| China's petroleum industry is an important one of the pillar industries. It is closely related to the development of national strategies, global political and economic developments. Under the powerful driving force of energy demand and oil industry, Geophysical logging has made rapid development. Conversely, the technological progress of geophysical logging has accelerated development of the oil industry as one of the top ten technologies. It is widely used in oil and gas field exploration and development in the entire process, providing an important basis for the information for petroleum geology and engineering and technical personnel to search for and evaluation of oil and gas and solving a series of geological problems.Currently, China National Logging Corporation (CNLC) has self-developed a set of logging equipment, along with the need for such equipment manufacturing vehicle logging data acquisition and processing system and rapid visual interpretation system. Logging data processing and interpretation is in accordance with the comprehensive geological scheduled tasks, using computers to log data processing, and integrated geological, logging and development of a comprehensive data analysis and interpretation to resolve the stratigraphic division, oil and gas reservoir, and useful deposits evaluation and exploration, and other geological and engineering technology issues, and to interpret the results in graphical form or in the form of data directly to them. It is the most important component which realizes technical and economic benefits last and direct.Data Access is the basis of the important components in logging data processing and comprehensive explanation of software. Each log data processing interprets software components necessary to support the bottom. Developing logging software faces the most important issue is defining data format and the preservation of design data access application program interface (API). At present, logging data with extensive sources, large format, different content and structure increases the complexity of software development on the data-processing software and explain, and more generally to its own software data format as the center, with the top-level data access module independent evolution. Therefore, in this paper, through a variety of data formats of extensive research, based on the object-oriented method and design patterns, bring forward a common data format solution for logging framework, providing a uniform data access interface for log interpretation software designed.Different sources of data format of the series modules, such as provided by standardized format organizations, or by third-party companies, or by their own solutions, using the framework of the package, can be easily added to the system, to add the support of the new format. Through the use of frameworks, makes the data format for has a good flexibility scalability, and maintainability which is necessary modules of this series in each of the logging data processing system, conducing to iteratively develop decoding part and the upper system. The solution of the expansibility is to edit XML configuration files and add or replace DLL to expand, instead of modifying codes, recompile circumstances. The realization is in means of design focusing on interface and the reflection mechanism in .NET.The whole work of designing framework mainly include two levels: data access layer and output object layer. Data access layer provides a universal data access interfaces to provide data services for the upper application; output object layer converts various output objects (such as conventional curve, waveform, etc.), log header, two-dimensional data table to the bottom concrete data format.Another work in this paper is to use the framework to realize various data format decoding, and to produce a simple data format conversion program using the data access interface of the framework. Universal data access interface is basically defined by the functional needs of upper system. When realize specific data format under the framework some problems encountered are as follows: in the format conversion, there are some differences between different file formats; the data access interface is universal, so inevitably data access interface and data format standards are inconsistent; some formats support user-defined information. Because of the reasons mentioned above, realizations for data format under the framework follow the following principles: This module can read and write standard and customized information of data format files generated by this module; this module can read and write standard information of data format files generated by external program, but not interpret the user-defined information; when data format conversion, it is allowed that loose some information that destination file doesn't support.Now the decoding for LAS has been done, and applied under the framework. The principles of other formats and LAS format are similar. The remaining work would be completed later.LAS is a used ASCII coding logging data format. Version 2.0 is the most logging data processing system product support and widely used. It is the industry's widely accepted data formats. However, its higher 3.0 version does not always support the current system, according to the author known only Schlumberger's Toolbox format conversion tools to support DLIS data file format conversion LAS 3.0 document. Given the practicality of LAS 2.0, LAS 3.0 will be gradually applied to actual production soon. Therefore, to achieve the decoding for LAS 3.0 also has an important and practical value.In this paper, when realize LAS format decoding, facade pattern is used to provide a simple access interface and some relative data structure to simplify. In the internal module design, the design principles are that data are separated from operation, and preferentially use composition for operations rather than inherit. Inherit is applied for data to represent the level of the LAS format specification. Inherit is useful for abstract and reuse. Composition is applied to construct operation classes instead of inherit. It ensures encapsulation and"black box"reuse. In LAS decoding module, multi-iteration algorithm is designed to simplify the program structure. LAS decoding module also supports LAS 2.0 and LAS 3.0 two versions, and supports the automatic identification for files versions. |
PDF Full Text Request