Search

Requirements-to-code traceability link recovery (RC-TLR) can establish connections between requirements and target code artifacts, which is critical for the maintenance and evolution of large software systems. However, to the best of our knowledge, there is no existing experimental study focused on state-of-the-art (SOTA) methods for the RC-TLR problem, and there is also a lack of uniform benchmarks for evaluating new methods in the field. We developed a framework to identify SOTA methods using the Systematic Literature Review method and applied it to research in the RC-TLR field from 2018 to 2023. Through experiments replication on 13 datasets using 6 methods, we observed that for information retrieval-based methods, Close Relations between Target artifacts-based method (CRT), TraceAbility Recovery by Consensual biTerms (TAROT), and Fine-grained TLR (FTLR) performed well on COEST dataset, while Combining Part-Of-Speech with information-retrieval techniques (Conpos) and TAROT achieve promising results in large datasets. As concerns machine learning-based methods, Random Forest consistently exhibits strong performances on all datasets. We hope that this study can provide a comparative benchmark for performance evaluation in the RC-TLR field. The resource repository that we have established is expected to alleviate the workload of researchers in performance analysis, and promote progress of the field.

Traceability links between issues and commits (issue-commit links recovery (ILR)) play a significant role in software maintenance tasks by enhancing developers’ observability in practice. Recent advancements in large language models, particularly pre-trained models, have improved the effectiveness of automated ILR. However, these models’ large parameter sizes and extended training time pose challenges in large software projects. Besides, existing methods often overlook the association and distinction among artifacts, leading to the generation of erroneous links. To mitigate these problems, this paper proposes a novel link recovery method called MTLink. It utilizes multi-teacher knowledge distillation (MTKD) to compress the model and employs an adaptive multi-task strategy to reduce information loss and improve link accuracy. Experiments are conducted on four open-source projects. The results show that (i) MTLink outperforms state-of-the-art methods; (ii) The multi-teacher knowledge distillation maintains accuracy despite model size reduction; (iii) The adaptive multi-task tracing method effectively handles confusion caused by similar artifacts and balances each task. In conclusion, MTLink offers an efficient solution for ILR in software traceability. The code is available at https://zenodo.org/records/10321150.

Abstract Requirements management (RM) plays an important role in requirements engineering. The development of machine learning (ML) is in full swing, and many ML software management techniques had been used to improve the performance of RM methods. However, as no research study is known that exists systematically to summarise the ML methods used in RM. To fill this gap, this paper adopts the systematic mapping study to survey the state‐of‐the‐art ML methods for RM primary studies and were finally selected in this mapping, which was published on 36 conferences and journals. The 24 factors affecting the ML method of RM are determined, of which 9, 11 and 4 are the three parts of RM, namely requirements baseline maintenance, requirements traceability and requirements change management separately. The 18 objectives of the ML method for RM are summarised, of which 6, 7 and 5 are the three parts of RM. The eight ML methods used in RM and their time sequence are summarised. The 18 evaluation indexes for RM in the ML method are determined, and the performance of these methods on these parameters is analysed. The research direction of this paper is of great significance to the research of researchers in demand management.

Nonlinear boundary control technique can offer superior dynamic behavior and eliminate the LCL-filter resonance in the grid-connected inverter. However, the chattering might be aroused by the distorted intracycle voltage information in the control law. It becomes worse when the nonlinear boundary controller is directly applied in three-phase three-wire inverter configuration. A hybrid current control scheme utilizing the decoupled predictive hysteresis capacitor current controller integrated with the linear grid current controller is proposed in the paper. By applying the 60° discontinuous pulse-width modulation scheme and using the virtual dual-buck decoupled structure, two separate sets of switching criteria for dictating the switching states of two specific half-bridge legs are formulated to avoid the interactions amongst phases. A small-signal ac model is also built to study the transfer characteristics of such hybrid current controlled inverter system. Simulation and experimental results have been offered to evaluate the effectiveness of the proposed method. Both results meet the theoretical predictions.

To ease the heavy burden of inverter control, including line current regulation, harmonic current suppression, filter resonance suppression, grid distortion suppression, phase locked loop etc., the study proposes a master–slave current regulation method for an LCL-filter-based grid-connected inverter operating under variable grid condition. The proposed system is composed of a grid-connected voltage-source inverter (GC-VSI) in series with a bidirectional voltage-source converter (B-VSC) at the inverter output. The GC-VSI is responsible for line current regulation, while the B-VSC takes charge of harmonic current suppression and grid impedance cancellation. Owing to low volt-ampere of the B-VSC, the efficiency of the inverter system is not sacrificed. To confirm its validity, a simulation platform with the proposed inverter system has been built. The simulation results demonstrate the performance enhancement of the proposed inverter system operating under different grid conditions.