Browsing by Subject "Time series analysis"

Now showing 1 - 2 of 2

Experimentelle Entwicklung einer modellbasierten prädiktiven Regelung für den flexiblen Betrieb von Biogasanlagen
(2023) Dittmer, Celina; Lemmer, Andreas
The transformation of the energy system requires controllable producers due to increasingly decentralised, fluctuating electricity generation from wind turbines and photovoltaics. Biogas plants can make a substantial contribution here by making plant operation more flexible and thus providing electricity as needed. Technical adjustments, such as the expansion of gas storage capacities and CHP output, can compensate for short-term fluctuations. However, in order to be able to shift the potential of electricity generation over longer periods of time, an adapted feed-in strategy is essential. The control of biogas production poses several challenges in practical implementation. First, the conversion of biomass into biogas is a complex process and must be considered individually for each biogas plant. Models developed so far use parameters for all characteristic process phases and influencing variables in order to be able to model anaerobic digestion. In contrast, biogas plants are often only rudimentarily equipped with measurement technology, so that corresponding parameters are not available. In this work, a model-predictive control of biogas plant operation was developed to enable demand-driven electricity generation. The aim was to develop models that are particularly well suited for practical use. Thus, for the first time, a successful application on almost all biogas plants could be possible without or with only minor adaptations to the existing measurement technology. All studies carried out in this thesis are based on a real-world laboratory, the "Unterer Lindenhof". This includes a practical biogas plant as well as an electrical consumption corresponding to that of a village with about 125 inhabitants. In a first step, forecasting models were evaluated to predict the electricity demand of the real-world laboratory over 48 hours in advance. Four models from the field of time series analysis were examined, one TBATS and three different ARIMA models. In an evaluation of 366 forecasts each, all four models performed sufficiently well to provide a set point for biogas plant operation, with average MAPE values of 13-16 %. Further investigations showed that forecasts can also be carried out over a period of up to 14 days without significant losses in forecast quality. In a further step, a model was developed to simulate biogas production. This is also based on time series analysis, or more precisely on a regression model. Thus, it differs significantly from previous developments in this field, which are mostly based on the complex ADM1. It turns out to be very advantageous that the developed simulation model uses as input parameters only historical data of the last four weeks of biogas production and the amount of solid substrates fed in, without considering their composition. The simulation of biogas production over 48 hours in advance is based on correlations resulting from these two data sets. An evaluation of the model over 366 simulations resulted in an average MAPE of 14-18 %. Data from both digesters of the biogas plant were used, which can be considered as independent systems, demonstrating the adaptability of the model. In a third step, the feeding schedule was developed for demand-based biogas production. For each 48 hours in advance, 1500 randomised feeding schedules were calculated. Some constraints were imposed, such as the maximum amount of substrate that is technically possible in the biogas plant. The biogas production expected from the feeding schedules could be calculated using the simulation model. By comparing the simulation with the desired biogas demand profile, the simulation with the least deviations could be determined and the appropriate feeding plan selected and implemented. The entire model predictive control system was used and thoroughly tested in a field trial at the real-world laboratory "Unterer Lindenhof". Over a period of 36 days, an average MAPE of less than 20 % was achieved in comparison between the real biogas production and the desired biogas demand. During the test period, the biogas demand was derived from the predicted electricity demand of the real-world laboratory. The investigations carried out show that the model-predictive control system developed enables demand-oriented electricity generation on full-scale and that, due to the models being very close to practice for the first time, adaptation to almost all biogas plants is possible.
Modeling and spatiotemporal mapping of water quality through remote sensing techniques: A case study of the Hassan Addakhil dam
(2021) El Ouali, Anas; El Hafyani, Mohammed; Roubil, Allal; Lahrach, Abderrahim; Essahlaoui, Ali; Hamid, Fatima Ezzahra; Muzirafuti, Anselme; Paraforos, Dimitrios S.; Lanza, Stefania; Randazzo, Giovanni
With its high water potential, the Ziz basin is one of the most important basins in Morocco. This paper aims to develop a methodology for spatiotemporal monitoring of the water quality of the Hassan Addakhil dam using remote sensing techniques combined with a modeling approach. Firstly, several models were established for the different water quality parameters (nitrate, dissolved oxygen and chlorophyll a) by combining field and satellite data. In a second step, the calibration and validation of the selected models were performed based on the following statistical parameters: compliance index R2, the root mean square error and p-value. Finally, the satellite data were used to carry out spatiotemporal monitoring of the water quality. The field results show excellent quality for most of the samples. In terms of the modeling approach, the selected models for the three parameters (nitrate, dissolved oxygen and chlorophyll a) have shown a good correlation between the measured and estimated values with compliance index values of 0.62, 0.56 and 0.58 and root mean square error values of 0.16 mg/L, 0.65 mg/L and 0.07 µg/L for nitrate, dissolved oxygen and chlorophyll a, respectively. After the calibration, the validation and the selection of the models, the spatiotemporal variation of water quality was determined thanks to the multitemporal satellite data. The results show that this approach is an effective and valid methodology for the modeling and spatiotemporal mapping of water quality in the reservoir of the Hassan Addakhil dam. It can also provide valuable support for decision-makers in water quality monitoring as it can be applied to other regions with similar conditions.