FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2325098103> ?p ?o ?g. }

• W2325098103 abstract "Emulsifier is the key equipment in the production of emulsion explosive. The condition of the emulsifier is very important to the production safety of the emulsion explosive. The paper focuses on emulsifier fault warning in the production of emulsion explosive. Now the emulsifier fault warning is based on the threshold. That method can not warn fault timely. The paper proposes a fault early warning method which is on the BP neural network. The neural network model is established by the method. The condition of the emulsifier which is important to the production safety can be predicted through the model. According to the simulation example, the model can predict the conduction of the emulsifier well. Introduction Emulsifier is the key equipment in the production of emulsion explosive. Emulsifier fault is the main cause of the most production safety accidents in the past ten years. So it is important to warning emulsifier fault. The existing fault warning is on the monitoring parameters. Fault warning is triggered if the parameters exceed the threshold [1]. While the accident causes by the emulsifier fault has often happened. The method is not effective to warn fault in time. To mix the oil phase solution and aqueous solution adequately, emulsifier runs in high speed. The temperature and flow rate of the oil phase solution or aqueous solution affect the condition of emulsifier. It is also affected by the pressure and current of the emulsifier [2]. Combining with the working principle of the emulsifier, the paper proposes a fault warning method on the BP neural network. Fault early warning process A large numbers of historical data about production line of emulsion explosive is stored in the configuration software. Based on the BP neural network and the historical data, a model is built. This method for fault warning is the improvement of emulsifier threshold alarm used widely now. Fig.1 shows the process for emulsifier fault early warning. sensors date preprocess monitoring online BP neural network prediction fault early warning threshold alarm feature extraction historicaldata Fig.1 Fault early warning process The selection of research object Emulsifier is a kind of rotating equipment. It produces emulsion matrix by stirring specific solution and oil phase. During the stirring process, emulsifier vibrates. The amplitude of 3rd International Conference on Machinery, Materials and Information Technology Applications (ICMMITA 2015) © 2015. The authors Published by Atlantis Press 1836 vibration is controlled within certain range in normal conditions. Abnormal vibration happens when something goes wrong with the emulsifier. So vibration is selected as the main index to show the condition of the emulsifier. According to the operation of the emulsifier, the other 7 parameters which are effected the vibration are selected for emulsifier fault early warning. The 7 parameters are the temperature and flow rate of the oil phase solution or aqueous solution, the pressure and current of the emulsifier, the temperature of the emulsion matrix. BP neural network algorithm Artificial neurons are the basic unit of the BP neural network. Fig.2 shows an artificial neurons model. In the Fig.2, xi is one of input and yi is one of the output,si is the input which is outside the neurons,wij is the weight between two artificial neurons [3]." @default.
• W2325098103 created "2016-06-24" @default.
• W2325098103 creator A5011643959 @default.
• W2325098103 creator A5041831011 @default.
• W2325098103 date "2015-01-01" @default.
• W2325098103 modified "2023-09-23" @default.
• W2325098103 title "The emulsifier fault early warning on BP neural network theory" @default.
• W2325098103 cites W2480510902 @default.
• W2325098103 doi "https://doi.org/10.2991/icmmita-15.2015.343" @default.
• W2325098103 hasPublicationYear "2015" @default.
• W2325098103 type Work @default.
• W2325098103 sameAs 2325098103 @default.
• W2325098103 citedByCount "0" @default.
• W2325098103 crossrefType "proceedings-article" @default.
• W2325098103 hasAuthorship W2325098103A5011643959 @default.
• W2325098103 hasAuthorship W2325098103A5041831011 @default.
• W2325098103 hasBestOaLocation W23250981031 @default.
• W2325098103 hasConcept C127313418 @default.
• W2325098103 hasConcept C154945302 @default.
• W2325098103 hasConcept C165205528 @default.
• W2325098103 hasConcept C175551986 @default.
• W2325098103 hasConcept C29825287 @default.
• W2325098103 hasConcept C41008148 @default.
• W2325098103 hasConcept C50644808 @default.
• W2325098103 hasConcept C76155785 @default.
• W2325098103 hasConceptScore W2325098103C127313418 @default.
• W2325098103 hasConceptScore W2325098103C154945302 @default.
• W2325098103 hasConceptScore W2325098103C165205528 @default.
• W2325098103 hasConceptScore W2325098103C175551986 @default.
• W2325098103 hasConceptScore W2325098103C29825287 @default.
• W2325098103 hasConceptScore W2325098103C41008148 @default.
• W2325098103 hasConceptScore W2325098103C50644808 @default.
• W2325098103 hasConceptScore W2325098103C76155785 @default.
• W2325098103 hasLocation W23250981031 @default.
• W2325098103 hasOpenAccess W2325098103 @default.
• W2325098103 hasPrimaryLocation W23250981031 @default.
• W2325098103 hasRelatedWork W2356557657 @default.
• W2325098103 hasRelatedWork W2365567737 @default.
• W2325098103 hasRelatedWork W2365745636 @default.
• W2325098103 hasRelatedWork W2372829958 @default.
• W2325098103 hasRelatedWork W2373962361 @default.
• W2325098103 hasRelatedWork W2376596949 @default.
• W2325098103 hasRelatedWork W2386387936 @default.
• W2325098103 hasRelatedWork W2394139458 @default.
• W2325098103 hasRelatedWork W1629725936 @default.
• W2325098103 hasRelatedWork W2140808311 @default.
• W2325098103 isParatext "false" @default.
• W2325098103 isRetracted "false" @default.
• W2325098103 magId "2325098103" @default.
• W2325098103 workType "article" @default.