The release of potassium during biomass combustion leads to several problems as the emissions of particle matter or formation of deposits. K release is mainly described in literature in a qualitative way and this work aims to develop a simplified model to quantitatively describe it at different stages. The proposed model has 4 reactions and 5 solid species, describing K release in 3 steps; during pyrolysis, KCl evaporation and carbonate dissociation. This release model is coupled into a single particle model and successfully validated with experiments conducted in a single particle reactor with spruce, straw and Miscanthus pellets at different temperatures. The model employs same kinetic parameters for the reactions in all cases, while different product compositions of the reactions are employed for each fuel, which is attributed to differences in composition. The proposed model correctly predicts the online release at different stages during conversion as well as the final release for each case.

The increased biomass utilization leads to the need of an efficient and flexible usage of available sources. Therefore, it is necessary to combust low-cost biogenic residues, which inherently have higher nitrogen contents that lead to increased NOx emissions. In order to tackle this issue a new combustion technology with double air staging and flue gas recirculation is under development. The technology also features an increased fuel bed height and very low oxygen concentrations in the fuel bed to reduce fuel bed temperatures. This work focuses on the CFD simulation of the formation and reduction of NOx emissions of in a small scale boiler (35 kWth). Compared to previously applied models, major modification concerning the heat and mass transfer in the fuel bed as well as the subsequent conversion in the freeboard were made. The fuel bed is modelled via representative fuel particles with a Lagrangian approach and a thermally thick particle model considering intra-particle
gradients. Due to the increased fuel bed height and the relatively low oxygen concentration the formation and cracking of tars has to be considered in the simulation. This heavily influences the formation and reduction of NOx and its precursors. The fuel bound nitrogen is released via the particle model in the form of NO during char burnout and via a lumped tar species during pyrolysis. The cracking of the lumped tar species is modelled via two global gas phase reactions that releases the NOx precursors NH3 and HCN. The cracking reactions are added to a skeletal reaction mechanism with 28 species and 102 reactions that includes the fate of the N species. The simulation results are compared to experimental data from test runs with spruce wood chips and Miscanthus pellets as fuels. The comparison showed good agreement for the test runs with wood chips, where the temperature distribution inside the fuel bed and the released species above the fuel bed were predicted well. The test runs with Miscanthus showed a greater deviation between the measured and simulated values. For both fuels the NOx reduction that was experimentally observed in the secondary combustion zone could not be predicted with reasonable agreement. Therefore, it is necessary to further investigate the cracking of the tars and the subsequent formation of the NOx precursors. The presented work forms the basis for further improvements of the numerical models and subsequently the optimization of the new technology.

During transportation and storage of wood pellets various gases are formed leading to toxic atmosphere. Various influencing factors and measures reducing off-gassing have already been investigated. The present study aims at applying an antioxidant, acetylsalicylic acid (ASA), to reduce off-gassing from wood pellets by lowering wood extractives oxidation. Therefore, acetylsalicylic acid was applied in industrial and laboratory pelletizing processes. Pine and spruce sawdust (ratio 1:1) were pelletized with adding 0-0.8% (m/m) ASA. Glass flasks measurements confirmed off-gassing reduction by adding ASA for all wood pellets investigated.The biggest effect was achieved by adding 0.8% (m/m) ASA in the industrial pelletizing experiments where the emission of volatile organic compounds (VOC_tot) was reduced by 82% and a reduction of carbon monoxide (CO) and carbon dioxide (CO₂) emissions by 70% and 51%, respectively, could be achieved. Even an addition of 0.05% (m/m) ASA led to off-gassing reduction by >10%. A six week storage experiment to investigate the long-term effectivity of ASA addition revealed, that antioxidant addition was effective in reducing CO-, CO₂- and VOC_tot-release, especially during the first four weeks of the storage experiment, after which time the relative reduction effect was significantly decreased.

This study focuses on the determination of alkali release from wood and straw pellets during combustion. The aim is to expand the knowledge on the K and Na release behaviour and to adopt chemiluminescence-based sensors for online monitoring of alkali detection which can be applied for the prevention of fouling formation in low quality biomass combustion plants. Flame emission spectrometry (FES) was used for optical detection of chemiluminescence spectra of K and Na using optical bandpass filters mounted on an ICCD (Intensified Charge Coupled Device) camera. FES data were verified by additional experiments with a single particle reactor (SPR) coupled with an inductively coupled plasma mass spectrometer (ICP-MS). Using both techniques, the release profiles of K and Na during a single pellet combustion at 1000 °C were determined and obtained K* and Na* emission intensities directly correlated with the results from the ICP-MS. It was determined that the emission intensity of alkali radicals depends on alkali concentrations in the samples and K and Na radical emission intensities increase with increasing alkali amounts in the samples. The ICP-MS data revealed that the release of K and Na mainly takes place during the stage of devolatilization. During devolatilization, almost all potassium and sodium are released from wood samples, while only 65–90% of K and 74–90% of Na are released from straw samples. Based on the results, the flame emission spectroscopy technique is capable to fully detect released alkali metals in the gas phase during combustion and proves a possibility to use flame emission sensors for monitoring the release of alkali species from biomass during combustion processes.

Modern central heating systems with logwood boilers are comprised of the boiler, a buffer storage and solar thermal collectors. Conventional control strategies for these heating systems do not coordinate the utilization of all components. This can lead to a sub-optimal operation of the entire heating system resulting in a loss of efficiency and increased pollutant emissions. This contribution presents a control strategy which considers all components of the heating system including the user and forecasts for the solar yield and heat demand. It determines and carries out an optimal operating strategy that improves the user utility and maximizes the heating system efficiency while also ensuring a clean and efficient combustion. The control strategy continuously learns the user behavior and instructs the user when to refill the logwood boiler and how much fuel to use. The new control strategy was verified through test runs performed at an experimental setup consisting of a commercially available logwood boiler with a nominal capacity of 28 kW , two buffer storages with a capacity of 1.5 m³ each and a heating device with a thermal output of up to 12 kW simulating a solar thermal collector. During these test runs, the CO emissions were reduced 93.6 %by in the main combustion phase, 7.1 % more solar yield was utilized, the buffer losses were reduced by - 16.9 % and the overall efficiency was increased by 3.1 % . Thus, the application of this control strategy resulted in a significantly improved user utility and heating system efficiency.

Mixed Integer Linear Programming (MILP) optimization algorithms provide accurate and clear solutions for Microgrid and Distributed Energy Resources projects. Full-scale optimization approaches optimize all time-steps of data sets (e.g., 8760 time-step and higher resolutions), incurring extreme and unpredictable run-times, often prohibiting such approaches for effective Microgrid designs. To reduce run-times down-sampling approaches exist. Given that the literature evaluates the full-scale and down-sampling approaches only for limited numbers of case studies, there is a lack of a more comprehensive study involving multiple Microgrids. This paper closes this gap by comparing results and run-times of a full-scale 8760 h time-series MILP to a peak preserving day-type MILP for 13 real Microgrid projects. The day-type approach reduces the computational time between 85% and almost 100% (from 2 h computational time to less than 1 min). At the same time the day-type approach keeps the objective function (OF) differences below 1.5% for 77% of the Microgrids. The other cases show OF differences between 6% and 13%, which can be reduced to 1.5% or less by applying a two-stage hybrid approach that designs the Microgrid based on down-sampled data and then performs a full-scale dispatch algorithm. This two stage approach results in 20–99% run-time savings.

Energy systems have increased in complexity in the past years due to the everincreasing integration of intermittent renewable energy sources such as solar thermal or wind power. Modern energy systems comprise different energy domains such as electrical power, heating and cooling which renders their control even more challenging. Employing supervisory controllers, so-called energy management systems (EMSs), can help to handle this complexity and to ensure the energy-efficient and cost-efficient operation of the energy system. One promising approach are optimization-based EMS, which can for example be modelled as stochastic mixed-integer linear programmes (SMILP). Depending on the problem size and control horizon, obtaining solutions for these in real-time is a difficult task. The progressive hedging (PH) algorithm is a practical way for splitting a large problem into smaller subproblems and solving them iteratively, thus possibly reducing the solving time considerably. The idea of the PH algorithm is to aggregate the solutions of subproblems, where artificial costs have been added. These added costs enforce that the aggregated solutions become non-anticipative and
are updated in every iteration of the algorithm. The algorithm is relatively simple to implement in practice, re-using almost all of a possibly existing deterministic implementations and can be easily parallelized.
Although it has no convergence guarantees in the mixed-integer linear case, it can nevertheless be used as a good heuristic for SMILPs. Recent theoretical results shown that for applying augmented Lagrangian functions in the context of mixed-integer programmes, any norm proofs to be a valid penalty function. This is not true for squared norms, like the squared L 2 -norm that is used in the classical progressive hedging algorithm. Building on these theoretical results, the use of the L 1 and L-infinity-norm in the PH algorithm is investigated in this paper. In order to incorporate these into the algorithm an adapted multiplier update step is proposed. Additionally a heuristic extension of the aggregation step and an adaptive penalty parameter update scheme from the literature is investigated. The advantages of the proposed modifications are demonstrated by means of illustrative examples, with the application to SMILP-based EMS in mind.

Recently, researchers have begun to study hybrid approaches to Microgrid techno-economic planning, where a reduced model optimizes the DER selection and sizing is combined with a full model that optimizes operation and dispatch. Though providing significant computation time savings, these hybrid models are susceptible to infeasibilities, when the size of the DER is insufficient to meet the energy balance in the full model during macrogrid outages. In this work, a novel hybrid optimization framework is introduced, specifically designed for resilience to macrogrid outages. The framework solves the same optimization problem twice, where the second solution using full data is informed by the first solution using representative data to size and select DER. This framework includes a novel constraint on the state of charge for storage devices, which allows the representation of multiple repeated days of grid outage, despite a single 24-h profile being optimized in the representative model. Multiple approaches to the hybrid optimization are compared in terms of their computation time, optimality, and robustness against infeasibilities. Through a case study on three real Microgrid designs, we show that allowing optimizing the DER sizing in both stages of the hybrid design, dubbed minimum investment optimization (MIO), provides the greatest degree of optimality, guarantees robustness, and provides significant time savings over the benchmark optimization.

This work presents a methodology to perform the scale-up of a solid fuel furnace to a higher heat output with maintaining or improving the burn-out quality. As basis to derive the scale-up concept, an example of a 35 kW screw burner for biomass fuels is investigated. Based on the Pi-theorem, the relevant dimensionless parameters are derived and similarity rules for the scale-up are proposed as follows: As initial conditions, the height to diameter ratio of the combustion chamber, the mean Reynolds number in the combustion chamber and the mean square velocity through the combustion chamber shall be kept constant or in the case of the Reynolds number may also increase. Additionally the effective momentum flux ratio between the secondary air injected in the combustion chamber and the gases from the pyrolysis and gasification section also shall be kept constant to maintain the mixing conditions between combustible gases and secondary air. Finally the thermal surface load on the screw also shall be kept constant. The influence of different scale-up approaches on thermal surface load, gas velocity, pressure losses, Reynolds number and height-to-diameter ratio are compared and discussed and a scaling approach to increase the heat output from 35 kW to 150 kW is described. For a theoretical validation of the scale-up, CFD simulations are performed to investigate the predicted pollutant emissions and the pressure loss for the scaled 150 kW furnace.

A key factor for the further distribution of biomass boilers in modern energy systems is the capability of changing the applied feedstock during normal plant operation. This is only possible with the application of advanced control strategies that utilize knowledge about the state variables and varying fuel properties. However, neither the state variables nor the fuel properties are measurable during plant operation and, thus, need to be estimated. This contribution presents a method for the simultaneous real-time estimation of the state variables and the fuel properties in fixed-bed biomass boilers which is a novel approach in the field of biomass boilers. The method bases on an Extended Kalman Filter using a nonlinear dynamic model and measurement data from the combustion process. The estimated variables are the masses of dry fuel and water in the fuel bed as well as the fuel's bulk density, water content, chemical composition and lower heating value. The proposed method is easy to implement and requires moderate computational effort which increases the potential of its application at actual biomass boilers. The proposed method is verified with simulation studies and by test runs performed at a representative small-scale fixed-bed biomass boiler. The estimation results show a good agreement with the actual values, demonstrating that the proposed method is capable of accurately estimating the biomass boiler's state variables and simultaneously its fuel properties. For this reason, the presented method is a key technology to ensure the further distribution of biomass boilers in modern energy systems.

The present study aims to present a comprehensive characterization of the surface of ash-layered olivine bed particles from dual fluidized bed gasification. It is well known from operation experience at industrial gasification plants that the bed material is activated during operation concerning its positive influence on gasification reactions. This is due to the built up of ash layers on the bed material particles; however, the chemical mechanisms are not well understood yet. Olivine samples from long-term operation in an industrial-scale gasification plant were investigated in comparison to fresh unused olivine. Changes of the surface morphology due to Ca-enrichment showed a significant increase of their surface area. Furthermore, the Ca-enrichment on the ash layer surface was distinctively associated to CaO being present. The presence of CaO on the surface was proven by adsorption tests of carbon monoxide as model compound. The detailed characterization contributes to a deeper understanding of the surface properties of ash layers and forms the basis for further investigations into their influence on gasification reactions.

The viability of thermochemical energy storage for a given application is often determined by the reaction kinetics under process conditions. For high exergetic efficiency the process needs to operate in close proximity to the reaction equilibrium. Thus, accurate kinetic models that include the effect of the reaction equilibrium are required.

In the present work, different parametrization methods for the equilibrium term in the General Kinetic Equation are evaluated by modeling the kinetics of two reaction systems relevant for thermochemical energy storage (CaC2O4 and CuO) from experimental data. A non-parametric modeling method based on tensor decompositions is used that allows for a purely data driven assessment of different parametrization methods.

Our analysis shows that including a suitable equilibrium term is crucial. Omitting the equilibrium term when modeling formation reactions can lead to seemingly negative activation energies. Our tests also show that for formation reactions, the reaction rate decreases much faster towards the equilibrium than theory predicts. We present an empirical modeling approach that can predict the reaction rate of gas-solid reactions, regardless of the shortcomings of theory. In this way, non-parametric modeling offers a powerful tool for applied research and may contribute to the advancement of the thermochemical energy storage technology.

The necessity of recycling anthropogenically used phosphorus to prevent aquatic eutrophication and decrease the economic dependency on mined phosphate ores encouraged recent research to identify potential alternative resource pools. One of these resource pools is the ash derived from the thermochemical conversion of sewage sludge. This ash is rich in phosphorus, although most of it is chemically associated in a way where it is not plant available. The aim of this work was to identify the P recovery potential of ashes from sewage sludge co-conversion processes with two types of agricultural residues, namely wheat straw (rich in K and Si) and sunflower husks (rich in K), employing thermodynamic equilibrium calculations. The results indicate that both the melting behavior and the formation of plant available phosphates can be enhanced by using these fuel blends in comparison with pure sewage sludge. This enhanced bioavailability of phosphates was mostly due to the predicted formation of K-bearing phosphates in the mixtures instead of Ca/Fe/Al phosphates in the pure sewage sludge ash. According to the calculations, gasification conditions could increase the degree of slag formation and enhance the volatilization of K in comparison with combustion conditions. Furthermore, the possibility of precipitating phosphates from ash melts could be shown. It is emphasized that the results of this theoretical study represent an idealized system since in practice, non-equilibrium influences such as kinetic limitations and formation of amorphous structures may be significant. However, applicability of thermodynamic calculations in the prediction of molten and solid phases may still guide experimental research to investigate the actual phosphate formation in the future.

Wood log stoves are a common residential heating technology that produce comparably high pollutant emissions. Within this work, a detailed CFD model for transient wood log combustion in stoves was developed, as a basis for its optimization. A single particle conversion model previously developed by the authors for the combustion of thermally thick biomass particles, i.e. wood logs, was linked with CFD models for flow and turbulence, heat transfer and gas combustion. The sub-models were selected based on a sensitivity analysis and combined into an overall stove model, which was then validated by simulations of experiments with a typical wood log stove, including emission measurements. The comparison with experimental results shows a good accuracy regarding flue gas temperature as well as CO2 and O2 flue gas concentrations. Moreover, the characteristic behavior of CO emissions could be described, with higher emissions during the ignition and burnout phases. A reasonable accuracy is obtained for CO emissions except for the ignition phase, which can be attributed to model simplifications and the stochastic nature of stove operation. Concluding, the CFD model allows a transient simulation of a stove batch for the first time and hence, is a valuable tool for process optimization.