Nikolay Markov

Model-driven Platform for Oil and Gas Enterprise Data Integration

The necessity of data interoperability is very important for the oil and gas enterprise environment. Different information systems and applications cover the range of the IT infrastructure and need to interoperate. Most of the industrial companies do not have common rational strategy for their automation. That leads to the lack of integration between purchased and implemented software that is necessary for various departments. Integration problems prevent companies from building common information space. An instrumental platform for oil and gas enterprise data integration has been considered as the problem solution. The proposed platform combines Service-Oriented Architecture principles and Model-Driven Approach and uses branch standard PRODML as a model for object mapping.

Example of implementation of MES “Magistral-Vostok” for oil and gas production enterprise

Todays trend in automation of the large industrial enterprises is gradual filling of intermediate layer between low level systems of automation (management information systems of technological process) and control systems of the enterprise on the basis of ERP and BI. Traditionally MES which users are, first of all, engineers and analysts of the main experts here apply for a link role. The overwhelming majority of the systems presented today in the market, have a narrow branch orientation. It is caused by complexity of balance between the flexibility of architectural and interface decisions necessary for expansion of systems, and convenience to the user, wishing to have the simple tool of the decision of the daily problems. Design and implementation of such systems is interfaced to a number of serious problems, as characteristic for working out of the big distributed information systems, and specific for MES. An experience of solving such problems on example of MES “Magistral-Vostok” by Tomsk polytechnical university is considered.

Entropy-based Intra-coding RDO estimation for HEVC

The neoteric H.265/HEVC video compression standard provides two times higher compression efficiency compared to the current industrial standard H.264/AVC. However the coding complexity increased. The main bottleneck of the encoder is rate-distortion optimization (RDO) estimation module as it involves numerous sequential syntax-based binary arithmetic coding (SBAC) loops. In this paper we present entropy-based RD estimation for HEVC intra compression. Our RDO implementation eliminates sequential block interdependencies and reduces RD estimation complexity. Our rate estimation algorithm provides the bit rate overhead between 0.6% to 6.0% on the reference HM encoder in All Intra Main configuration, while compression time reduction is 20%.

Iterative intra prediction search for H.265/HEVC

Emerging video compression standard H.265/HEVC provides up to 2 times better compression efficiency compared to H.264/AVC standard. However, it has higher computational complexity. In this paper, we propose iterative intra prediction search for H.265/HEVC encoder to reduce the number of prediction modes for estimation. We base on our research of computational complexity and usage frequency of H.265/HEVC intra prediction. On JCT-VC test sequences, we got about 40% encoding time reduction for HM 10.1 intra-only coding with negligible bitrate increase and PSNR quality degradation. We also offer some additional speed-up techniques, including fast prediction error estimation.

Contemporary video compression standards: H.265/HEVC, VP9, VP10, Daala

In this paper we compare compression efficiency of the latest video coding standards H.265/HEVC, VP9, VP10 and Daala to H.264/AVC with the help of reference video encoders available. Experimental results show that H.265/HEVC on average provides 37–40% bitrate savings, VP9 provides 9–11%, VP10 — 10–12% bitrate savings, while Daala provides 8–9% bitrate overhead on average.

Fractal compression of grayscale and color images: Tools and results

This paper presents fast fractal image compression algorithm, based on quadtree image partitioning, Fisher block classification, block preparations and multithreaded coding process. Investigation results on efficiency of fractal still grayscale and color images compression are presented with comparison with JPEG results.

Manufacturing execution systems intellectualization: oil and gas implementation sample

Up-to-date trend in industrial automation is implementation of Manufacturing Execution Systems (MES) everywhere and in oil and gas industry. Conception of MES is constantly in progress. Many researches suppose that analytical features, available for low-end users (engineers, dispatchers, geologists, etc.) are necessary in manufacturing management, but today there is no ready-to-use framework applicable to make intelligent manufacturing systems for oil and gas industry. A model-driven approach of MES intellectualization and an original iMES framework proposed. iMES based on functions of the traditional MES (within MESA-11 model), business intelligence (BI)-methods (On-Line Analytical Processing & Data Mining) and production markup language (industrial data standard for oil and gas production). Case study of well tests results validation using iMES framework is considered.

Information Technologies in Industrial Enterprises Production Assets Management

The paper discusses problems of representing information about production assets in databases of information systems in tasks of assets registration and recording. Main approaches of representing asset registration data in a relational database are considered. An improved approach based on inheritance implementation is proposed. Its advantage is good enough flexibility, good performance and reach security. Appraisal information about implementation of proposed solutions in some oil-and-gas enterprises of Western Siberia is given.

Fast Search for Intra Prediction Mode in H.265/HEVC Video Compression

Mechanical engineering, chemical engineering and other industries have a high demand for the video compression systems that are used, e.g., in CCTV and video sensing. The newest video compression standard H.265/HEVC provides the compression rate of 100–300 times to the uncompressed video. The side effect is the increase of a computational complexity of the compression system. This high complexity obstructs the industrial implementation of H.265/HEVC video compression systems. One of the main objectives is to reduce intra compression complexity. In this paper, we present our algorithm of a fast search for intra prediction mode in H.265/HEVC video compression. The algorithm on average provides only 1.9% bitrate increase with 41% timesaving.

Development tools for common information space of distributed industrial company

Analysis of distributed industrial companies automation problems shows that most of the companies have a lot of stand-alone heterogeneous information systems and applications. It indicates to urgent problem of the system integration scientific basis establishing and, in practice, companys common information space development. This work presents original software tools for developing of common information space of distributed industrial company by means of information systems and applications data integration. Software tools principles of realization are offered and integrated model of enterprise industrial data is designed. The main components of the tools are a core and adapters that provide interaction between the core and the applications (or information systems). Algorithms of data integration participants coordination, data models structural matching, etc. are developed. The approach of data collecting from automated process control systems and delivering them to the top levels of enterprise management is offered. The results of developed algorithms and approach efficiency analysis are considered. The features of software tools architecture and development are described. The results of the software tools implementation within common information space development for oil and gas production company with big amount of remote fields are given.