Wind power forecasting is of great significance in grid dispatching. This paper proposes a statistical model based on feature classification least squares support vector machine, which can predict short-term wind powe...
详细信息
Wind power forecasting is of great significance in grid dispatching. This paper proposes a statistical model based on feature classification least squares support vector machine, which can predict short-term wind power. First of all, this paper analyzes the data of an actual power plant. After analyzing the data, it is found that there is uncertainty in the existence of multiple powers at the same wind speed. Then, in order to resolve this uncertainty, the wind speed and wind speed trend samples are density clustered according to the DBSCAN method. The clustering results are divided into several categories, and the samples of different categories are modeled by least squares support vector machines. Finally, the effectiveness of the proposed prediction model is compared with that of unclassified samples through the prediction model. Simulation results show that the designed model has higher prediction power accuracy.
The online differentiation of a signal contaminated with bounded noise is addressed. A differentiator is developed that generates a Lipschitz continuous output, is exact in the absence of noise, and provides the optim...
The online differentiation of a signal contaminated with bounded noise is addressed. A differentiator is developed that generates a Lipschitz continuous output, is exact in the absence of noise, and provides the optimal worst-case accuracy among all possible exact differentiators when noise is present. This combination of features is not shared by any previously existing differentiator. Tuning of the developed differentiator is very simple, requiring only the knowledge of a bound for the second-order derivative of the signal. The approach consists in regularizing the possibly highly noisy output of a recently introduced linear adaptive robust exact differentiator and feeding it to a first-order sliding-mode filter designed to maintain optimal accuracy. The proposed regularization and filtering of this output allows trading the speed with which exactness is obtained for the feature of a Lipschitz continuous, hence less noisy, output. An illustrative example is provided to highlight the features of the developed differentiator.
This paper presents a control strategy for a planar three-link underactuated manipulator(UM) with a passive first link based on a wavelet neural network(WNN) model. Firstly, by using the particle swarm optimizatio...
详细信息
This paper presents a control strategy for a planar three-link underactuated manipulator(UM) with a passive first link based on a wavelet neural network(WNN) model. Firstly, by using the particle swarm optimization(PSO) algorithm, the target angles of all links are calculated according to the established kinematic model and the given target position. Then, a WNN model is trained to describe the coupling relationship between the passive link and the second link. The difference between the current angle and the target angle of the passive link is converged to zero by repeatedly controlling the second link to rotate an angle which is calculated by the trained WNN model. Next, the active links are controlled to rotate to their target angles with low speeds. Finally, the effectiveness of the proposed control strategy is verified through experimental results.
Imbalanced data with skewed class distributions and different misclassification costs is common in many real-world applications. Traditional classification approach does not work well for imbalanced data, because they...
详细信息
Imbalanced data with skewed class distributions and different misclassification costs is common in many real-world applications. Traditional classification approach does not work well for imbalanced data, because they assume equal costs for each class. To deal with this problem, cost-sensitive approaches assign different misclassification costs for different classes without disrupting the true original distributions of samples. However, due to lack of prior knowledge, the misclassification costs are usually unknown and hard to choose in practice. Whats more, even instances in the same class may have different misclassification costs. As an extension of class-dependent costs, this paper presents a composite cost-sensitive deep neural network(CCS-DNN) for imbalanced classification. A specifically-designed cost-sensitive matrix, which is composed of exampledependent costs and class-dependent costs, is embedded into the loss function to improve the classification performance. And the parameters of both the cost-sensitive matrix and the network are jointly optimized during training. The results of comparative experiments on some benchmark datasets indicate that the CCS-DNN performs better than other baseline methods.
In the field of face recognition and analysis, eye state detection is an essential step, which is the prerequisite and breakthrough of drowsiness estimation and auxiliary driving. This paper presents an eye state dete...
详细信息
In the field of face recognition and analysis, eye state detection is an essential step, which is the prerequisite and breakthrough of drowsiness estimation and auxiliary driving. This paper presents an eye state detection method based on Weight Binarization Convolution Neural Network(WBCNN). The weight of the network is constrained by binarization, which can limit the weight to 1 or-1, reducing the power dissipation and internal storage considerably. The human eye state features which can be extracted by convolution neural network effectively, and binary network not only contributes to reducing the storage size of the model, but also accelerates the computation. Experiments on eye state detection were conducted on the Closed Eyes in the wild(CEW) and FER2013 Databases, from which the results show that our method achieved average test accuracy of 97.41%on CEW. We used the FER2013 facial expression database for pre-training, which can make up for the lack of CEW training samples. The computational speed of non-binary is slower than binary network. Moreover, less storage capacity is required by our method.
The identification and positioning of insulator shed feature is of vital significance for insulator cleaning. However,there exist few texture features the insulator and its surface is reflective, which makes it diffic...
详细信息
The identification and positioning of insulator shed feature is of vital significance for insulator cleaning. However,there exist few texture features the insulator and its surface is reflective, which makes it difficult to identify and locate the convex points on the surface of the insulator shed. In order to identify and locate the insulator shed, we propose a method to extract the convex points of the insulator skirt based on the binary image of the insulator and the position of the central axis. Binocular vision technology is applied to obtain the depth change curve of the insulator on its axis, with the help of the depth threshold method and principal component analysis(PCA) method. At the same time, the pixel coordinates on the central axis are used to traverse the depth on the central axis to extract the shed bumps, thereby achieving the spatial positioning of the insulator shed bumps. The experiments show that the bumps on the insulator’s central axis can be well marked and used in the insulator cleaning robot system.
With the development of single-cell sequencing technology, it is a hot research topic to identify the cell types using single-cell sequencing data, and many single-cell clustering algorithms have been developed to stu...
详细信息
With the development of single-cell sequencing technology, it is a hot research topic to identify the cell types using single-cell sequencing data, and many single-cell clustering algorithms have been developed to study this issue. These methods capture partial information of single-cell sequencing data, and obtain the different performance on the same data set. Combining these different results into one can improve the accuracy and validity. Here, we proposed ECBN, Ensemble Clustering based on B ayesian Network. ECBN can ensemble several different results of state-of-the-art single cell clustering methods, such as Seurat, CIDR, SC3 and t-SNE+k-means, and generate a more optimal clustering result through Bayesian network. Experiments are carried on the 5 single cell data sets and compared with 4 individual single cell clustering methods and 3 integrative *** size of experiment data sets ranges from 822 to 3605 and the results show that our method can achieve good ***, ECBN can also use the graphical regularization to lighten the limitation which is generated by the different basis results.
Seismic simulation shaking table plays a key role in many industrial structural dynamic ***-precision waveform repetition has a great impact on the experimental accuracy of structural dynamic *** equipment of seismic ...
详细信息
Seismic simulation shaking table plays a key role in many industrial structural dynamic ***-precision waveform repetition has a great impact on the experimental accuracy of structural dynamic *** equipment of seismic simulation shaker table has obvious uncertainties in the modeling process,so it is difficult to describe the seismic simulation shaker table with a more accurate mathematical ***,this paper adopt the Linear Active Disturbance Rejection control(LADRC) to achieve high-precision reproduction of the waveform of the seismic simulation *** method will all the model uncertainty and external disturbance as total disturbance,using the extended state observer(ESO) to estimate the total disturbance,and by using state feedback control law of generalized disturbance compensation,to get better performance of resistance to *** method does not rely on the model of the specimen and does not need to obtain prior knowledge of the *** this paper,the inertial and single-degree-of-freedom elastic specimens are simulated for vibration test,and the control effect of this strategy is fully verified by comparison with the traditional PID control method.
Inconsistent feeder impedances in microgrids easily lead to uneven reactive power output of the inverter. This paper proposes an improved current-based droop control strategy for this problem. The droop coefficient is...
详细信息
Inconsistent feeder impedances in microgrids easily lead to uneven reactive power output of the inverter. This paper proposes an improved current-based droop control strategy for this problem. The droop coefficient is adjusted by the inverter capacity ratio, and the reactive current measured at the point of common coupling(PCC) is used as a reference value for improved differential control to compensate the voltage and control its reactive power output. It is found that the improved droop control has good adaptive ability and stability. The simulation results also prove the correctness and feasibility of the proposed strategy.
In the tuning system based on vision sensor, the object recognition and localization is very important. In order to meet the practical tuning of the microwave cavity filter by the mechanical arm, a global recognition ...
详细信息
In the tuning system based on vision sensor, the object recognition and localization is very important. In order to meet the practical tuning of the microwave cavity filter by the mechanical arm, a global recognition and localization method of multiple adjustable screws is proposed, which combines image mosaic and template matching. A high-definition global image of microwave cavity filter is composed of multiple local images according to feature point registration, and then all adjustable screws in the global image are recognized by template matching algorithm based on shape feature, which solves the problem of limited field of vision of a single industrial camera. It realizes the recognition, localization and numbering of all tuning objects in the whole tuning range, and transmits the coordinate information corresponding to the number to the mechanical arm. This method can be used to tune various types of microwave cavity filters, and it has certain versatility. The experimental results show that the recognition effect of the screws is favorable when the above method is applied to the actual tuning of 12-step microwave cavity filter with the hand-eye integrated four-axis mechanical arm. And it meets the requirements of tuning accuracy.
暂无评论