Abstract
Evaluating and documenting the robustness of forecasting models to different input conditions is important for their responsible deployment in real-world applications. Time series forecasting models often exhibit degraded performance in the form of unusually large errors, high uncertainty, or hubris (high errors coupled with low uncertainty). Traditional stress testing approaches rely on manually designed adverse scenarios that fail to systematically identify unknown stress factors, in which data characteristics indicate potential issues. To overcome this limitation, this paper introduces MAST (Meta-learning and data Augmentation for Stress Testing), a novel method for stress testing forecasting models. MAST leverages model outputs (error scores and prediction intervals) to automatically identify and characterize input conditions that induce stress. Specifically, MAST is a binary probabilistic classifier that predicts the likelihood of forecasting model stress based on time series features. An additional contribution is a novel time series data augmentation approach based on oversampling or synthetic time series generation, that improves the information about stress factors in the input space, resulting in increased stress classification performance. Experiments were conducted using 6 benchmark datasets containing a total of 97.829 time series. We demonstrate how MAST is able to identify and explain input conditions that lead to manifestations of stress, namely large errors, high uncertainty, or hubris.

Abstract
Abstract Generative AI has been rapidly adopted by the software development industry in various ways, offering innovative approaches to transforming requirements into working software. Combining Generative AI with Test-Driven Development (TDD) presents a creative method to accelerate this transformation. However, questions remain about ChatGPT’s readiness for this challenge, including the techniques and best practices required for success and the scenarios where this approach can consistently deliver results. To explore these questions, we designed a study where a group of master’s students performed programming assignments using TDD, first independently and then with the support of ChatGPT. The three assignments represent distinct scenarios: mathematical calculations (function), text processing (class), and system integration (class with dependencies). We performed a qualitative analysis of the submitted code and reports identifying key strategies that significantly influence success rates, such as providing contextual information, separating instructions in prompts following an iterative process, and assisting AI in fixing errors. Among the scenarios, the integration task achieved the highest performance. This study highlights the potential of leveraging Generative AI in TDD for software development and presents a list of effective strategies to maximize its impact. By applying these positive strategies and avoiding identified pitfalls, this research marks a step toward establishing best practices for integrating Generative AI with TDD in software engineering.

Abstract
This chapter proposes a day-ahead scheduling framework in an energy hub (EH), integrating different energy conversion and storage technologies to efficaciously fulfill various types of load demands. The mentioned EH is capable of synchronously managing electrical, cooling, and heat load demands. The system is equipped with a combined heat and power (CHP) generating unit that efficiently supplies both heat and electricity. Furthermore, there are an electric heat pump and a boiler that also supply the heating load, while the heater is specifically employed for direct heating usage. The system includes an absorption chiller to supply a cooling load. This chiller absorbs waste heat from the CHP unit, resulting in improved energy efficiency. Battery storage systems enable the efficient use of energy by storing surplus power during times of low demand for future consumption. In addition, solar photovoltaic panels are included to capture renewable energy, therefore decreasing reliance on traditional energy sources and mitigating environmental consequences. The EH also includes a saltwater desalination technology operating together with the energy network to ensure the supply of freshwater, which is especially vital in dry areas. The desalination process is fueled by both renewable and produced thermal energy, thus maximizing resource use and reducing operating costs. The presented scheduling model has been formulated within a mixed-integer linear programming framework, implemented in GAMS, and solved by using the CPLEX solver to ensure optimal operation and minimum computational burden. This chapter provides a broad guideline of how the integrated systems operate. © 2025 Elsevier B.V., All rights reserved.

Abstract

Abstract

Abstract
This paper presents a fast and accurate optimization technique for optimal power flow (OPF) that can be conveniently applied to transmission and distribution systems. The method is based on the branch flow and DC optimal power flow (DCOPF) models. As the branch flow model is independent of the bus voltage angle, the model needs further development to enable use in meshed transmission systems. Thus, this paper adds the bus voltage angle constraint as a key constraint to the branch flow model so that the voltage angle can also be used in the power flow model in addition to the voltage magnitude control. The problem is based on second-order programming and modeled as a quadratically-constrained programming (QCP) problem solved using the CPLEX solver in GAMS. The functionality of the proposed model is tested utilizing four standard distribution systems, three transmission systems, a combined transmission-distribution network. The studied distribution systems include the 33-bus, 69-bus, 118-bus distribution (118-D) test systems, and 730-bus distribution system (730-D). Additionally, the studied transmission systems include 9-bus, 30-bus, and 118-bus transmission (118-T) test systems. The combined transmission-distribution system included the 9-bus transmission system with three connected distribution systems. The simulation results obtained from the developed technique are compared to those obtained from a conventional optimal flow model. The power losses and the absolute error of the solution are used as the two metrics to compare the methods' performance for distribution networks. The absolute error of the solution derived from the proposed hybrid OPF compared to MATPOWER for the 33-bus system is 0.00198 %. For the 69-bus system, the error is 0.00044 %. In addition, for the 118-D and 730-D systems, the absolute errors are 0.0026 %, and 0.05 %, respectively. For the transmission network, the operating costs and the solution absolute error are the two metrics used for comparing the proposed hybrid OPF model and MATPOWER. The results indicate the superior performance of the hybrid OPF model to the Newton-Raphson method in MATPOWER in terms of operating cost. In this regard, cost reductions relative to values given by MATPOWER are 0.0005 %, 0.838 %, and 0.015 %, for the 9-bus, 30-bus, and 118-T systems, respectively. The simulation studies demonstrate the performance of the presented branch flow-based model in solving the OPF problem with accurate results.