Open Access

The transition to a fully renewable energy grid requires better forecasting of demand at the low-voltage level to increase efficiency and ensure reliable control. However, high fluctuations and increasing electrification cause huge forecast variability, not reflected in traditional point estimates. Probabilistic load forecasts take uncertainties into account and thus allow more informed decision-making for the planning and operation of low-carbon energy systems. We propose an approach for flexible conditional density forecasting of short-term load based on Bernstein polynomial normalizing flows, where a neural network controls the parameters of the flow. In an empirical study with 3639 smart meter customers, our density predictions for 24h-ahead load forecasting compare favorably against Gaussian and Gaussian mixture densities. Furthermore, they outperform a non-parametric approach based on the pinball loss, especially in low-data scenarios.

The World Health Organization recognizes physical activity as an influencing domain on quality of life. Monitoring, evaluating, and supervising it by wearable devices can contribute to the early detection and progress assessment of diseases such as Alzheimer’s, rehabilitation, and exercises in telehealth, as well as abrupt events such as a fall. In this work, we use a non-invasive and non-intrusive flexible wearable device for 3D spine pose measurement to monitor and classify physical activity. We develop a comprehensive protocol that consists of 10 indoor, 4 outdoor, and 8 transition states activities in three categories of static, dynamic, and transition in order to evaluate the applicability of the flexible wearable device in human activity recognition. We implement and compare the performance of three neural networks: long short-term memory (LSTM), convolutional neural network (CNN), and a hybrid model (CNN-LSTM). For ground truth, we use an accelerometer and strips data. LSTM reached an overall classification accuracy of 98% for all activities. The CNN model with accelerometer data delivered better performance in lying down (100%), static (standing = 82%, sitting = 75%), and dynamic (walking = 100%, running = 100%) positions. Data fusion improved the outputs in standing (92%) and sitting (94%), while LSTM with the strips data yielded a better performance in bending-related activities (bending forward = 49%, bending backward = 88%, bending right = 92%, and bending left = 100%), the combination of data fusion and principle components analysis further strengthened the output (bending forward = 100%, bending backward = 89%, bending right = 100%, and bending left = 100%). Moreover, the LSTM model detected the first transition state that is similar to fall with the accuracy of 84%. The results show that the wearable device can be used in a daily routine for activity monitoring, recognition, and exercise supervision, but still needs further improvement for fall detection.

Digital federated platforms and data cooperatives for secure, trusted and sovereign data exchange will play a central role in the construction industry of the future. With the help of platforms, cooperatives and their novel value creation, the digital transformation and the degree of organization of the construction value chain can be taken to a new level of collaboration. The goal of this research project was to develop an experimental prototype for a federated innovation data platform along with a suitable exemplary use case. The prototype is to serve the construction industry as a demonstrator for further developments and form the basis for an innovation platform. It exemplifies how an overall concept is concretely implemented along one or more use cases that address high-priority industry pain points. This concept will create a blueprint and a framework for further developments, which will then be further established in the market. The research project illuminates the perspective of various governance innovations to increase industry collaboration, productivity and capital project performance and transparency as well as the overall potential of possible platform business models. However, a comprehensive expert survey revealed that there are considerable obstacles to trust-based data exchange between the key stakeholders in the industry value network. The obstacles to cooperation are predominantly not of a technical nature but rather of a competitive, predominantly trust-related nature. To overcome these obstacles and create a pre-competitive space of trust, the authors therefore propose the governance structure of a data cooperative model, which is discussed in detail in this paper.

In tomato drying, degradation in final quality may occur based on the drying method used and predrying preparation. Hence, this research was conducted to evaluate the effect of different predrying treatments on physicochemical quality and drying kinetics of twin-layer-solar-tunnel-dried tomato slices. During the experimental work, tomato slices of var. Galilea were used. As predrying treatments, 0.5% calcium chloride (CaCl2), 0.5% ascorbic acid (C6H8O6), 0.5% citric acid (C6H8O7), and 0.5% sodium chloride (NaCl) were used. The tomato samples were sliced to 5 mm thickness, socked in the pretreatments for ten minutes, and dried in a twin layer solar tunnel dryer under the weather conditions of Jimma, Ethiopia. Untreated samples were used as control. The moisture losses from the samples were monitored by weighing samples at 2 h interval from each treatment. SAS statistical software version 9.2 was used for analyzing data on the physicochemical quality of tomato slices in CRD with three replications. From the experimental result, it was observed that dried tomato slices pretreated with 0.5% ascorbic acid gave the best retention of vitamin C and total phenolic content with a high sugar/acid ratio. Better retention of lycopene and fast drying were observed in dried tomato slices pretreated with 0.5% sodium chloride, and pretreating tomatoes with 0.5% citric acid resulted in better color values than the other treatments. Compared to the control, pretreating significantly preserved the overall quality of dried tomato slices and increased the moisture removal rate in the twin layer solar tunnel dryer.

The present contribution proposes a novel method for the indirect measurement of the ground reaction forces (GRF) induced by a pedestrian during walking on a vibrating structure. Its main idea is to formulate and solve an inverse problem in the time domain with the aim of finding the optimal time dependent moving point force describing the GRF of a pedestrian (input data), which minimizes the difference between a set of computed and a set of measured structural responses (output data). The solution of the inverse problem is addressed by means of the gradient-based trust region optimization strategy. The moving force identification process uses output data from a set of acceleration and displacement time histories recorded at different locations on the structure. The practicability and the accuracy of the proposed GRF identification method is firstly evaluated using simulated measurements, which revealed a high accuracy, robustness and stability of the results in relation to high noise levels. Subsequently, a comprehensive experimental validation process using real measurement data recorded on the HUMVIB experimental footbridge on the campus of the Technical University of Darmstadt (Germany) was carried out. Besides the conventional sensors for the acquisition of structural responses, an array of biomechanical force plates as well as classical load cells at the supports were used for measurement reference GRFs needed in the experimental validation process. The results show that the proposed method delivers a very accurate estimation of the GRF induced by a subject during walking on the experimental structure.