Open Access

In Ghana, unreliable public grid infrastructure greatly impacts rural healthcare, where diesel generators are commonly used despite their high financial and environmental costs. Photovoltaic (PV)-hybrid systems offer a sustainable alternative, but require robust, predictive control strategies to ensure reliability. This study proposes a sector-specific Model Predictive Control (MPC) approach, integrating advanced load and meteorological forecasting for optimal energy dispatch. The methodology includes a long-short-term memory (LSTM)-based load forecasting model with probabilistic Monte Carlo dropout, a customized Numerical Weather Prediction (NWP) model based on the Weather Research and Forecasting (WRF) framework, and deep learning-based All-Sky Imager (ASI) nowcasting to improve short-term solar predictions. By combining these forecasting methods into a seamless prediction framework, the proposed MPC optimizes system performance while reducing reliance on fossil fuels. This study benchmarks the MPC against a traditional rule-based dispatch system, using data collected from a rural health facility in Kologo, Ghana. Results demonstrate that predictive control greatly reduces both economic and ecological costs. Compared to rule-based dispatch, diesel generator operation and fuel consumption are reduced by up to 61.62% and 47.17%, leading to economical and ecological cost savings of up to 20.7% and 31.78%. Additionally, system reliability improves, with battery depletion events during blackouts decreasing by up to 99.42%, while wear and tear on the diesel generator and battery are reduced by up to 54.93% and 37.34%, respectively. Furthermore, hyperparameter tuning enhances MPC performance, introducing further optimization potential. These findings highlight the effectiveness of predictive control in improving energy resilience for critical healthcare applications in rural settings.

The shift towards renewable energy sources has heightened the interest in solar photovoltaic (SPV) systems, particularly in grid-connected configurations, to enhance energy security and reduce carbon emissions. Grid-tied SPVs face power quality challenges when specific grid codes are compromised. This study investigates and upgrades an integrated 90 kWp solar plant within a distribution network, leveraging data from Ghana's Energy Self-Sufficiency for Health Facilities (EnerSHelF) project. The research explores four scenarios for SPV placement optimization using dynamic programming and the Conditional New Adaptive Foraging Tree Squirrel Search Algorithm (CNAFTSSA). A Python-based simulation identifies three scenarios, high load nodes, voltage drop nodes, and system loss nodes, as the points for placing PV for better performance. The analysis revealed 85 %, 82.88 %, and 100 % optimal SPV penetration levels for placing the SPV at high load, voltage drop, and loss nodes. System active power losses were reduced by 72.97 %, 71.52 %, and 70.15 %, and reactive power losses by 73.12 %, 71.86 %, and 68.11 %, respectively, by placing the SPV at the above three categories of nodes. The fourth scenario applies to CNAFTSSA, achieving 100 % SPV penetration and reducing active and reactive power losses by 72.33 % and 72.55 %, respectively. This approach optimizes the voltage regulation (VR) from 24.92 % to 4.16 %, outperforming the VR of PV placement at high load nodes, voltage drop nodes, and loss nodes, where the voltage regulations are 5.25 %, 9.36 %, and 9.64 %, respectively. The novel CNAFTSSA for optimal SPV placement demonstrates its effectiveness in achieving higher penetration levels and improving system losses and VR. The findings highlight the effectiveness of strategic SPV placement and offer a comprehensive methodology that can be adapted for similar power distribution systems.

Accurate global horizontal irradiance (GHI) forecasting is critical for integrating solar energy into the power grid and operating solar power plants. The Weather Research and Forecasting model with its solar radiation extension (WRF-Solar) has been used to forecast solar irradiance in different regions around the world. However, the application of the WRF-Solar model to the prediction of GHI in West Africa, particularly Ghana, has not yet been investigated. The aim of this study is to evaluate the performance of the WRF-Solar model for predicting GHI in Ghana, focusing on three automatic weather stations (Akwatia, Kumasi and Kologo) for the year 2021. We used two one-way nested domains (D1 = 15 km and D2 = 3 km) to investigate the ability of the fully coupled WRF-Solar model to forecast GHI up to 72-hour ahead under different atmospheric conditions. The initial and lateral boundary conditions were taken from the ECMWF high-resolution operational forecasts. Our findings reveal that the WRF-Solar model performs better under clear skies than cloudy skies. Under clear skies, Kologo performed best in predicting 72-hour GHI, with a first day nRMSE of 9.62 %. However, forecasting GHI under cloudy skies at all three sites had significant uncertainties. Additionally, WRF-Solar model is able to reproduce the observed GHI diurnal cycle under high AOD conditions in most of the selected days. This study enhances the understanding of the WRF-Solar model’s capabilities and limitations for GHI forecasting in West Africa, particularly in Ghana. The findings provide valuable information for stakeholders involved in solar energy generation and grid integration towards optimized management in the region.

This work proposes a novel approach for probabilistic end-to-end all-sky imager-based nowcasting with horizons of up to 30 min using an ImageNet pre-trained deep neural network. The method involves a two-stage approach. First, a backbone model is trained to estimate the irradiance from all-sky imager (ASI) images. The model is then extended and retrained on image and parameter sequences for forecasting. An open access data set is used for training and evaluation. We investigated the impact of simultaneously considering global horizontal (GHI), direct normal (DNI), and diffuse horizontal irradiance (DHI) on training time and forecast performance as well as the effect of adding parameters describing the irradiance variability proposed in the literature. The backbone model estimates current GHI with an RMSE and MAE of 58.06 and 29.33 W m−2, respectively. When extended for forecasting, the model achieves an overall positive skill score reaching 18.6 % compared to a smart persistence forecast. Minor modifications to the deterministic backbone and forecasting models enables the architecture to output an asymmetrical probability distribution and reduces training time while leading to similar errors for the backbone models. Investigating the impact of variability parameters shows that they reduce training time but have no significant impact on the GHI forecasting performance for both deterministic and probabilistic forecasting while simultaneously forecasting GHI, DNI, and DHI reduces the forecast performance.

Solar photovoltaic power output is modulated by atmospheric aerosols and clouds and thus contains valuable information on the optical properties of the atmosphere. As a ground-based data source with high spatiotemporal resolution it has great potential to complement other ground-based solar irradiance measurements as well as those of weather models and satellites, thus leading to an improved characterisation of global horizontal irradiance. In this work several algorithms are presented that can retrieve global tilted and horizontal irradiance and atmospheric optical properties from solar photovoltaic data and/or pyranometer measurements. The method is tested on data from two measurement campaigns that took place in the Allgäu region in Germany in autumn 2018 and summer 2019, and the results are compared with local pyranometer measurements as well as satellite and weather model data. Using power data measured at 1 Hz and averaged to 1 min resolution along with a non-linear photovoltaic module temperature model, global horizontal irradiance is extracted with a mean bias error compared to concurrent pyranometer measurements of 5.79 W m−2 (7.35 W m−2) under clear (cloudy) skies, averaged over the two campaigns, whereas for the retrieval using coarser 15 min power data with a linear temperature model the mean bias error is 5.88 and 41.87 W m−2 under clear and cloudy skies, respectively. During completely overcast periods the cloud optical depth is extracted from photovoltaic power using a lookup table method based on a 1D radiative transfer simulation, and the results are compared to both satellite retrievals and data from the Consortium for Small-scale Modelling (COSMO) weather model. Potential applications of this approach for extracting cloud optical properties are discussed, as well as certain limitations, such as the representation of 3D radiative effects that occur under broken-cloud conditions. In principle this method could provide an unprecedented amount of ground-based data on both irradiance and optical properties of the atmosphere, as long as the required photovoltaic power data are available and properly pre-screened to remove unwanted artefacts in the signal. Possible solutions to this problem are discussed in the context of future work.

This dataset contains all the relevant data for the algorithms described in the paper "Irradiance and cloud optical properties from solar photovoltaic systems", which were developed within the framework of the MetPVNet project.

This  repository contains the data to reproduce the analysis conducted in the publication:  Witthuhn, J., Hünerbein, A., Filipitsch, F., Wacker, S., Meilinger, S., and Deneke, H., Aerosol properties and aerosol-radiation interactions in clear sky conditions over Germany.

Intention: Within the research project EnerSHelF (Energy-Self-Sufficiency for Health Facilities in Ghana), i. a. energy-meteorological and load-related measurement data are collected, for which an overview of the availability is to be presented on a poster. Context: In Ghana, the total electricity consumed has almost doubled between 2008 and 2018 according to the Energy Commission of Ghana. This goes along with an unstable power grid, resulting in power outages whenever electricity consumption peaks. The blackouts called "dumsor" in Ghana, pose a severe burden to the healthcare sector. Innovative solutions are needed to reduce greenhouse gas emissions and improve energy and health access.

The accurate forecasting of solar radiation plays an important role for predictive control applications for energy systems with a high share of photovoltaic (PV) energy. Especially off-grid microgrid applications using predictive control applications can benefit from forecasts with a high temporal resolution to address sudden fluctuations of PV-power. However, cloud formation processes and movements are subject to ongoing research. For now-casting applications, all-sky-imagers (ASI) are used to offer an appropriate forecasting for aforementioned application. Recent research aims to achieve these forecasts via deep learning approaches, either as an image segmentation task to generate a DNI forecast through a cloud vectoring approach to translate the DNI to a GHI with ground-based measurement (Fabel et al., 2022; Nouri et al., 2021), or as an end-to-end regression task to generate a GHI forecast directly from the images (Paletta et al., 2021; Yang et al., 2021). While end-to-end regression might be the more attractive approach for off-grid scenarios, literature reports increased performance compared to smart-persistence but do not show satisfactory forecasting patterns (Paletta et al., 2021). This work takes a step back and investigates the possibility to translate ASI-images to current GHI to deploy the neural network as a feature extractor. An ImageNet pre-trained deep learning model is used to achieve such translation on an openly available dataset by the University of California San Diego (Pedro et al., 2019). The images and measurements were collected in Folsom, California. Results show that the neural network can successfully translate ASI-images to GHI for a variety of cloud situations without the need of any external variables. Extending the neural network to a forecasting task also shows promising forecasting patterns, which shows that the neural network extracts both temporal and momentarily features within the images to generate GHI forecasts.

An aircraft plume model has been developed on the basis of two coupled trajectory box models. Two boxes, one for plume and one for background conditions, are coupled by means of a mixing parameterization based on turbulence theory. The model considers comprehensive gas phase chemistry for the tropopause region including acetone, ethane and their oxidation products. Heterogeneous halogen, N2O5 and HOx chemistry on various types of background and aircraft-induced aerosols (liquid and ice) is considered, using state-of-the-art solubility dependent uptake coefficients for liquid phase reactions. The microphysical scheme allows for coagulation, gas-diffusive particle growth and evaporation, so that the particle development from 1s after emission to several days can be simulated. Model results are shown, studying emissions into the upper troposphere as well as into the lowermost stratosphere for contrail and non-contrail conditions. We show the microphysical and chemical evolution of spreading plumes and use the concept of mean plume encounter time, tl, to define effective emission and perturbation indices (EEIs and EPIs) for the North Atlantic Flight Corridor (NAFC) showing EEI(NOy) and EPI(O3) for various background conditions, such as relative humidity, local time of emission, and seasonal variations. Our results show a high sensitivity of EEI and EPIs on the exact conditions under which emissions take place. The difference of EEIs with and without considering plume processes indicates that these processes cannot be neglected.

For the winter 1999/2000 transport of air masses out of the vortex to mid-latitudes and ozone destruction inside and outside the northern polar vortex is studied to quantify the impact of earlier winter (before March) polar ozone destruction on mid-latitude ozone. Nearly 112 000 trajectories are started on 1 December 1999 on 6 different potential temperature levels between 500–600 K and for a subset of these trajectories photo-chemical box-model calculations are performed. We linked a decline of −0.9% of mid-latitude ozone in this layer occurring in January and February 2000 to ozone destruction inside the vortex and successive transport of these air masses to mid-latitudes. Further, the impact of denitrification, PSC-occurrence and anthropogenic chlorine loading on future stratospheric ozone is determined by applying various scenarios. Lower stratospheric temperatures and denitrification were found to play the most important role in the future evolution of polar ozone depletion.

From a synopsis of field, laboratory and model studies at T>205 K as well as from the field experiments POLSTAR at T<205 K we derive a general picture of the partitioning of nitric acid (HNO3) in cirrus clouds and a new hypothesis on the uptake of HNO3 on ice particles: A substantial part of nitric acid remains in the gas phase under cirrus cloud conditions. The HNO3 removed from the gas phase is distributed between interstitial aerosol and ice particles in dependence on the temperature and ice surface, respectively. In cold cirrus clouds with small ice surface areas (T <205 K) the partitioning is strongly in favour of interstitial ternary solution particles while in warmer cirrus clouds with large ice surface areas the uptake on ice dominates. Consequently, denitrification via sedimenting ice particles may occur only in the -more frequently occurring- warm cirrus clouds The HNO3 coverage on ice is found to be different for ice particles and ice films. On ice films the coverage can increase with decreasing temperature from about 0.1 to 0.8 monolayer, while that on ice particles is found to decrease with temperature and PHNO3 from 0.1 to 0.001 monolayer. An HNO3 uptake behaviour following dissociative Langmuir isotherms where the coverage decreases for descending temperatures may explain the observations for ice particles From a comparison of the HNO3 measurements with model calculations it is found that (i) the global model of Lawrence and Crutzen (1998) overestimates the HNO3 partitioning in favour of the ice particles (ii) the Langmuir surface chemistry model of Tabazadeh et al. (1999) overestimates HNO3 coverages for temperatures ≤210 K More appropriate coverages are calculated when implementing in that model a temperature dependent function for the adsorption free energy (ΔGads (T)), which is empirically derived from the coverage measurements.

We examine the effect of nanometer-sized aircraft-induced aqueous sulfuric acid (H2SO4/H2O) particles on atmospheric ozone as a function of temperature. Our calculations are based on a previously derived parameterization for the regional-scale perturbations of the sulfate surface area density due to air traffic in the North Atlantic Flight Corridor (NAFC) and a chemical box model. We confirm large scale model results that at temperatures T>210 K additional ozone loss -- mainly caused by hydrolysis of BrONO2 and N2O5 -- scales in proportion with the aviation-produced increase of the background aerosol surface area. However, at lower temperatures (< 210 K) we isolate two effects which efficiently reduce the aircraft-induced perturbation: (1) background particles growth due to H2O and HNO3 uptake enhance scavenging losses of aviation-produced liquid particles and (2) the Kelvin effect efficiently limits chlorine activation on the small aircraft-induced droplets by reducing the solubility of chemically reacting species. These two effects lead to a substantial reduction of heterogeneous chemistry on aircraft-induced volatile aerosols under cold conditions. In contrast we find contrail ice particles to be potentially important for heterogeneous chlorine activation and reductions in ozone levels. These features have not been taken into consideration in previous global studies of the atmospheric impact of aviation. Therefore, to parameterize them in global chemistry and transport models, we propose the following parameterisation: scale the hydrolysis reactions by the aircraft-induced surface area increase, and neglect heterogeneous chlorine reactions on liquid plume particles but not on ice contrails and aircraft induced ice clouds.

This thesis contributes to a better understanding of the effect of heterogeneous chemistry on ozone in the tropopause region. As part of the German research project ALTO, it especially focuses on the impact of aircraft emissions on heterogeneous ozone chemistry in this region. This is an important question as ozone is a strong greenhouse gas, whose radiative effect, is strongest near the tropopause. In general, the treatment of heterogeneous processes on background and aviation-produced particles requires the consideration of processes ranging from nanometer to continental scale. For this reason the present modeling work includes a treatment of small scale processes as well as the development and subsequent application of parameterisations. Three numerical trajectory box models considering highly detailed microphysical and chemical processes have been developed: (a) an aircraft plume model including coagulation, chemistry and plume dilution, (b) a particle-size resolved microphysical box model and, (c) a comprehensive photo-chemical box model.