Distributed Spark Automatic Monitoring And Control System For Pipeline

Factory production process is not standardized,and the safeguard measures are not sound enough.Dust explosion accidents occur sometimes.Detecting and eliminating ignition sources and eliminating through spark monitoring equipment is an efficient precaution.It is necessary to improve the supervision.Current commercial watching equipment does not have network function,which cannot meet the new needs.This thesis designs spark moni-toring equipment and builds a remote realtime monitoring platform for unified management.It further ensures safe production.This paper designed and realized a distributed,embedded pipeline spark automatic moni-toring and control system.It realtime watches the production pipeline spark and equipment operating status and other information.It also synchronizes data to the cloud server.It is convenient for the safety supervision department and factory personnel to oversee,maintain and manage the system.That improves the safety of production supervision,the overall pro-tection of safe production.This big system contains three parts: control system,regional management system and subn-ode system.The control system software mainly includes the user interface,database man-agement and network communication.The regional management module is responsible for linking the various subnodes in the region.It also collects the operation data of the subn-odes,completes the configuration and fulfills test and other commands issued by the control system.The subnode system is responsible for completing the specific business functions.Subnode module provides many different types with unified interface.So,new subnode modules can be quickly developed as needed.Compared with other commercial devices,the system has increased network functions.It focuses on enhancing scalability and stability,and has certain advantages in performance,security and ease of use.Because of the complexity of the factory environment,the signal sent by wireless commu-nication is susceptible.That causes network congestion.This thesis analyzes the classical cache management mechanism in the DTN.Based on the network environment,this thesis presents a caching management mechanism based on the importance of data.It contains three parts of management strategy: Divide all messages into two priority queues,malfunction events and spark events.Important malfunction event message queue send preferentially.Based on the number of copies,TTL,and spark levels of spark event messages,the design formula calculates the importance of the message,sorting and managing the messages.The ACK response mechanism is designed to remove useless copies of messages from the net-work,and to further reduce the network load.The cache management mechanism designed in this thesis effectively improves the success rate of network message delivery.It guar-antees the success rate and delivery speed of important data.It completes the performance verification through simulation test.Compared with the traditional DO method,the over-all success rate of message transmission is increased by 6%.The success rate of important message transmission is increased by 20%.And the transmission delay is decreased by 30%.The system has been deployed in several factories and tested for more than one year.Several problems were found and repaired.The system has been upgraded and maintained.It meets the design needs.