Research on configuration design and operation effect evaluation for ultra high voltage (UHV) vertical insulator cleaning robot

2019 ◽  
Vol 47 (1) ◽  
pp. 90-101 ◽  
Author(s):  
Yu Yan ◽  
Wei Jiang ◽  
An Zhang ◽  
Qiao Min Li ◽  
Hong Jun Li ◽  
...  
Keyword(s):  
Neural Network ◽  
Motion Planning ◽  
High Voltage ◽  
Bp Neural Network ◽  
Evaluation Method ◽  
Effect Evaluation ◽  
Content Type ◽  
Network Training ◽  
Cleaning Robot ◽  
Physical Prototype

Purpose This study aims to the three major problems of low cleaning efficiency, high labor intensity and difficult to evaluate the cleaning effect for manual insulators cleaning in ultra high voltage (UHV) converter station, the purpose of this paper is to propose a basic configuration of UHV vertical insulator cleaning robot with multi-freedom-degree mechanical arm system on mobile airborne platform and its innovation cleaning operation motion planning. Design/methodology/approach The main factors affecting the insulators cleaning effect in the operation process have been analyzed. Because of the complex coupling relationship between the influencing factors and the insulators cleaning effect, it is difficult to establish its analytical mathematical model. Combining the non-linear mapping and approximation characteristics of back propagation (BP) neural network, the insulator cleaning effect evaluation can be abstracted as a non-linear approximation process from actual cleaning effect to ideal cleaning effect. An evaluation method of robot insulator cleaning effect based on BP neural network has been proposed. Findings Through the BP neural network training, the robot cleaning control parameters can be obtained and used in the robot online operation control, so that the better cleaning effect can be also obtained. Finally, a physical prototype of UHV vertical insulator cleaning robot has been developed, and the effectiveness and engineering practicability of the proposed robot configuration, cleaning effect evaluation method are all verified by simulation experiments and field operation experiments. At the same time, this method has the remarkable characteristics of sound versatility, strong adaptability, easy expansion and popularization. Originality/value An UHV vertical insulator cleaning robot operation system platform with multi-arm system on airborne platform has been proposed. Through the coordinated movement of the manipulator each joint, the manipulator can be positioned to the insulator strings, and the insulator can be cleaned by two pairs high-pressure nozzles located at the double manipulator. The influence factors of robot insulator cleaning effect have been analyzed. The BP neural network model of insulator cleaning effect evaluation has been established. The evaluation method of robot insulator cleaning effect based on BP neural network has also been proposed, and the corresponding evaluation result can be obtained through the network training. Through the system integration design, the robot physical prototype has been developed. For the evaluation of other operation effects of power system, the validity and engineering practicability of the robot mechanism, motion planning and the method for evaluating the effect of robot insulator cleaning have been verified by simulation and field operation experiments.

scholarly journals Spatiotemporal Monitoring and Evaluation Method for Sand-Filling of Immersed Tube Tunnel Foundation

2021 ◽  
Vol 11 (3) ◽  
pp. 1084
Author(s):  
Peng Wu ◽  
Ailan Che
Keyword(s):  
Neural Network ◽  
Elastic Wave ◽  
Frequency Domain ◽  
Bp Neural Network ◽  
Evaluation Method ◽  
Monitoring And Evaluation ◽  
Strong Nonlinearity ◽  
Filling Process ◽  
Time Frequency ◽  
Feature Parameters

The sand-filling method has been widely used in immersed tube tunnel engineering. However, for the problem of monitoring during the sand-filling process, the traditional methods can be inadequate for evaluating the state of sand deposits in real-time. Based on the high efficiency of elastic wave monitoring, and the superiority of the backpropagation (BP) neural network on solving nonlinear problems, a spatiotemporal monitoring and evaluation method is proposed for the filling performance of foundation cushion. Elastic wave data were collected during the sand-filling process, and the waveform, frequency spectrum, and time–frequency features were analysed. The feature parameters of the elastic wave were characterized by the time domain, frequency domain, and time-frequency domain. By analysing the changes of feature parameters with the sand-filling process, the feature parameters exhibited dynamic and strong nonlinearity. The data of elastic wave feature parameters and the corresponding sand-filling state were trained to establish the evaluation model using the BP neural network. The accuracy of the trained network model reached 93%. The side holes and middle holes were classified and analysed, revealing the characteristics of the dynamic expansion of the sand deposit along the diffusion radius. The evaluation results are consistent with the pressure gauge monitoring data, indicating the effectiveness of the evaluation and monitoring model for the spatiotemporal performance of sand deposits. For the sand-filling and grouting engineering, the machine-learning method could offer a better solution for spatiotemporal monitoring and evaluation in a complex environment.

Multidimensional research on agrometeorological disasters based on grey BP neural network

2020 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Bingjun Li ◽  
Shuhua Zhang
Keyword(s):  
Neural Network ◽  
Natural Disaster ◽  
Bp Neural Network ◽  
Henan Province ◽  
Assessment System ◽  
Systematic Analysis ◽  
Content Type ◽  
Analysis And Evaluation ◽  
Grain Crops ◽  
Disaster Vulnerability

PurposeThe purpose of this study to provide a reference basis for effectively managing the risk of agrometeorological disasters in Henan Province, speeding up the establishment of a scientific and reasonable system of agrometeorological disasters prevention and reduction and guaranteeing grain security.Design/methodology/approachFirstly, according to the statistical data of areas covered by natural disaster, areas affected by natural disaster, sown area of grain crops and output of grain crops from 1979 to 2018 in Henan Province, China. We have constructed an agrometeorological disaster risk assessment system for Henan province, China, which is composed of indicators such as rate covered by natural disaster, rate affected by natural disaster, disaster coefficient of variation and disaster vulnerability. The variation characteristics of agrometeorological disasters in Henan Province and their effects on agricultural production are analyzed. Secondly, the grey relational analysis method is used to analyze the relation degree between the main agrometeorological disaster factors and the output of grain crops of Henan Province. Based on the grey BP neural network, the rate covered by various natural disaster and the rate affected by various natural disaster are simulated and predicted.FindingsThe results show that: (1) the freeze injury in the study period has a greater contingency, the intensity of the disaster is also greater, followed by floods. Droughts, windstorm and hail are Henan Province normal disasters. (2) According to the degree of disaster vulnerability, the ability to resist agricultural disasters in Henan Province is weak. (3) During the study period, drought and flood are the key agrometeorological disasters affecting the grain output of Henan Province, China.Practical implicationsThe systematic analysis and evaluation of agrometeorological disasters are conducive to the sustainable development of agriculture, and at the same time, it can provide appropriate and effective measures for the assessment and reduction of economic losses and risks.Originality/valueBy calculating and analyzing the rate covered by natural disaster, the rate affected by natural disaster, disaster coefficient of variation and disaster vulnerability of crops in Henan Province of China and using grey BP neural network simulation projections for the rate covered by various natural disaster and the rate affected by various natural disaster, the risk assessment system of agrometeorological disasters in Henan is constructed, which provides a scientific basis for systematic analysis and evaluation of agrometeorological disasters.

Research on the Optimized Algorithms on Neural Network

2012 ◽  
Vol 605-607 ◽  
pp. 2175-2178
Author(s):  
Xiao Qin Wu
Keyword(s):  
Neural Network ◽  
Bp Neural Network ◽  
Simulated Annealing Algorithm ◽  
Fitness Function ◽  
Strong Stability ◽  
Training Algorithm ◽  
Network Training ◽  
The Neural Network ◽  
Annealing Algorithm ◽  
Proper Operation

In order to overcome the disadvantage of neural networks that their structure and parameters were decided stochastically or by one’s experience, an improved BP neural network training algorithm based on genetic algorithm was proposed.In this paper,genetic algorithms and simulated annealing algorithm that optimizes neural network is proposed which is used to scale the fitness function and select the proper operation according to the expected value in the course of optimization,and the weights and thresholds of the neural network is optimized. This method is applied to the stock prediction system.The experimental results show that the proposed approach have high accuracy,strong stability and improved confidence.

Prediction and analysis of compressive strength of recycled aggregate thermal insulation concrete based on GA-BP optimization network

2020 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Jinsong Tu ◽  
Yuanzhen Liu ◽  
Ming Zhou ◽  
Ruixia Li
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Compressive Strength ◽  
Thermal Insulation ◽  
Bp Neural Network ◽  
Back Propagation ◽  
Training Sample ◽  
Recycled Aggregate ◽  
Generalization Performance ◽  
Content Type

Purpose This paper aims to predict the 28-day compressive strength of recycled thermal insulation concrete more accurately. Design/methodology/approach The initial weights and thresholds of BP neural network are improved by genetic algorithm on MATLAB 2014 a platform. Findings Genetic algorithm–back propagation (GA-BP) neural network is more stable. The generalization performance of the complex is better. Originality/value The GA-BP neural network based on the training sample data can better realize the strength prediction of recycled aggregate thermal insulation concrete and reduce the complex orthogonal experimental process. GA-BP neural network is more stable. The generalization performance of the complex is better.

Data Mining with BP Neural Network Algorithm Based MapReduce

2013 ◽  
Vol 380-384 ◽  
pp. 2915-2919 ◽  
Author(s):  
Jian Ming Cui ◽  
Yan Xin Ye
Keyword(s):  
Neural Network ◽  
Data Mining ◽  
Bp Neural Network ◽  
Resource Constraints ◽  
Training Time ◽  
Computing Platform ◽  
Network Training ◽  
Network Algorithm ◽  
Neural Network Algorithm ◽  
Cloud Computing Platform

Traditional massive data mining with BP neural network algorithm, resource constraints of the ordinary stand-alone platform and scalability bottlenecks and classification process serialization due to classification inefficient results, and also have an impact on the classification accuracy. In this paper, the Detailed description of the flow of execution of the BP neural network parallel algorithm in Hadoop's MapReduce programming model.Experimental results show that: the BP neural network under the cloud computing platform can greatly shorten the network training time, better parallel efficiency and good scalability.

Application of BP Neural Network Based on Immune Particle Swarm Optimization for Fault Diagnosis of Power Transformer

2013 ◽  
Vol 448-453 ◽  
pp. 3605-3609
Author(s):  
Yu Xin Zhang ◽  
Yu Liu
Keyword(s):  
Neural Network ◽  
Particle Swarm Optimization ◽  
Fault Diagnosis ◽  
Bp Neural Network ◽  
Power Transformer ◽  
Particle Swarm ◽  
Swarm Optimization ◽  
Network Training ◽  
Gradient Descent Algorithm ◽  
Ipso Algorithm

Cloing and hypermutation of immune theory were used in optimization on particle swarm optimization (PSO), an immune particle swarm optimization (IPSO) algorithm was proposed , which overcome the problem of premature convergence on PSO. IPSO was used in BP Neural Network training to overcome slow convergence speed and easily getting into local dinky value of gradient descent algorithm. BP Neural Network trained by IPSO was used to fault diagnosis of power transformer, it has high accuracy after experimental verification and to meet the power transformer diagnosis engineering requirements.

Sign in / Sign up

Export Citation Format

Share Document