Appropriate reliance on code clones significantly reduces development costs and hastens the development process. Reckless cloning, in contrast, reduces code quality and ultimately adds costs and time. To avoid this scenario, many researchers have proposed methods for clone detection and refactoring. The developed techniques, however, are only reliably capable of detecting clones that are either entirely identical or that only use modified identifiers, and do not provide clone-type information. This paper proposes a two-pass clone classification technique that uses a tree-based convolution neural network (TBCNN) to detect multiple clone types, including clones that are not wholly identical or to which only small changes have been made, and automatically classify them by type. Our method was validated with BigCloneBench, a well-known and wildly used dataset of cloned code. Our experimental results validate that our technique detected clones with an average rate of 96% recall and precision, and classified clones with an average rate of 78% recall and precision.

Energy Recovery Battery Testing Systems (ERBTS) plays an important role in battery manufacture. The conventional ERBTS configuration contains a fundamental transformer, and a bidirectional Direct Current (DC)–DC and Alternating Current (AC)–DC converter. All ERBTS are connected in parallel, thus constituting a special and complicated AC micro-grid system. Aiming at addressing their low energy recovery efficiency, complex grid-connected control algorithm issues for islanded detection, and complicated power circuit topology issues, a hierarchical DC-link voltage hybrid AC–DC micro-grid that contains composite energy storing devices is proposed. Moreover, an energy management optimal scheme for the proposed scheme is put forward. The system configuration of the proposed scheme is described in detail. Compared to the conventional scheme, the proposed scheme has the merits of simplified power circuit topology, no need for phase synchronous control, and much higher energy recovery efficiency and reliability. The validity and effectiveness of the proposed technique is verified through numerous experimental results.

In software testing, generating test data is quite expensive and time-consuming. The manual generation of an appropriately large set of test data to satisfy a specified coverage criterion carries a high cost and requires significant human effort. Currently, test automation has come at the cost of low quality. In this paper, we are motivated to propose a model-based approach utilizing the activity diagram of the system under test as a test base, focusing on its data flow aspect. The technique is incorporated with a search-based optimization heuristic to fully automate the test data generation process and deliver test cases with more improved quality. Our experimental investigation used three open-source software systems to assess and compare the proposed technique with two alternative approaches. The experimental results indicate the improved fault-detection performance of the proposed technique, which was 11.1% better than DFAAD and 38.4% better than EvoSuite, although the techniques did not differ significantly in terms of statement and branch coverage. The proposed technique was able to detect more computation-related faults and tends to have better fault detection capability as the system complexity increases.

Energy Management of a Grid-connected High Power Energy Recovery Battery Testing System Energy recovery;Battery testing system;Islanded detection;Micro-grid; Energy recovery battery testing systems (ERBTS) have been widely used in battery manufactures. All the ERBTS are connected in parallel which forms a special and complicated micro-grid system, which has the shortcomings of low energy recovery efficiency, complex grid-connected control algorithms issues for islanded detection, and complicated power circuit topology issues. To solve those shortcomings, a DC micro-grid system is proposed, the released testing energy has the priority to be reutilized between various testing system within the local grid, Compared to conventional scheme, the proposed system has the merits of a simplified power circuit topology, no needs for synchronous control, and much higher testing efficiency. The testing energy can be cycle-used inside the local micro-grid. The additional energy can be recovered to AC-grid. Numerous experimental comparison results between conventional and proposed scheme are provided to demonstrate the validity and effectiveness of the proposed technique.

최근 들어 초분광 영상을 이용하여 식물의 생육상태 분석 연구가 이루어지고 있다. 하지만 물리적 요인과 데이터의 복잡성 등 많은 요인이 초분광 영상 분석을 어렵게 한다. 본 논문은 초분광 영상과 DNN을 이용하여 작물의 생육상태 분류 기법을 제안한다. 초분광 영상을 통해 작물의 주요 정보를 얻고 DNN이 효과적으로 동작할 수 있도록 작은 패치 크기로 데이터를 자르고, 회전 시키는 등 데이터 전처리 기법을 사용한다. 실험에서는 파프리카 잎을 정상 잎, 해충 피해를 입은 잎 등 4종류의 잎과 배경으로 분류하였으며, 실험 결과 90.9%의 정확도를 보였다. 제안된 기법은 데이터 생성 방법이 DNN에 부담을 주지 않으며, 기존의 RGB 영상으로는 판단하기 어려운 다양한 생육 상태를 분류할 수 있다.

소프트웨어 개발에서 코드 클론은 비용 절감과 개발 속도 증가를 가져오는 요소로 작용한다. 그러나 코드의 무분별한 복사는 코드의 품질을 낮추고, 버그 수정에 많은 비용과 시간을 필요로 한다. 코드 클론을 검출하기 위한 기존 기술들은 온전히 복사된 코드 혹은 식별자만 수정한 코드 이외의 클론은 검출하기 어렵고, 클론의 유형 분류는 사람이 직접 수행해야 한다. 본 논문은 다양한 형태의 클론 검출과 검출된 클론의 유형을 자동으로 분류하는 것을 목표로 TBCNN(Tree-Based Convolution Neural Network) 기반 클론 분류 방법을 제안한다. 제안한 방법을 코드 클론 검출 방법 검증에 많이 사용하는 BigCloneBench 데이터 셋을 이용하여 실험하였다. 그 결과, 제안한 방법은 78%의 재현율과 정밀도로 4가지 유형의 코드 클론을 분류할 수 있었다.