The collective behavior of certain animals and insects has the characteristic of self-organization. The simple interactions among individuals can produce complex adaptive patterns at the level of the group. Recently,n...
详细信息
The collective behavior of certain animals and insects has the characteristic of self-organization. The simple interactions among individuals can produce complex adaptive patterns at the level of the group. Recently,new scientific investigation pointed out that desert locusts show extreme phenotypic plasticity in transforming between the lonely phase and the swarming gregarious phase depending on the population density,which is controlled by a serotonin called 5-hydroxytryptamine( 5HT). In this paper,based on the mechanism of the locusts' collective behavior,a new particle swarm optimization technique called LBPSO is studied. The number of swarms is selfadaptively adjusted by the acquired outstanding particles coming from behind the previous global best solution. The swarm sizes are related to the corresponding serotonin 5HT,which is determined by the optimization parameters such as global best and iteration number. And each swarm adopts one of three rules below according to its density, generalized social evolution strategy, generalized cognition evolution strategy and the independent moving strategy. A comparative study of LBPSO,social particle swarm optimization( SPSO), improved SPSO and the standard particle swarm optimization( StdPSO) on their abilities of tracking optima is carried out. And the results under four static benchmark functions and a dynamic function generator moving peaks benchmark( MPB)show that LBPSO outperforms the other three functions in both static and dynamic landscapes due to the introduced locusts' collective behavior.
A single model cannot satisfy the high-precision prediction requirements given the high nonlinearity between *** contrast,ensemble models can effectively solve this *** key factors for improving the accuracy of ensemb...
详细信息
A single model cannot satisfy the high-precision prediction requirements given the high nonlinearity between *** contrast,ensemble models can effectively solve this *** key factors for improving the accuracy of ensemble models are namely the high accuracy of a submodel,the diversity between subsample sets and the optimal ensemble *** study presents an improved ensemble modeling method to improve the prediction precision and generalization capability of the *** proposed method first uses a bagging algorithm to generate multiple subsample ***,an indicator vector is defined to describe these subsample ***,subsample sets are selected on the basis of the results of agglomerative nesting clustering on indicator vectors to maximize the diversity between ***,these subsample sets are placed in a stacked autoencoder for ***,XGBoost algorithm,rather than the traditional simple average ensemble method,is imported to ensemble the model during *** machine learning public datasets and atmospheric column dry point dataset from a practical industrial process show that our proposed method demonstrates high precision and improved prediction ability.
The steam system is an important part of the utility systems in process industry. The energy consumption and operation cost of the existing plant were increased due to the inefficient configuration of the steam system...
详细信息
Semi-supervised dimensionality reduction is an important research area for data classification. A new linear dimensionality reduction approach, global inference preserving projection (GIPP), was proposed to perform ...
详细信息
Semi-supervised dimensionality reduction is an important research area for data classification. A new linear dimensionality reduction approach, global inference preserving projection (GIPP), was proposed to perform classification task in semi-supervised case. GIPP provided a global structure that utilized the underlying discriminative knowledge of unlabeled samples. It used path-based dissimilarity measurement to infer the class label information for unlabeled samples and transformd the diseriminant algorithm into a generalized eigenequation problem. Experimental results demonstrate the effectiveness of the proposed approach.
In chemicalprocesses, there are both nonlinearity and slow dynamics among process variables. Meanwhile, at the beginning of faults, slight variation may not prompt the significant parametric changes. Traditional faul...
详细信息
To implement self-adaptive control parameters, a hybrid differential evolution algorithm integrated with particle swarm optimization (PSODE) is proposed. In the PSODE, control parameters are encoded to be a symbioti...
详细信息
To implement self-adaptive control parameters, a hybrid differential evolution algorithm integrated with particle swarm optimization (PSODE) is proposed. In the PSODE, control parameters are encoded to be a symbiotic individual of original individual, and each original individual has its own symbiotic individual. Differential evolution ( DE) operators are used to evolve the original population. And, particle swarm optimization (PSO) is applied to co-evolving the symbiotic population. Thus, with the evolution of the original population in PSODE, the symbiotic population is dynamically and self-adaptively adjusted and the realtime optimum control parameters are obtained. The proposed algorithm is compared with some DE variants on nine functious. The results show that the average performance of PSODE is the best.
Near-infrared( NIR) spectroscopy has been widely employed as a process analytical tool( PAT) in various fields; the most important reason for the use of this method is its ability to record spectra in real time to cap...
详细信息
Near-infrared( NIR) spectroscopy has been widely employed as a process analytical tool( PAT) in various fields; the most important reason for the use of this method is its ability to record spectra in real time to capture process properties. In quantitative online applications,the robustness of the established NIR model is often deteriorated by process condition variations,nonlinear of the properties or the high-dimensional of the NIR data set. To cope with such situation,a novel method based on principal component analysis( PCA) and artificial neural network( ANN) is proposed and a new sample-selection method is mentioned. The advantage of the presented approach is that it can select proper calibration samples and establish robust model effectively. The performance of the method was tested on a spectroscopic data set from a refinery process. Compared with traditional partial leastsquares( PLS),principal component regression( PCR) and several other modeling methods, the proposed approach was found to achieve good accuracy in the prediction of gasoline properties. An application of the proposed method is also reported.
In this paper, we propose an unsupervised learning method for jointly estimating monocular depth and ego-motion, which is capable to recover the absolute scale of global camera trajectory. In order to solve the genera...
详细信息
Vision-based formation control offers an alternate solution for unmanned aerial vehicles (UAVs) to work together in the external position system denied environment. In this paper, we present a vision-based formation c...
详细信息
In this paper, we propose a framework for trajectory planning in a 3D dynamic environment where other non-cooperative agents may obstruct the active quadrotor. A trajectory predictor is designed for the non-cooperativ...
详细信息
暂无评论