Load forecasting is of great significance to the development of new power systems.With the advancement of smart grids,the integration and distribution of distributed renewable energy sources and power electronics devi...Load forecasting is of great significance to the development of new power systems.With the advancement of smart grids,the integration and distribution of distributed renewable energy sources and power electronics devices have made power load data increasingly complex and volatile.This places higher demands on the prediction and analysis of power loads.In order to improve the prediction accuracy of short-term power load,a CNN-BiLSTMTPA short-term power prediction model based on the Improved Whale Optimization Algorithm(IWOA)with mixed strategies was proposed.Firstly,the model combined the Convolutional Neural Network(CNN)with the Bidirectional Long Short-Term Memory Network(BiLSTM)to fully extract the spatio-temporal characteristics of the load data itself.Then,the Temporal Pattern Attention(TPA)mechanism was introduced into the CNN-BiLSTM model to automatically assign corresponding weights to the hidden states of the BiLSTM.This allowed the model to differentiate the importance of load sequences at different time intervals.At the same time,in order to solve the problem of the difficulties of selecting the parameters of the temporal model,and the poor global search ability of the whale algorithm,which is easy to fall into the local optimization,the whale algorithm(IWOA)was optimized by using the hybrid strategy of Tent chaos mapping and Levy flight strategy,so as to better search the parameters of the model.In this experiment,the real load data of a region in Zhejiang was taken as an example to analyze,and the prediction accuracy(R2)of the proposed method reached 98.83%.Compared with the prediction models such as BP,WOA-CNN-BiLSTM,SSA-CNN-BiLSTM,CNN-BiGRU-Attention,etc.,the experimental results showed that the model proposed in this study has a higher prediction accuracy.展开更多
Motivated by the business model called“community group buying”(CGB),which has emerged in China and some countries in Southeast Asia,such as Singapore and Indonesia,we develop algorithms that could help CGB platforms...Motivated by the business model called“community group buying”(CGB),which has emerged in China and some countries in Southeast Asia,such as Singapore and Indonesia,we develop algorithms that could help CGB platforms match consumers with stage-stations(the picking up center under the CGB mode).By altering the fundamental design of the existing hierarchy algorithms,improvements are achieved.It is proven that our method has a faster running speed and greater space efficiency.Our algorithms avoid traversal and compress the time complexities of matching a consumer with a stage-station and updating the storage information to O(logM)and O(MlogG),where M is the number of stage-stations and G is that of the platform’s stock-keeping units.Simulation comparisons of our algorithms with the current methods of CGB platforms show that our approaches can effectively reduce delivery costs.An interesting observation of the simula-tions is worthy of note:Increasing G may incur higher costs since it makes inventories more dispersed and delivery prob-lems more complicated.展开更多
To improve the global search ability and imaging quality of electrical resistivity imaging(ERI) inversion, a two-stage learning ICPSO algorithm of radial basis function neural network(RBFNN) based on information crite...To improve the global search ability and imaging quality of electrical resistivity imaging(ERI) inversion, a two-stage learning ICPSO algorithm of radial basis function neural network(RBFNN) based on information criterion(IC) and particle swarm optimization(PSO) is presented. In the proposed method, IC is applied to obtain the hidden layer structure by calculating the optimal IC value automatically and PSO algorithm is used to optimize the centers and widths of the radial basis functions in the hidden layer. Meanwhile, impacts of different information criteria to the inversion results are compared, and an implementation of the proposed ICPSO algorithm is given. The optimized neural network has one hidden layer with 261 nodes selected by AKAIKE's information criterion(AIC) and it is trained on 32 data sets and tested on another 8 synthetic data sets. Two complex synthetic examples are used to verify the feasibility and effectiveness of the proposed method with two learning stages. The results show that the proposed method has better performance and higher imaging quality than three-layer and four-layer back propagation neural networks(BPNNs) and traditional least square(LS) inversion.展开更多
Pantograph-catenary contact force provides the main basis for evaluation of current quality collection; however,the pantograph-catenary contact force is largely affected by the catenary irregularities.To analyze the c...Pantograph-catenary contact force provides the main basis for evaluation of current quality collection; however,the pantograph-catenary contact force is largely affected by the catenary irregularities.To analyze the correlated relationship between catenary irregularities and pantograph-catenary contact force,a method based on nonlinear auto-regressive with exogenous input(NARX) neural networks was developed.First,to collect the test data of catenary irregularities and contact force,the pantograph/catenary dynamics model was established and dynamic simulation was conducted using MATLAB/Simulink.Second,catenary irregularities were used as the input to NARX neural network and the contact force was determined as output of the NARX neural network,in which the neural network was trained by an improved training mechanism based on the regularization algorithm.The simulation results show that the testing error and correlation coefficient are 0.1100 and 0.8029,respectively,and the prediction accuracy is satisfactory.And the comparisons with other algorithms indicate the validity and superiority of the proposed approach.展开更多
To improve the computational efficiency of the reliability-based design optimization(RBDO) of flexible mechanism, particle swarm optimization-advanced extremum response surface method(PSO-AERSM) was proposed by integr...To improve the computational efficiency of the reliability-based design optimization(RBDO) of flexible mechanism, particle swarm optimization-advanced extremum response surface method(PSO-AERSM) was proposed by integrating particle swarm optimization(PSO) algorithm and advanced extremum response surface method(AERSM). Firstly, the AERSM was developed and its mathematical model was established based on artificial neural network, and the PSO algorithm was investigated. And then the RBDO model of flexible mechanism was presented based on AERSM and PSO. Finally, regarding cross-sectional area as design variable, the reliability optimization of flexible mechanism was implemented subject to reliability degree and uncertainties based on the proposed approach. The optimization results show that the cross-section sizes obviously reduce by 22.96 mm^2 while keeping reliability degree. Through the comparison of methods, it is demonstrated that the AERSM holds high computational efficiency while keeping computational precision for the RBDO of flexible mechanism, and PSO algorithm minimizes the response of the objective function. The efforts of this work provide a useful sight for the reliability optimization of flexible mechanism, and enrich and develop the reliability theory as well.展开更多
An isothermal compressive experiment using Gleeble 1500 thermal simulator was studied to acquire flow stress at different deformation temperatures, strains and strain rates. The artificial neural networks with the err...An isothermal compressive experiment using Gleeble 1500 thermal simulator was studied to acquire flow stress at different deformation temperatures, strains and strain rates. The artificial neural networks with the error back propagation(BP) algorithm was used to establish constitutive model of 2519 aluminum alloy based on the experiment data. The model results show that the systematical error is small(δ=3.3%) when the value of objective function is 0.2, the number of nodes in the hidden layer is 5 and the learning rate is 0.1. Flow stresses of the material under various thermodynamic conditions are predicted by the neural network model, and the predicted results correspond with the experimental results. A knowledge-based constitutive relation model is developed.展开更多
The static performance of inflatable structures has been well studied and the dynamic deployment simulation has received much attention. However, very few studies focus on its deflation behavior. Although there are se...The static performance of inflatable structures has been well studied and the dynamic deployment simulation has received much attention. However, very few studies focus on its deflation behavior. Although there are several dynamic finite element algorithms that can be applied to the deflation simulation, their computation costs are expensive, especially for large scale structures. In this work, a simple method based on classic thermodynamics and the analytical relationship between air and membrane was proposed to efficiently analyze the air state variables under the condition of ventilation. Combined with failure analysis of static bearing capacity, a fast incremental analytical method was presented to predict both elastic and post wrinkling deflation process of inflatable structures. Comparisons between simplified analysis, dynamic finite element simulation, and a full-scale experimental test are presented and the suitability of this simple method for solving the air state and predicting the deflation behavior of inflatable structures is proved.展开更多
Considering multi-factor influence, a forecasting model was built. The structure of BP neural network was designed, and immune algorithm was applied to optimize its network structure and weight. After training the dat...Considering multi-factor influence, a forecasting model was built. The structure of BP neural network was designed, and immune algorithm was applied to optimize its network structure and weight. After training the data of power demand from the year 1980 to 2005 in China, a nonlinear network model was obtained on the relationship between power demand and the factors which had impacts on it, and thus the above proposed method was verified. Meanwhile, the results were compared to those of neural network optimized by genetic algorithm. The results show that this method is superior to neural network optimized by genetic algorithm and is one of the effective ways of time series forecast.展开更多
An on-demand distributed clustering algorithm based on neural network was proposed. The system parameters and the combined weight for each node were computed, and cluster-heads were chosen using the weighted clusterin...An on-demand distributed clustering algorithm based on neural network was proposed. The system parameters and the combined weight for each node were computed, and cluster-heads were chosen using the weighted clustering algorithm, then a training set was created and a neural network was trained. In this algorithm, several system parameters were taken into account, such as the ideal node-degree, the transmission power, the mobility and the battery power of the nodes. The algorithm can be used directly to test whether a node is a cluster-head or not. Moreover, the clusters recreation can be speeded up.展开更多
文摘Load forecasting is of great significance to the development of new power systems.With the advancement of smart grids,the integration and distribution of distributed renewable energy sources and power electronics devices have made power load data increasingly complex and volatile.This places higher demands on the prediction and analysis of power loads.In order to improve the prediction accuracy of short-term power load,a CNN-BiLSTMTPA short-term power prediction model based on the Improved Whale Optimization Algorithm(IWOA)with mixed strategies was proposed.Firstly,the model combined the Convolutional Neural Network(CNN)with the Bidirectional Long Short-Term Memory Network(BiLSTM)to fully extract the spatio-temporal characteristics of the load data itself.Then,the Temporal Pattern Attention(TPA)mechanism was introduced into the CNN-BiLSTM model to automatically assign corresponding weights to the hidden states of the BiLSTM.This allowed the model to differentiate the importance of load sequences at different time intervals.At the same time,in order to solve the problem of the difficulties of selecting the parameters of the temporal model,and the poor global search ability of the whale algorithm,which is easy to fall into the local optimization,the whale algorithm(IWOA)was optimized by using the hybrid strategy of Tent chaos mapping and Levy flight strategy,so as to better search the parameters of the model.In this experiment,the real load data of a region in Zhejiang was taken as an example to analyze,and the prediction accuracy(R2)of the proposed method reached 98.83%.Compared with the prediction models such as BP,WOA-CNN-BiLSTM,SSA-CNN-BiLSTM,CNN-BiGRU-Attention,etc.,the experimental results showed that the model proposed in this study has a higher prediction accuracy.
基金supported by the National Natural Science Foundation of China(71991464,71921001)Fundamental Research Funds for the Central Universities,General Program(WK2040000053)Key Program(YD2040002027)。
文摘Motivated by the business model called“community group buying”(CGB),which has emerged in China and some countries in Southeast Asia,such as Singapore and Indonesia,we develop algorithms that could help CGB platforms match consumers with stage-stations(the picking up center under the CGB mode).By altering the fundamental design of the existing hierarchy algorithms,improvements are achieved.It is proven that our method has a faster running speed and greater space efficiency.Our algorithms avoid traversal and compress the time complexities of matching a consumer with a stage-station and updating the storage information to O(logM)and O(MlogG),where M is the number of stage-stations and G is that of the platform’s stock-keeping units.Simulation comparisons of our algorithms with the current methods of CGB platforms show that our approaches can effectively reduce delivery costs.An interesting observation of the simula-tions is worthy of note:Increasing G may incur higher costs since it makes inventories more dispersed and delivery prob-lems more complicated.
基金Project(41374118)supported by the National Natural Science Foundation,ChinaProject(20120162110015)supported by Research Fund for the Doctoral Program of Higher Education,China+3 种基金Project(2015M580700)supported by the China Postdoctoral Science Foundation,ChinaProject(2016JJ3086)supported by the Hunan Provincial Natural Science Foundation,ChinaProject(2015JC3067)supported by the Hunan Provincial Science and Technology Program,ChinaProject(15B138)supported by the Scientific Research Fund of Hunan Provincial Education Department,China
文摘To improve the global search ability and imaging quality of electrical resistivity imaging(ERI) inversion, a two-stage learning ICPSO algorithm of radial basis function neural network(RBFNN) based on information criterion(IC) and particle swarm optimization(PSO) is presented. In the proposed method, IC is applied to obtain the hidden layer structure by calculating the optimal IC value automatically and PSO algorithm is used to optimize the centers and widths of the radial basis functions in the hidden layer. Meanwhile, impacts of different information criteria to the inversion results are compared, and an implementation of the proposed ICPSO algorithm is given. The optimized neural network has one hidden layer with 261 nodes selected by AKAIKE's information criterion(AIC) and it is trained on 32 data sets and tested on another 8 synthetic data sets. Two complex synthetic examples are used to verify the feasibility and effectiveness of the proposed method with two learning stages. The results show that the proposed method has better performance and higher imaging quality than three-layer and four-layer back propagation neural networks(BPNNs) and traditional least square(LS) inversion.
基金Project(20120009110035)supported by Specialized Research Fund for the Doctoral Program of Higher Education of ChinaProject(2011BAG01B05)supported by National Key Technology Research and Development Program of ChinaProject(2011AA110501)supported by National High-tech Research and Development Program of China
文摘Pantograph-catenary contact force provides the main basis for evaluation of current quality collection; however,the pantograph-catenary contact force is largely affected by the catenary irregularities.To analyze the correlated relationship between catenary irregularities and pantograph-catenary contact force,a method based on nonlinear auto-regressive with exogenous input(NARX) neural networks was developed.First,to collect the test data of catenary irregularities and contact force,the pantograph/catenary dynamics model was established and dynamic simulation was conducted using MATLAB/Simulink.Second,catenary irregularities were used as the input to NARX neural network and the contact force was determined as output of the NARX neural network,in which the neural network was trained by an improved training mechanism based on the regularization algorithm.The simulation results show that the testing error and correlation coefficient are 0.1100 and 0.8029,respectively,and the prediction accuracy is satisfactory.And the comparisons with other algorithms indicate the validity and superiority of the proposed approach.
基金Projects(51275138,51475025)supported by the National Natural Science Foundation of ChinaProject(12531109)supported by the Science Foundation of Heilongjiang Provincial Department of Education,China+1 种基金Projects(XJ2015002,G-YZ90)supported by Hong Kong Scholars Program,ChinaProject(2015M580037)supported by Postdoctoral Science Foundation of China
文摘To improve the computational efficiency of the reliability-based design optimization(RBDO) of flexible mechanism, particle swarm optimization-advanced extremum response surface method(PSO-AERSM) was proposed by integrating particle swarm optimization(PSO) algorithm and advanced extremum response surface method(AERSM). Firstly, the AERSM was developed and its mathematical model was established based on artificial neural network, and the PSO algorithm was investigated. And then the RBDO model of flexible mechanism was presented based on AERSM and PSO. Finally, regarding cross-sectional area as design variable, the reliability optimization of flexible mechanism was implemented subject to reliability degree and uncertainties based on the proposed approach. The optimization results show that the cross-section sizes obviously reduce by 22.96 mm^2 while keeping reliability degree. Through the comparison of methods, it is demonstrated that the AERSM holds high computational efficiency while keeping computational precision for the RBDO of flexible mechanism, and PSO algorithm minimizes the response of the objective function. The efforts of this work provide a useful sight for the reliability optimization of flexible mechanism, and enrich and develop the reliability theory as well.
文摘An isothermal compressive experiment using Gleeble 1500 thermal simulator was studied to acquire flow stress at different deformation temperatures, strains and strain rates. The artificial neural networks with the error back propagation(BP) algorithm was used to establish constitutive model of 2519 aluminum alloy based on the experiment data. The model results show that the systematical error is small(δ=3.3%) when the value of objective function is 0.2, the number of nodes in the hidden layer is 5 and the learning rate is 0.1. Flow stresses of the material under various thermodynamic conditions are predicted by the neural network model, and the predicted results correspond with the experimental results. A knowledge-based constitutive relation model is developed.
基金Projects(51178263,51378307)supported by the National Natural Science Foundation of China
文摘The static performance of inflatable structures has been well studied and the dynamic deployment simulation has received much attention. However, very few studies focus on its deflation behavior. Although there are several dynamic finite element algorithms that can be applied to the deflation simulation, their computation costs are expensive, especially for large scale structures. In this work, a simple method based on classic thermodynamics and the analytical relationship between air and membrane was proposed to efficiently analyze the air state variables under the condition of ventilation. Combined with failure analysis of static bearing capacity, a fast incremental analytical method was presented to predict both elastic and post wrinkling deflation process of inflatable structures. Comparisons between simplified analysis, dynamic finite element simulation, and a full-scale experimental test are presented and the suitability of this simple method for solving the air state and predicting the deflation behavior of inflatable structures is proved.
基金Project(70373017) supported by the National Natural Science Foundation of China
文摘Considering multi-factor influence, a forecasting model was built. The structure of BP neural network was designed, and immune algorithm was applied to optimize its network structure and weight. After training the data of power demand from the year 1980 to 2005 in China, a nonlinear network model was obtained on the relationship between power demand and the factors which had impacts on it, and thus the above proposed method was verified. Meanwhile, the results were compared to those of neural network optimized by genetic algorithm. The results show that this method is superior to neural network optimized by genetic algorithm and is one of the effective ways of time series forecast.
基金Project (A1420060159) supported by the National Basic Research of China project (60234030) supported by the National Natural Science Foundation of China project(05005A) supported by Youth Foundation of Central South University of Forestry & Technology
文摘An on-demand distributed clustering algorithm based on neural network was proposed. The system parameters and the combined weight for each node were computed, and cluster-heads were chosen using the weighted clustering algorithm, then a training set was created and a neural network was trained. In this algorithm, several system parameters were taken into account, such as the ideal node-degree, the transmission power, the mobility and the battery power of the nodes. The algorithm can be used directly to test whether a node is a cluster-head or not. Moreover, the clusters recreation can be speeded up.
