algorithm for optimizing the Pareto set in dynamic problem spaces.
- Description: Abstract: This paper presents an algorithm based on dynamic multiobjective optimization (DMO) which employs a single randomly mutating time-variant archive to balance convergence and diversity in order to efficiently select the final, non-dominated Pareto set. The algorithm is tested on selected dynamic optimization benchmark functions and improvement in the performance of the single archive approach is demonstrated by improved performance metrics and overall computational time.
- Full Text:
Adaptive optimal digital PID controller
- Authors: Sun, Yanxia , Wang, Zenghui
- Date: 2015
- Subjects: PID controller , Adaptive control , Parameter tuning , Particle swarm optimization
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/21202 , uj:16123 , Citation: Sun, Y. & Wang, Z. 2015. Adaptive optimal digital PID controller. Applied mechanics and materials 789-790:1021-1026.
- Description: Abstract: It is necessary to change the parameters of PID controller if the parameters of plants change or there are disturbances. Particle swarm optimization algorithm is a powerful optimization algorithm to find the global optimal values in the problem space. In this paper, the particle swarm optimization algorithm is used to identify the model of the plant and the parameter of digital PID controller online. The model of the plant is identified online according to the absolute error of the real system output and the identified model output. The digital PID parameters are tuned based on the identified model and they are adaptive if the model is changed. Simulations are done to validate the proposed method comparing with the classical PID controller. , Originally presented at 2014 International Conference on Mechatronics, Automation and Manufacturing (ICMAM 2014), Beijing, October 24-26, 2014.
- Full Text:
- Authors: Sun, Yanxia , Wang, Zenghui
- Date: 2015
- Subjects: PID controller , Adaptive control , Parameter tuning , Particle swarm optimization
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/21202 , uj:16123 , Citation: Sun, Y. & Wang, Z. 2015. Adaptive optimal digital PID controller. Applied mechanics and materials 789-790:1021-1026.
- Description: Abstract: It is necessary to change the parameters of PID controller if the parameters of plants change or there are disturbances. Particle swarm optimization algorithm is a powerful optimization algorithm to find the global optimal values in the problem space. In this paper, the particle swarm optimization algorithm is used to identify the model of the plant and the parameter of digital PID controller online. The model of the plant is identified online according to the absolute error of the real system output and the identified model output. The digital PID parameters are tuned based on the identified model and they are adaptive if the model is changed. Simulations are done to validate the proposed method comparing with the classical PID controller. , Originally presented at 2014 International Conference on Mechatronics, Automation and Manufacturing (ICMAM 2014), Beijing, October 24-26, 2014.
- Full Text:
Adaptive sharing scheme based sub-swarm multi-objective PSO
- Authors: Sun, Yanxia , Wang, Zenghui
- Date: 2017
- Subjects: Multi-objective PSO , Adaptive sharing scheme , Sub-swarm
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/247883 , uj:25750 , Citation: Sun, Y. & Wang, Z. 2017. Adaptive sharing scheme based sub-swarm multi-objective PSO.
- Description: Abstract: To improve the optimization performance of multi-objective particle swarm optimization, a new sub-swarm method, where the particles are divided into several sub-swarms, is proposed. To enhance the quality of the Pareto front set, a new adaptive sharing scheme, which depends on the distances from nearest neighbouring individuals, is proposed and applied. In this method, the first sub-swarms particles dynamically search their corresponding areas which are around some points of the Pareto front set in the objective space, and the chosen points of the Pareto front set are determined based on the adaptive sharing scheme. The second sub-swarm particles search the rest objective space, and they are away from the Pareto front set, which can promote the global search ability of the method. Moreover, the core points of the first sub-swarms are dynamically determined by this new adaptive sharing scheme. Some Simulations are used to test the proposed method, and the results show that the proposed method can achieve better optimization performance comparing with some existing methods.
- Full Text:
- Authors: Sun, Yanxia , Wang, Zenghui
- Date: 2017
- Subjects: Multi-objective PSO , Adaptive sharing scheme , Sub-swarm
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/247883 , uj:25750 , Citation: Sun, Y. & Wang, Z. 2017. Adaptive sharing scheme based sub-swarm multi-objective PSO.
- Description: Abstract: To improve the optimization performance of multi-objective particle swarm optimization, a new sub-swarm method, where the particles are divided into several sub-swarms, is proposed. To enhance the quality of the Pareto front set, a new adaptive sharing scheme, which depends on the distances from nearest neighbouring individuals, is proposed and applied. In this method, the first sub-swarms particles dynamically search their corresponding areas which are around some points of the Pareto front set in the objective space, and the chosen points of the Pareto front set are determined based on the adaptive sharing scheme. The second sub-swarm particles search the rest objective space, and they are away from the Pareto front set, which can promote the global search ability of the method. Moreover, the core points of the first sub-swarms are dynamically determined by this new adaptive sharing scheme. Some Simulations are used to test the proposed method, and the results show that the proposed method can achieve better optimization performance comparing with some existing methods.
- Full Text:
An improved ensemble learning approach for the prediction of heart disease risk
- Mienye, Ibomoiye Domor, Sun, Yanxia, Wang, Zenghui
- Authors: Mienye, Ibomoiye Domor , Sun, Yanxia , Wang, Zenghui
- Date: 2020
- Subjects: Ensemble learning , Machine learning , CART
- Language: English
- Type: Article
- Identifier: http://hdl.handle.net/10210/430433 , uj:37100 , Citation: Mienye, I.D., Sun, Y. & Wang, Z. 2020. An improved ensemble learning approach for the prediction of heart disease risk.
- Description: Abstract: According to the world health organization (WHO), cardiovascular diseases are the leading cause of death worldwide, and identifying those at risk can help prevent sudden deaths. Some computational methods have been proposed to predict the patient’s heart disease risk. Meanwhile, the accurate prediction of diseases is a critical aspect of machine learning and to further enhance the classification of heart disease risk, this paper proposes an improved ensemble learning approach. The mean of the data columns is used to partition the dataset into smaller subsets, and then a classification and regression tree (CART) algorithm is applied to model each partition. Randomization is introduced during the data partitioning. The resulting ensemble produces a robust model for the prediction of heart disease risk. To evaluate the effectiveness of the proposed method, the Cleveland and Framingham heart disease datasets are used. When compared with other machine learning methods and some recent scholarly works, the proposed method showed significant improvement.
- Full Text:
- Authors: Mienye, Ibomoiye Domor , Sun, Yanxia , Wang, Zenghui
- Date: 2020
- Subjects: Ensemble learning , Machine learning , CART
- Language: English
- Type: Article
- Identifier: http://hdl.handle.net/10210/430433 , uj:37100 , Citation: Mienye, I.D., Sun, Y. & Wang, Z. 2020. An improved ensemble learning approach for the prediction of heart disease risk.
- Description: Abstract: According to the world health organization (WHO), cardiovascular diseases are the leading cause of death worldwide, and identifying those at risk can help prevent sudden deaths. Some computational methods have been proposed to predict the patient’s heart disease risk. Meanwhile, the accurate prediction of diseases is a critical aspect of machine learning and to further enhance the classification of heart disease risk, this paper proposes an improved ensemble learning approach. The mean of the data columns is used to partition the dataset into smaller subsets, and then a classification and regression tree (CART) algorithm is applied to model each partition. Randomization is introduced during the data partitioning. The resulting ensemble produces a robust model for the prediction of heart disease risk. To evaluate the effectiveness of the proposed method, the Cleveland and Framingham heart disease datasets are used. When compared with other machine learning methods and some recent scholarly works, the proposed method showed significant improvement.
- Full Text:
Analysis of a fractional order nonlinear system based on the frequency domain approximation
- Authors: Wang, Zenghui , Sun, Yanxia
- Date: 2014
- Subjects: Nonlinear system , Fractional order system , Chaos , Hyperchaos
- Language: English
- Type: Article
- Identifier: http://hdl.handle.net/10210/17800 , uj:15926 , Doi: 10.14355/ijepr.2014.0304.02 , Citation: Wang, Z., Sun, Y. 2014. Analysis of a fractional order nonlinear system based on the frequency domain approximation. International Journal of Engineering Practical Research (IJEPR), 3(4): 74-77.
- Description: Abstract: The dynamics of nonlinear system is very complicated especially the fractional nonlinear system since they can be found in many areas of engineering and science. The dynamics of the Lorenz system with fractional derivatives is analysed based on the frequency approximation. For a given range of parameters where the non‐fractional Lorenz system has periodic orbits, it is found that the fractional Lorenz system exhibits chaos and hyperchaos. A striking finding is that the fractional Lorenz system exhibits hyperchaos, although the total system order is less than 3, which is contrary to the well known conclusion that hyperchaos cannot occur in the integer‐order continuous‐time autonomous system of order less than 4. Finally, a reasonable explanation is offered for this complicated dynamical phenomenon.
- Full Text:
- Authors: Wang, Zenghui , Sun, Yanxia
- Date: 2014
- Subjects: Nonlinear system , Fractional order system , Chaos , Hyperchaos
- Language: English
- Type: Article
- Identifier: http://hdl.handle.net/10210/17800 , uj:15926 , Doi: 10.14355/ijepr.2014.0304.02 , Citation: Wang, Z., Sun, Y. 2014. Analysis of a fractional order nonlinear system based on the frequency domain approximation. International Journal of Engineering Practical Research (IJEPR), 3(4): 74-77.
- Description: Abstract: The dynamics of nonlinear system is very complicated especially the fractional nonlinear system since they can be found in many areas of engineering and science. The dynamics of the Lorenz system with fractional derivatives is analysed based on the frequency approximation. For a given range of parameters where the non‐fractional Lorenz system has periodic orbits, it is found that the fractional Lorenz system exhibits chaos and hyperchaos. A striking finding is that the fractional Lorenz system exhibits hyperchaos, although the total system order is less than 3, which is contrary to the well known conclusion that hyperchaos cannot occur in the integer‐order continuous‐time autonomous system of order less than 4. Finally, a reasonable explanation is offered for this complicated dynamical phenomenon.
- Full Text:
Analysis of the effect of parameter variation on a dynamic cost function for distributed energy resources : a DER-CAM case study
- Essiet, Ima O., Sun, Yanxia, Wang, Zenghui
- Authors: Essiet, Ima O. , Sun, Yanxia , Wang, Zenghui
- Date: 2018
- Subjects: Evolutionary algorithm , Pareto front , Distributed energy resource
- Language: English
- Type: Article
- Identifier: http://hdl.handle.net/10210/274901 , uj:29350 , Citation: Essiet, I.O., Sun, Y. & Wang, Z. 2018. Analysis of the effect of parameter variation on a dynamic cost function for distributed energy resources : a DER-CAM case study.
- Description: Abstract: This paper investigates the effect of selected strategies of distributed energy resources (DER) on an energy cost function, which optimizes the allocation of distributed energy resources for a mid-rise apartment building. This is achieved by comparison of parameter optimization results for both a high- and low-level optimizer respectively. The optimization process is carried out using the following approach: (1) a two-objective function is constructed with one objective function similar to that of the high-level optimizer (DER-CAM); (2) an evolutionary algorithm (EA) with modified selection capability is used to optimize the two-objective function problem in (1) for 4 selected cases of DER utilization previously optimized in DER-CAM. (3) the optimization results of the low-level optimizer are compared with the outcome of DER-CAM optimization for the 4 selected cases. This is done to establish the capability of DER-CAM as an effective tool for optimal distributed energy resource allocation. Results obtained demonstrate the effect of load shifting and solar photovoltaic (PV) panels with power exporting capability on the optimization of the cost function. The Pareto-based MOEA approach has also proved to be effective in observing the interactions between objective function parameters. Mean inverted generational distance (MIGD) values obtained over 10 runs for each of the 4 cases considered show that a DER combination of PV panel, battery storage, heat pump and load shifting outperforms the other strategies in 70% of the total simulation runs.
- Full Text:
- Authors: Essiet, Ima O. , Sun, Yanxia , Wang, Zenghui
- Date: 2018
- Subjects: Evolutionary algorithm , Pareto front , Distributed energy resource
- Language: English
- Type: Article
- Identifier: http://hdl.handle.net/10210/274901 , uj:29350 , Citation: Essiet, I.O., Sun, Y. & Wang, Z. 2018. Analysis of the effect of parameter variation on a dynamic cost function for distributed energy resources : a DER-CAM case study.
- Description: Abstract: This paper investigates the effect of selected strategies of distributed energy resources (DER) on an energy cost function, which optimizes the allocation of distributed energy resources for a mid-rise apartment building. This is achieved by comparison of parameter optimization results for both a high- and low-level optimizer respectively. The optimization process is carried out using the following approach: (1) a two-objective function is constructed with one objective function similar to that of the high-level optimizer (DER-CAM); (2) an evolutionary algorithm (EA) with modified selection capability is used to optimize the two-objective function problem in (1) for 4 selected cases of DER utilization previously optimized in DER-CAM. (3) the optimization results of the low-level optimizer are compared with the outcome of DER-CAM optimization for the 4 selected cases. This is done to establish the capability of DER-CAM as an effective tool for optimal distributed energy resource allocation. Results obtained demonstrate the effect of load shifting and solar photovoltaic (PV) panels with power exporting capability on the optimization of the cost function. The Pareto-based MOEA approach has also proved to be effective in observing the interactions between objective function parameters. Mean inverted generational distance (MIGD) values obtained over 10 runs for each of the 4 cases considered show that a DER combination of PV panel, battery storage, heat pump and load shifting outperforms the other strategies in 70% of the total simulation runs.
- Full Text:
Asynchronous and stochastic dimension updating PSO and its application to parameter estimation for frequency modulated (FM) sound waves
- Authors: Sun, Yanxia , Wang, Zenghui
- Date: 2016
- Subjects: Swarm optimization , Asynchronous updating , Stochastic dimension updating
- Language: English
- Type: Conference proceeding
- Identifier: http://hdl.handle.net/10210/92362 , uj:20221 , Citation: Sun, Y. & Wang, Z. 2016. Asynchronous and stochastic dimension updating PSO and its application to parameter estimation for frequency modulated (FM) sound waves.
- Description: Abstract: The particle velocity and position updating plays very important role for achieving good optimization performance for Particle Swarm Optimization (PSO). This paper analyzed the performance of asynchronously updating PSO and synchronously updating PSO by simulation and found that the asynchronously updating way can achieve better optimization performance than the synchronously updating way. Moreover, the convergence of asynchronously PSO is faster than the synchronously PSO, which means there is spare time to achieve better optimization performance based on some techniques. Here we proposed stochastic dimension updating technique which means only some dimensions of position will be updated. Several benchmark functions have been used to validate the proposed method and the proposed method is also applied to the parameter estimation for frequency modulated Sound Waves.
- Full Text:
- Authors: Sun, Yanxia , Wang, Zenghui
- Date: 2016
- Subjects: Swarm optimization , Asynchronous updating , Stochastic dimension updating
- Language: English
- Type: Conference proceeding
- Identifier: http://hdl.handle.net/10210/92362 , uj:20221 , Citation: Sun, Y. & Wang, Z. 2016. Asynchronous and stochastic dimension updating PSO and its application to parameter estimation for frequency modulated (FM) sound waves.
- Description: Abstract: The particle velocity and position updating plays very important role for achieving good optimization performance for Particle Swarm Optimization (PSO). This paper analyzed the performance of asynchronously updating PSO and synchronously updating PSO by simulation and found that the asynchronously updating way can achieve better optimization performance than the synchronously updating way. Moreover, the convergence of asynchronously PSO is faster than the synchronously PSO, which means there is spare time to achieve better optimization performance based on some techniques. Here we proposed stochastic dimension updating technique which means only some dimensions of position will be updated. Several benchmark functions have been used to validate the proposed method and the proposed method is also applied to the parameter estimation for frequency modulated Sound Waves.
- Full Text:
Big data analysis for gas sensor using convolutional neural network and ensemble of evolutionary algorithms
- Essieta, Ima, Sun, Yanxia, Wang, Zenghui
- Authors: Essieta, Ima , Sun, Yanxia , Wang, Zenghui
- Date: 2019
- Subjects: Convolutional neural network (CNN), Deep learning, Classification accuracy
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/397337 , uj:33021
- Description: Abstract : Big data analysis has gained popularity over the years as a result of developments in computing and electronics. Several methods have been proposed in literature for efficiently mining data from dedicated databases and a wide range of electronic sensors. However, as the volume of data grows, diversity and velocity of the data also grows (sometimes exponentially). Neural networks have been proposed in literature for optimal big data mining; however, they suffer from problems of over-fitting and under-fitting. In this paper, an ensemble of evolutionary algorithms is proposed, namely: improved non-dominated sorting genetic algorithm (NSGA), differential evolution (DE) and multi-objective evolutionary algorithm based on dominance and decomposition (MOEAD/D). These algorithms are each combined with a convolutional neural network (CNN), and performance is evaluated using root mean square error (RMSE) and mean absolute percentage error (MAPE). The test data consists of gas sensor readings obtained from an array of 16 metal oxide semiconductor sensors. The gases being detected are Carbon Monoxide/Ethylene in air, and Methane/Ethylene in air. 4,178,504 data points were collected over an uninterrupted 12-hour period. Preliminary results show improved RMSE and MAPE values over 50 learning cycles compared to a case where the CNN learned on its own.
- Full Text:
- Authors: Essieta, Ima , Sun, Yanxia , Wang, Zenghui
- Date: 2019
- Subjects: Convolutional neural network (CNN), Deep learning, Classification accuracy
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/397337 , uj:33021
- Description: Abstract : Big data analysis has gained popularity over the years as a result of developments in computing and electronics. Several methods have been proposed in literature for efficiently mining data from dedicated databases and a wide range of electronic sensors. However, as the volume of data grows, diversity and velocity of the data also grows (sometimes exponentially). Neural networks have been proposed in literature for optimal big data mining; however, they suffer from problems of over-fitting and under-fitting. In this paper, an ensemble of evolutionary algorithms is proposed, namely: improved non-dominated sorting genetic algorithm (NSGA), differential evolution (DE) and multi-objective evolutionary algorithm based on dominance and decomposition (MOEAD/D). These algorithms are each combined with a convolutional neural network (CNN), and performance is evaluated using root mean square error (RMSE) and mean absolute percentage error (MAPE). The test data consists of gas sensor readings obtained from an array of 16 metal oxide semiconductor sensors. The gases being detected are Carbon Monoxide/Ethylene in air, and Methane/Ethylene in air. 4,178,504 data points were collected over an uninterrupted 12-hour period. Preliminary results show improved RMSE and MAPE values over 50 learning cycles compared to a case where the CNN learned on its own.
- Full Text:
Cask theory based parameter optimization for particle swarm optimization
- Authors: Wang, Zenghui , Sun, Yanxia
- Date: 2013
- Subjects: PSO , Parameter optimization , Try and error method , Nested optimization method , Cask theory
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/21292 , uj:16138 , Citation: Wang, Z. & Sun, Y. 2013. Cask theory based parameter optimization for particle swarm optimization. Lecture Notes in Computer Sciences (LNCS) 7928: 137–143. ISSN: 0302-9743
- Description: Abstract: To avoid the bored try and error method of finding a set of parameters of Particle Swarm Optimization (PSO) and achieve good optimization performance, it is desired to get an adaptive optimization method to search a good set of parameters. A nested optimization method is proposed in this paper and it can be used to search the tuned parameters such as inertia weight, acceleration coefficients c1 and c2, and so on. This method considers the cask theory to achieve a better optimization performance. Several famous benchmarks were used to validate the proposed method and the simulation results showed the efficiency of the proposed method. , Originally presented at Fourth International Conference on Swarm Intelligence (ICSI 2013), Harbin, China, 12-15, June, 2013.
- Full Text:
- Authors: Wang, Zenghui , Sun, Yanxia
- Date: 2013
- Subjects: PSO , Parameter optimization , Try and error method , Nested optimization method , Cask theory
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/21292 , uj:16138 , Citation: Wang, Z. & Sun, Y. 2013. Cask theory based parameter optimization for particle swarm optimization. Lecture Notes in Computer Sciences (LNCS) 7928: 137–143. ISSN: 0302-9743
- Description: Abstract: To avoid the bored try and error method of finding a set of parameters of Particle Swarm Optimization (PSO) and achieve good optimization performance, it is desired to get an adaptive optimization method to search a good set of parameters. A nested optimization method is proposed in this paper and it can be used to search the tuned parameters such as inertia weight, acceleration coefficients c1 and c2, and so on. This method considers the cask theory to achieve a better optimization performance. Several famous benchmarks were used to validate the proposed method and the simulation results showed the efficiency of the proposed method. , Originally presented at Fourth International Conference on Swarm Intelligence (ICSI 2013), Harbin, China, 12-15, June, 2013.
- Full Text:
Chaotic particle swarm optimization
- Sun, Yanxia, Qi, Guoyuan, Wang, Zenghui, Van Wyk, Barend Jacobus, Hamam, Yskandar
- Authors: Sun, Yanxia , Qi, Guoyuan , Wang, Zenghui , Van Wyk, Barend Jacobus , Hamam, Yskandar
- Date: 2009
- Subjects: Chaos , Particle swarm optimization , Neural network , Convergence
- Language: English
- Identifier: http://hdl.handle.net/10210/22298 , uj:16185 , Citation: Sun, Y. et al. 2009. Chaotic particle swarm optimization. 2009 World Summit on Genetic and Evolutionary Computation (2009 GEC Summit), Shanghai, China, June 12-14, 2009. p. 505-510. ISBN:16-05-58326-X.
- Description: Abstract: A new particle swarm optimization (PSO) algorithm with has a chaotic neural network structure, is proposed. The structure is similar to the Hop¯eld neural network with transient chaos, and has an improved ability to search for globally optimal solution and does not su®er from problems of premature convergence. The presented PSO model is discrete-time discrete-state. The bifurcation diagram of a particle shows that it converges to a stable fixed point from a strange attractor, guaranteeing system convergence.
- Full Text:
- Authors: Sun, Yanxia , Qi, Guoyuan , Wang, Zenghui , Van Wyk, Barend Jacobus , Hamam, Yskandar
- Date: 2009
- Subjects: Chaos , Particle swarm optimization , Neural network , Convergence
- Language: English
- Identifier: http://hdl.handle.net/10210/22298 , uj:16185 , Citation: Sun, Y. et al. 2009. Chaotic particle swarm optimization. 2009 World Summit on Genetic and Evolutionary Computation (2009 GEC Summit), Shanghai, China, June 12-14, 2009. p. 505-510. ISBN:16-05-58326-X.
- Description: Abstract: A new particle swarm optimization (PSO) algorithm with has a chaotic neural network structure, is proposed. The structure is similar to the Hop¯eld neural network with transient chaos, and has an improved ability to search for globally optimal solution and does not su®er from problems of premature convergence. The presented PSO model is discrete-time discrete-state. The bifurcation diagram of a particle shows that it converges to a stable fixed point from a strange attractor, guaranteeing system convergence.
- Full Text:
E-education in an open distance university
- Authors: Wang, Zenghui , Sun, Yanxia
- Date: 2014
- Subjects: E-education , Open distance university , University of South Africa , e-system
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/21137 , uj:16113 , Citation: Wang, Z. & Sun, Y. 2014. E-education in an open distance university. Proceedings of International Conference on Advanced Education and Management [ICAEM2014] Beijing, China, Jan.04~Jan.05, 2014. p. 141-145. ISBN: 978-1-60595-153-9
- Description: Abstract: With the development of internet, software and the relevant techniques, the e-education becomes more and more attractive. As one of most famous open distance universities, the University of South Africa (UNISA) has partly realized e-education. This paper describes and investigates the existing e-education system of UNISA. There are many advantages, such as realizing paperless office, lower cost, high efficiency, and so on, using this e-education system for teaching and learning. Moreover, the challenges and solving methods are also studied and discussed.
- Full Text:
- Authors: Wang, Zenghui , Sun, Yanxia
- Date: 2014
- Subjects: E-education , Open distance university , University of South Africa , e-system
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/21137 , uj:16113 , Citation: Wang, Z. & Sun, Y. 2014. E-education in an open distance university. Proceedings of International Conference on Advanced Education and Management [ICAEM2014] Beijing, China, Jan.04~Jan.05, 2014. p. 141-145. ISBN: 978-1-60595-153-9
- Description: Abstract: With the development of internet, software and the relevant techniques, the e-education becomes more and more attractive. As one of most famous open distance universities, the University of South Africa (UNISA) has partly realized e-education. This paper describes and investigates the existing e-education system of UNISA. There are many advantages, such as realizing paperless office, lower cost, high efficiency, and so on, using this e-education system for teaching and learning. Moreover, the challenges and solving methods are also studied and discussed.
- Full Text:
Elitism set based particle swarm optimization and its application
- Authors: Sun, Yanxia , Wang, Zenghui
- Date: 2017
- Subjects: Particle swarm optimization , Statistical method , Topology structure
- Language: English
- Type: Article
- Identifier: http://hdl.handle.net/10210/242178 , uj:24972 , Citation: Sun, Y. & Wang, Z. 2017. Elitism set based particle swarm optimization and its application.
- Description: Abstract: Topology plays an important role for Particle Swarm Optimization (PSO) to achieve good optimization performance. It is difficult to find one topology structure for the particles to achieve better optimization performance than the others since the optimization performance not only depends on the searching abilities of the particles, also depends on the type of the optimization problems. Three elitist set based PSO algorithm without using explicit topology structure is proposed in this paper. An elitist set, which is based on the individual best experience, is used to communicate among the particles. Moreover, to avoid the premature of the particles, different statistical methods have been used in these three proposed methods. The performance of the proposed PSOs is compared with the results of the standard PSO 2011 and several PSO with different topologies, and the simulation results and comparisons demonstrate that the proposed PSO with adaptive probabilistic preference can achieve good optimization performance.
- Full Text:
- Authors: Sun, Yanxia , Wang, Zenghui
- Date: 2017
- Subjects: Particle swarm optimization , Statistical method , Topology structure
- Language: English
- Type: Article
- Identifier: http://hdl.handle.net/10210/242178 , uj:24972 , Citation: Sun, Y. & Wang, Z. 2017. Elitism set based particle swarm optimization and its application.
- Description: Abstract: Topology plays an important role for Particle Swarm Optimization (PSO) to achieve good optimization performance. It is difficult to find one topology structure for the particles to achieve better optimization performance than the others since the optimization performance not only depends on the searching abilities of the particles, also depends on the type of the optimization problems. Three elitist set based PSO algorithm without using explicit topology structure is proposed in this paper. An elitist set, which is based on the individual best experience, is used to communicate among the particles. Moreover, to avoid the premature of the particles, different statistical methods have been used in these three proposed methods. The performance of the proposed PSOs is compared with the results of the standard PSO 2011 and several PSO with different topologies, and the simulation results and comparisons demonstrate that the proposed PSO with adaptive probabilistic preference can achieve good optimization performance.
- Full Text:
Enhanced NSGA based on adaptive crossover rate and reference points
- Essiet, Ima O., Sun, Yanxia, Wang, Zenghui
- Authors: Essiet, Ima O. , Sun, Yanxia , Wang, Zenghui
- Date: 2017
- Subjects: Optimization , Inverted generational distance , Reference points
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/256932 , uj:26981 , Citation: Essiet, I.O., Sun, Y. & Wang, Z. 2017. Enhanced NSGA based on adaptive crossover rate and reference points.
- Description: Abstract: The challenges of many-objective optimization are investigated; and one new algorithm, which is based on the NSGA-II, is proposed for multi-objective optimization in this paper. The reference points and an adaptable crossover rate are combined in the algorithm to improve the performance of NSGA-II. The performance of NSGA for optimizing the many objective search space is examined with and without the proposed algorithm through a constrained two-objective problem with up to 40 dimensions. Simulation results show that the proposed algorithm improves the performance of NSGA for the selected test problem in generations where a non-dominated set is not obtained by 39%.
- Full Text:
- Authors: Essiet, Ima O. , Sun, Yanxia , Wang, Zenghui
- Date: 2017
- Subjects: Optimization , Inverted generational distance , Reference points
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/256932 , uj:26981 , Citation: Essiet, I.O., Sun, Y. & Wang, Z. 2017. Enhanced NSGA based on adaptive crossover rate and reference points.
- Description: Abstract: The challenges of many-objective optimization are investigated; and one new algorithm, which is based on the NSGA-II, is proposed for multi-objective optimization in this paper. The reference points and an adaptable crossover rate are combined in the algorithm to improve the performance of NSGA-II. The performance of NSGA for optimizing the many objective search space is examined with and without the proposed algorithm through a constrained two-objective problem with up to 40 dimensions. Simulation results show that the proposed algorithm improves the performance of NSGA for the selected test problem in generations where a non-dominated set is not obtained by 39%.
- Full Text:
Fully connected multi-objective particle swarm optimizer based on neural network
- Authors: Wang, Zenghui , Sun, Yanxia
- Date: 2011
- Subjects: Multi-objective optimization , Particle swarm optimization , Neural network , Pareto front , Non domination
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/22087 , uj:16157 , Citation: Wang, Z. & Sun, Y. 2011. Fully connected multi-objective particle swarm optimizer based on neural network. Lecture notes in computer science 6838:170-177.
- Description: Abstract: In this paper, a new model for multi-objective particle swarm optimization (MOPSO) is proposed. In this model, each particle’s behavior is influenced by the best experience among its neighbors, its own best experience and all its components. The influence among different components of particles is implemented by the on-line training of a multi-input Multi-output back propagation (BP) neural network. The inputs and outputs of the BP neural network are the particle position and its the ’gradient descent’ direction vector to the less objective value according to the definition of no-domination, respectively. Therefore, the new structured MOPSO model is called a fully connected multi-objective particle swarm optimizer (FCMOPSO). Simulation results and comparisons with exiting MOPSOs demonstrate that the proposed FCMOPSO is more stable and can improve the optimization performance. , Originally presented at Fourth International Conference on Information and Computing (ICIC 2011), Phuket Island, Thailand 25 – 27 April 2011.
- Full Text:
- Authors: Wang, Zenghui , Sun, Yanxia
- Date: 2011
- Subjects: Multi-objective optimization , Particle swarm optimization , Neural network , Pareto front , Non domination
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/22087 , uj:16157 , Citation: Wang, Z. & Sun, Y. 2011. Fully connected multi-objective particle swarm optimizer based on neural network. Lecture notes in computer science 6838:170-177.
- Description: Abstract: In this paper, a new model for multi-objective particle swarm optimization (MOPSO) is proposed. In this model, each particle’s behavior is influenced by the best experience among its neighbors, its own best experience and all its components. The influence among different components of particles is implemented by the on-line training of a multi-input Multi-output back propagation (BP) neural network. The inputs and outputs of the BP neural network are the particle position and its the ’gradient descent’ direction vector to the less objective value according to the definition of no-domination, respectively. Therefore, the new structured MOPSO model is called a fully connected multi-objective particle swarm optimizer (FCMOPSO). Simulation results and comparisons with exiting MOPSOs demonstrate that the proposed FCMOPSO is more stable and can improve the optimization performance. , Originally presented at Fourth International Conference on Information and Computing (ICIC 2011), Phuket Island, Thailand 25 – 27 April 2011.
- Full Text: