| Since the development of the experimental type of the pulse experiments in the field of fusion to the steady-state long pulse experiments,the ways of processing the key segments such as data acquisition,transmission,storage and visualization in the steady-state long pulse experiments are intimately related to the experimental data feedback and the analysis of the experimental results.Therefore,the study of the data processing of the vital aspects of the steady-state long pulse experiments is the key to ensure the smooth experiment and the steady-state operation of the system,which is of great significance for further adjustment of the experimental parameters and system optimization.This paper focuses on the data processing of the above key aspects of the steady-state long-pulse experiments,and proposes the design and implementation of the real-time data processing system for steady-state long-pulse experiments based on the control,data access and communication(CODAC)system design theory proposed by the organization of International Thermonuclear Experimental Reactor(ITER).Then,the corresponding prototype system is implemented according to the design scheme.After system testing,the relevant test results show that the system implemented according to this design scheme can meet the needs of the experiment.The main research work conducted in this paper is as follows.1)Based on CODAC system theory,this paper proposes a distributed design scheme for real-time data processing system in steady-state long pulse experimental mode based on Experimental Physics and Industrial Control system(EPICS),and constructs a three-layer system design model with “task processing”,“storage processing” and “interaction processing” as the core concepts.2)A study of the data processing in the data acquisition subsystem is conducted,and an asynchronous data processing scheme for steady-state long pulse experiments is proposed.And based on the double buffer algorithm and MMAP technology,a data processing scheme for data exchange between memories is proposed.Then,the system debugging is completed simultaneously.In addition,considering the scalability and compatibility of the system,a device model is proposed based on EPICS to unify the device abstraction processing format and standardize the subsequent device development process.3)A study of the design of data storage management and server control subsystem is conducted.Based on TCP/IP protocol,the data transmission structure under highspeed sampling is proposed.And considering the possible system problems caused by hardware and software abnormalities and the storage limitation of EPICS,this paper adopts the service model of hot standby dual computer so that it can provide data storage and guarantee services while also providing a hardware and software implementation platform for system status monitoring and system remote services.4)A study of the control mode and data processing of the central control subsystem is conducted.In response to the diverse requirements of the experiments,the central control subsystem provides local and remote operation modes to meet the data processing requirements under different objective conditions;In addition,considering the concurrency exceptions that may be brought by different operation modes,this paper fully analyzes the potential concurrency problems of the system and proposes a reasonable concurrency handling mechanism.Finally,the integration and testing of the system are completed,and the results show that the overall operation of the system completed by relying on this design scheme is stable and reliable,and can meet the basic requirements of long pulse experiments. |
