| The distributed multi-domain modeling and simulation system based on Web service has become a hotspot of distributed multi-domain modeling and simulation system in the next generation of system simulation.Data volume of modelica simulation is too large to make the simulation data storage space consumption to be large,Simulation data shared and transferred between computers is time-consuming,which is not conducive to the cooperation of experts in many fields and seriously hindered the further research and development of the Modelica modeling and simulation system based on Web services.Efficient compression and storage of Modelica simulation data is an important method to improve the efficiency of Modelica simulation data storage and transmission in the Internet environment.In this paper,the data characteristics and application characteristics of Modelica simulation data are summarized from the principle of Modelica modeling and simulation.Then based on the data characteristics and application characteristics of the Modelica simulation data,the compression and storage of the Modelica simulation data is studied from three aspects: the compression algorithm,the memory structure and the memory coding.Combined with the application characteristics of Modelica simulation data,the storage structure of the Modelica simulation data is to be re-designed.This kind of storage structure is to divide the Modelica simulation data by the variables to storage,and then organize the file tree and build the multi-level index table.This method not only can clearly reflect the logical relationship between the data variables of Modelica simulation,but also reduce the time they read the query,and more suitable for applied in the storage of Modelica simulation data in the Web environment.Aiming at Modelica simulation data and combining with its own data characteristics,an improved algorithm of SDT compression algorithm,DCSDT is proposed,and the test results show that the compression error of DCSDT algorithm in Modelica simulation data is reduced by 14%~20% compared with SDT algorithm,and the compression ratio is increased by 19%,especially in the compression of periodic data,the compression ratio is significantly increased.The coding method is designed for the Modelica simulation data compressed by the DCSDT algorithm.The time stamp is stored in γ code.For integer data of Modelica simulation data,first,transfer the diffenrence of data sequence,and then use variable length byte coding.Finally,a MYDTI code is designed for floating point data of Modelica simulation data.The main idea of MYDTI coding is to transform the floating point number into two integers,and using variable length coding for storage. |