The partial classification algorithm is mainly used to predict the popularity of network news and to explore the best model to predict the popularity of network news, so as to help network news service providers predi...
详细信息
The partial classification algorithm is mainly used to predict the popularity of network news and to explore the best model to predict the popularity of network news, so as to help network news service providers predict the popularity of news before publication. The popularity of network news is predicted according to the data analysis process: first, UCI data sets are pre-processed;secondly, feature selection is conducted for the data sets by using recursive feature elimination algorithm;then modelling and analysis is carried out, and finally through the confusion matrix, risk map and ROC(Receiver Operating Characteristic) chart performance evaluation, the performance of the model is compared and analyzed. Through comparison, it is found that random forest is the best prediction model.
For economic goals, power maximization and mechanical load minimization are quite important for variable-speed wind turbines in the modern wind energy industry. However, these turbines are generally non-linear and hig...
详细信息
For economic goals, power maximization and mechanical load minimization are quite important for variable-speed wind turbines in the modern wind energy industry. However, these turbines are generally non-linear and highly coupled non-affine systems. In this work, an economic model predictive control is proposed to guarantee the economical control operation, meanwhile, improves the closed-loop optimal performance. Based on the 5-MW NREL wind turbine reference model with considering the tower fore-aft motion, the proposed control scheme is compared with the classical MPC scheme based on stepwise wind speed variations. The simulations result indicates the proposed controller effectiveness over the classical MPC.
Valley load regulations of thermal generators are one of the key measures to increase the utilization of wind power in China. For generators in valley load regulations, their coal consumption rates are much higher tha...
详细信息
Valley load regulations of thermal generators are one of the key measures to increase the utilization of wind power in China. For generators in valley load regulations, their coal consumption rates are much higher than those in basic operating conditions, and such characteristics are not considered in existing power system dispatch models. In this paper, a unit commitment(UC) model considering thermal generators in valley load regulations is presented, and coal consumption of generators are formulated as piecewise linear functions. The model can be solved by using existing mixed integer linear programming methods. Based on practical data of wind farm and thermal generator, numerical examples on the IEEE 30-bus system are presented. By comparing the results of our model with those of the conventional one, it is shown that valley load regulations of thermal generators lead to reduced generation costs and increased wind power utilization.
State Grid Corporation of China has accumulated a large number of electric power fault textual data based on ICT customer services. Effectively classifying and analyzing these texts can provide important clues for res...
详细信息
State Grid Corporation of China has accumulated a large number of electric power fault textual data based on ICT customer services. Effectively classifying and analyzing these texts can provide important clues for resolving new faults, and therefore helps customer service staffs provide more accurate fault diagnosis solutions. Because an occurred fault is possibly related to multiple classification labels, it is challenging to effectively classify the faults. In this paper, we present an ensemble learning based multi-label classification approach to analyzing electric power fault text data. Firstly, the power fault report data is pre-processed by word segmentation and stop word removal according to the structure of fault data. Each of fault text is represented as a TF-IDF vector. Then, we combine Binary Relevance with the Gradient Boosting ensemble learning algorithm for multi-label classification of fault texts. At last, the related experiments were made, and the experimental results show that our method is better than the traditional approaches such as Binary Relevance based on Logistic Regression and ML-KNN for fault text classification.
There are various deficiencies in network fault diagnosis methods but the need for fault diagnosis is further *** order to adapt to the current development needs,the paper applies the artificial neural network concept...
详细信息
There are various deficiencies in network fault diagnosis methods but the need for fault diagnosis is further *** order to adapt to the current development needs,the paper applies the artificial neural network concept to network fault *** at the slow convergence speed of traditional neural networks and the tendency to fall into the local optimal solution,first we attempt to add the momentum factor and then adopt rough set preprocessing *** simulation results show that the improved algorithm has a certain advantages for the original BP algorithm,and it has a certain value for future fault diagnosis research.
The traditional empirical tuning method of Linear Active Disturbance Rejection control (LADRC) parameters is obtained without considering the actuator rate limit. Once the actuator rate is limited, then the input and ...
详细信息
The traditional empirical tuning method of Linear Active Disturbance Rejection control (LADRC) parameters is obtained without considering the actuator rate limit. Once the actuator rate is limited, then the input and output signals of the actuator cannot be synchronized. Considering that there is no efficient strategy for tuning LADRC parameters when the actuator rate is limited. And the parameters tuning method of LADRC is complex and difficult. Taking the traditional LADRC parameter tuning value as the benchmark and combining with the Pigeon Inspired Optimization (PIO) algorithm, a LADRC parameter tuning method is proposed by constructing the quadratic objective function and changing the weight factor of the control increment. The simulation results show that the LADRC parameters optimization method based on PIO can keep the input and output signals of the actuator synchronized.
Wind direction information is of great significance to both wind energy assessment and wind power characteristic analysis. How to divide the wind direction sectors, while taking into account the continuous random fluc...
详细信息
Wind direction information is of great significance to both wind energy assessment and wind power characteristic analysis. How to divide the wind direction sectors, while taking into account the continuous random fluctuation of the wind direction in the time and space dimension, has become a hot topic worthy of study. This paper presents a wind sector division method considering spontaneous aggregation characteristics of wind-direction measured data. Firstly, the basic knowledge about Markov clustering(MCL) algorithm used in wind direction division is introduced. Secondly, the relative variations of wind direction are calculated for the periodic wind direction data, and its probability distribution is estimated statistically. The uniform division interval for the discrete states of Markov chain is then determined according to the cumulative probability quantile boundary of the relative wind direction variations. Then, the discrete states of wind direction are regarded as the nodes of a directional graph. And the graph using state transition matrix as the connection weight is clustered by the MCL algorithm. The evaluation indexes are also defined to judge the clustering performance. Finally, the actual data of the wind farm is adopted to verify the effectiveness of clustering results. The simulation results show that using MCL algorithm can fully utilize spontaneous aggregation characteristics of the wind direction to obtain effective wind-direction division results.
Batteries are widely used in power industry, so it is very important to ensure their safety. Therefore, the working parameters and working conditions of batteries should be checked regularly to ensure that all paramet...
详细信息
Batteries are widely used in power industry, so it is very important to ensure their safety. Therefore, the working parameters and working conditions of batteries should be checked regularly to ensure that all parameters are within the normal range. If abnormal conditions occur, timely protective measures can be taken to ensure that the whole power supply system will not outwork. Traditional manual measurement is time-consuming, laborious and inaccurate. Nowadays, the development of sensor technology and internet technology can be applied in battery measurement and online monitoring system. Considering the huge number of backup energy batteries, the large number of clients uploading data, the large amount of instantaneous access, and the scalability of system functions, backstage data terminals adopt micro-service architecture, which provides good reliability and scalability for the system.
Medical ultrasound devices are widely used in clinic because of its convenience, rapid and non-invasive. But ultrasound (US) images have the characteristics of large speckle noise, unclear target and low brightness. S...
详细信息
ISBN:
(数字)9781728144801
ISBN:
(纸本)9781728144818
Medical ultrasound devices are widely used in clinic because of its convenience, rapid and non-invasive. But ultrasound (US) images have the characteristics of large speckle noise, unclear target and low brightness. Since the deep learning theory has been developed, the accuracy of the tasks in the field of image has been greatly improved. In this paper, a deep neural network structure is established to automatic detect and track the liver vessel targets. Firstly, the dataset is augmented and preprocessed using histogram equalization. Secondly, the RetinaNet is implemented to extract the region of interest (ROI) in the US image. Then, the U-net is used to extract the features of the ROI, and deconvolution is implemented to restore the feature matrix to the size of the original image, which realize the automatic segmentation of blood vessels. Finally, the LSTM network is used to predict the information of vessels in the subsequent image. Experimental results show that the proposed algorithm is fast and robust. The accuracy of the ROI detection is 98.9%. The average error of the distance of the center point of the target is less than 1 mm.
Prosthetic hands are of a great interest. For this reason, control laws are required to improve the performances to the optimal level. Two approaches have been proposed to enhance the transient performances, ensure be...
详细信息
ISBN:
(纸本)9781665407830
Prosthetic hands are of a great interest. For this reason, control laws are required to improve the performances to the optimal level. Two approaches have been proposed to enhance the transient performances, ensure better accuracy and strengthen the robustness against the plant parameter variations. A Fractional Proportional Integral controller (FPI) tuned by Ant Colony optimization (ACO) algorithm has been synthesized and tested by simulations. Then a composite control law based on the Input Shaping approach and a FPI (IS-FPI) has been proposed. Performances have been discussed to evaluate the results and confirm the proposed technics.
暂无评论