The intermittent nature of wind energy raised multiple challenges to the power systems and is the biggest challenge to declare wind energy a reliable source. One solution to overcome this problem is wind energy forecasting. A precise forecast can help to develop appropriate incentives and well-functioning electric markets. The paper presents a comprehensive review of existing research and current developments in deterministic wind speed and power forecasting. Firstly, we categorize wind forecasting methods into four broader classifications: input data, time-scales, power output, and forecasting method. Secondly, the performance of wind speed and power forecasting models is evaluated based on 634 accuracy tests reported in twenty-eight published articles covering fifty locations of ten countries. From the analysis, the most significant errors were witnessed for the physical models, whereas the hybrid models showed the best performance. Although, the physical models have a large normalized root mean square error values but have small volatility. The hybrid models perform best for every time horizon. However, the errors almost doubled at the medium-term forecast from its initial value. The statistical models showed better performance than artificial intelligence models only in the very short term forecast. Overall, we observed the increase in the performance of forecasting models during the last ten years such that the normalized mean absolute error and normalized root mean square error values reduced to about half the initial values.

Markov chains (MC) are statistical models used to predict very short to short-term wind speed accurately. Such models are generally trained with a single moving window. However, wind speed time series do not possess an equal length of behavior for all horizons. Therefore, a single moving window can provide reasonable estimates but is not an optimal choice. In this study, a forecasting model is proposed that integrates MCs with an adjusting dynamic moving window. The model selects the optimal size of the window based on a similar approach to the leave-one-out method. The traditional model is further optimized by introducing a self-adaptive state categorization algorithm. Instead of synthetically generating time series, the modified model directly predicts one-step ahead wind speed. Initial results indicate that adjusting the moving window MC prediction model improved the forecasting performance of a single moving window approach by 50%. Based on preliminary findings, a novel hybrid model is proposed integrating maximal overlap discrete wavelet transform (MODWT) with auto-regressive integrated moving average (ARIMA) and adjusting moving window MC. It is evident from the literature that MC models are suitable for predicting residual sequences. However, MCs were not considered as a primary forecasting model for the decomposition-based hybrid approach in any wind forecasting studies. The improvement of the novel model is, on average, 55% for single deep learning models and 30% for decomposition-based hybrid models.

Human gut microbiota can provide comprehensive information about the health of a host but the tools to collect microbiome samples are not currently available. A standalone wireless robotic capsule that has been developed in this study, collects the microbiota both from lumen (capsule surrounding) and intestinal wall (mucosa layer) for the first time. First, a two-way shape memory alloy (SMA) spring actuation system was developed by tackling the high-drain current requirement of SMAs. The actuator can produce up to 800 mN force that was sufficient to collect samples. Second, successful encapsulation of the collected sample to avoid contamination was realised by testing 3 main sealing materials. Third, the robotic capsule was tested in a gut simulator that mimics in-vivo environment to ensure successful and safe travel of the capsule along the gastrointestinal tract. Finally an in vitro experimental setup that keeps an intestine alive for 6 hours was used to optimise the sample collection. The capsule collected 128 $\mu$ L and 107 $\mu$ L samples (which are sufficient quantities for microbiome analysis) from duodenual and ileal tissues of a sheep. The proposed robotic capsule has a potential to become a vital apparatus for clinicians to sample human and animal gut in the future.

Using robotic capsules for assessing gut health has been an emerging field since the early 2000s with researchers attempting to perform diagnosis, monitoring, and therapeutic functions inside the gut. The knowledge of peristaltic forces inside the intestine are crucial for designing the actuation mechanism of robotic capsules, however, the impact of peristalsis on a capsule has not yet been quantified. In this work, an analytical model is presented to study the peristaltic movement of the small intestine. For the first time, finite element simulations were conducted in COMSOL multiphysics to generate intestinal peristaltic forces, and analyze their impact on a robotic capsule. A capsule prototype (30 × φ12 mm) was developed to measure the peristaltic forces from living intestinal tissue, while an embedded system was used simultaneously to record the live data from the capsule-intestine interaction. In in vitro experiments, the intestine applied an average axial force of 226 mN and contraction cycles of 9 times/min, while the capsule prototype experienced maximum radial force of 180 mN. A specialized in vitro setup is developed to keep fresh ex vivo intestine samples alive for up to 6 h, while the capsule prototype measured the intestinal forces from the living tissue. This in vitro experimental setup provided an excellent model for the in vivo environment in terms of generating peristaltic movements, hence this force analysis will help in developing efficient prototypes for locomotion, anchoring, localization, biopsy, drug delivery, and sampling mechanisms for robotic capsules.

Recent studies suggest that human gut microbiota can act as a bio-marker for human health. Also, it can function as a potential tool to understand stress and anxiety. However, the conventional tools have limitations acquiring samples of gut microbiota without contamination. In this work, an untethered robotic capsule prototype is developed that can actively collect the microbiota from the mucosa layer of the small intestine for the first time with the potential to avoid the upstream and downstream contamination. An analytical model for quantifying the peristaltic forces and developing two-way shape memory alloy spring actuator is presented. For the first time, a novel two-way shape memory alloy spring actuator (5 mm x $\phi ~4$ mm) is used to perform the sampling inside the gut. The spring actuator can apply 675 mN force, which is sufficient to perform in vivo sampling. A specialised experimental setup that can keep the freshly dissected intestine alive for 6 hours is utilised to test the robotic capsule. The robotic capsule prototype has collected an average of $200~\mu L$ and $112~\mu L$ sample from living pig duodenal and ileal tissues respectively i.e. in the presence of peristaltic forces. The robotic capsule was also tested on intestine of other species including cow and sheep and collected an average of $160~\mu L$ and $185~\mu L$ of content respectively from the living post-mortem tissues. The collected sample size for all the species is feasible to analyse the microbiota through next generation sequencing techniques. The experimental setup is a reliable proxy to in-vivo behaviour and the robotic capsule experimental result is promising in terms of in situ collection of microbiota.