Mechanical and Thermal Modelling of Biomass Compression for the Design of a High Density Baler

Fibrous biomass and crop residues are traditionally used as a feedstock and bedding material for animals. Recently, they are increasingly used for renewable heat and power generation. Fibrous biomass in its unprocessed form has a low bulk density, which makes its transport over longer distances inefficient. The specific transportation costs can be reduced by compressing the biomass to higher bulk densities. Traditional ways to densify biomass are baling, briquetting and pelletisation. Densification through baling by an extrusion process can be performed at higher capacities than briquetting and pelletisation, which makes it more suitable for use in the field. Therefore, baling is traditionally used for densification of dried green forages (hay) such as grass and luzerne and the stems (straw) of grain crops like barley, wheat, oats and rice straw. A large number of process parameters related to biomass properties and machine design influence the process and thus affect product quality, machine efficiency and especially energy consumption. Due to the complexity of extrusion based densification where multiple process parameters interact simultaneously and due to the variation in the properties of the biomass to be handled, biomass densification machines often operate sub-optimal. Improvements to the machine design are typically tested during dedicated field tests where machines are subjected to different baling conditions. Those field tests often require significant investments, long time for preparation and are limited by weather conditions. Computer simulations could provide a cost efficient alternative for these field tests. Finite element analysis (FEA) is considered particularly useful for detailed analysis and design optimization of the biomass densification process through extrusion. The focus of this PhD research was therefore to model the mechanical and thermal behaviour of crop residues during densification in large square balers to increase bale density in an efficient way.
To describe the mechanical behaviour of crop residues at the bulk level, a rheological model involving exponential springs was developed and implemented into a commercial software package for FEA. Model parameters were determined by using data obtained from compression setups and dedicated field tests. The model was then validated on data that include various operating conditions and fibrous materials obtained through dedicated tests on compression setups and field tests with a large square baler. The developed 2D and 3D FEA models were used to describe the densification process in the compression setups and large square balers, and to identify the most important factors that influence process performance. Design changes which could improve the machine performance were identified and simulated. Finally, the designs which were considered most promising based on the simulations were identified.

As the backpressure on the bale has a large impact on the final density reached in a large square baler, the friction between the bale and the bale chamber walls determines the machine performance and product quality. This friction also generates friction heat, which increases the temperature of the bale and the doors in contact. This temperature increase was found to influence the mechanical properties of the materials in contact and thus also the overall process performance (bale density and level of pressure on doors). Therefore, the influence of temperature on the coefficient of friction between bale and steel structures was also investigated. It was concluded that the rising contact temperature decreases the coefficient of friction, which further results in a decrease in baling backpressure and lower product quality.

To analyse the influence of friction heat on temperature progression, mathematical models describing the thermal processes in the biomass material and the surrounding compression chamber have been implemented as well. Material properties and model parameters were extracted from available literature and field tests. The prediction performance of the model has been validated by fitting the estimated friction heat and comparing the simulated temperature evolution in the biomass and chamber walls to the corresponding values measured on a baler during dedicated field tests. From these simulations the heat losses during baling were estimated to be 11.5 kW. The simulation model was then used to propose suggestions for further process improvements related to thermal aspects of the process. One of these suggestions was to decrease the contact temperature resulting from the generated friction heat in order to maintain a high coefficient of friction between bale and steel structures. In this way, a high level of backpressure could be maintained during baler operation, resulting in a higher density for a similar pressure on the walls of the compression chamber.