The pulp and paper industry is an energy-intensive industrial sector that faces several challenges such as increased competition and rising feedstock and energy prices. To adress this, it is crucial for the industry to improve the material and energy efficiencies to the greatest possible extent. Process integration methods like pinch analysis and mathematical programming are useful tools for evaluating possible process alternatives, i.e. applications of new technologies, changes to new equipment and/or different operating conditions. Development of industrial energy combines is an interesting approach towards an overall optimization of energy and material flows. One problem is often that there are a large number of essentially different actors and financers that are interested in studying other parameters than those that are normally investigated in process integration studies, for example national or regional economics and various social values.In this project, engineering, economic and statistical tools and methods have been applied separately as well as in combination for different types of investigations conducted at the paper and pulp mill Billerud Karlsborg AB in Kalix, Sweden. One main objective has been to develop a process integration model of the mill based on the reMIND method to be used for introductory process simulations of the existing mill configuration. Additionally, pinch analysis has been used to identify alternatives for energy savings in the mill. Another objective has been to develop a regional economic market model (ReCOM) that should be suitable for analysis and predictions of price changes on relevant feedstock markets. A more simplified model based on the reMIND method has been used for intitial studies on how the mill can be turned into a biorefinery. The main purpose of that work has been to investigate if biomass gasification can be economically interesting for the mill and if so, under what boundary conditions. A statistical technique, conjoint analysis, has been used to study and analyze the attitude of employed people at the mill to changes in the production process that may affect for example the local and global environment etc. Finally, possible interactions between the different models and tools have been investigated.The reMIND modelling of the existing mill configuration has showed several alternatives to save steam and fuel. For example, if the wood-chips supplied to the digester is pre-heated from a temperature of 0°C to say 60°C by the use of low grade residual heat, approximately 1.5 ton per hour of 10 bar steam or 5 ton per hour of biomass fuels can theoretically be saved. Furthermore, if the inlet liquor temperature to effect 4 of the evaporation plant increases from 85 to 105°C, the steam used for evaporation decreases from 77 to 66 ton per hour and as a consequence, the biomass fuel supply to the bark boiler decreases from 51 to 39 ton per hour. This, however, also leads to a slightly reduced electricity production, from 35 to 34 MW due to a reduced production of the high pressure steam.The results from the developed ReCOM model, suggest that only none to small changes in the fibrous input prices from an increase in the fuel price (affecting the forestry sector) and a small price increase as a result from a reduced supply of purchasable wood-chips and pulp wood. The small effect that increasing fuel prices has on the fibrous input prices can largely be explained by the relatively small cost share that fuels have in the forestry sector. An increase of the labour costs would most likely have a larger impact. As for the price effect from a reduction in the supply of purchasable wood-chips, there is a substitution possibility between purchased and internally produced wood-chips for the pulp mill. However, when the limit for how much internally produced wood-chips is reached its will probably results in larger price effectsThe Pinch study of the mill indicated that there is a theoretical steam-saving potential of 18.5 MW, corresponding to 12% of the current steam demand. Two different retrofit proposals were suggested for how to achieve specific steam saving levels in practice. According to a basic retrofit proposal, a steam saving of 5.8 MW could be achieved at an investment cost of 7 MSEK while a more rigorous retrofit would enable steam savings of 11 MW at an investment cost of 14.5 MSEK. An approach for using these results in a reMIND model of the mill has also been proposed.The results from the more simplified reMIND modelling have showed that if the mill starts to produce DME via biomass gasification, the necessary policy support to make it economically feasible ranges from 92-561 SEK per MWh biofuel (DME) over four different future scenarios. This could be compared to the Swedish exemption from energy tax on biofuels, which currently amounts to approximately 275 SEK per MWh. It is also concluded that biomass gasification results in a larger net CO2 reduction when integrated with the pulp and paper mill, than when the mill and the gasification plant operate separately.The conjoint analysis showed that it is possible to find groups of respondents that were unknown prior to the study. If an organisation wants to implement a change in the process, conjoint analysis can be used to identify groups of participants with similar preferences and then tailor information to suit these specific groups.Many possibilities for the different models to interact have been identified and illustrated. The interaction between the reMIND method and ReCOM is based on exchanging information on fibrous input prices and quantities and conducted through an iterative process. The results indicate that the models can interact to produce more robust and reliable conclusions regarding optimal resource utilization suggesting that the described approach is feasible and that further research efforts can be made to extend the models. Pinch analysis and reMIND modelling has in other studies shown to be able to interact iteratively. In this study, the retrofit proposals obtained from the pinch analysis could serve as inputs to future reMIND modelling. Another interaction between reMIND and pinch analysis that has been identified during the project is to use pinch analysis to evaluate the opportunity to pre-heat certain process streams. The results from a conjoint analysis are quantitative in form of regression coefficients. However, to use these numbers for example in a Pareto front analysis will be difficult as the numbers has no monetary, energy or emission unit. Nonetheless, conjoint analysis can interact in many different ways with ReCOM as well as the reMIND models. For example, to choose scenarios to be modelled in ReCOM where the factors in the conjoint analysis can be tailored to indicate how the market would respond in a hypothetical situation. Conjoint analysis can be used to weight different factors in the reMIND model. The weighting can possibly also be used in the ReCOM model.This work has illustrated how the various engineering, economic and statistical methods and tools can be used both separately and in combination to help an industry towards more energy-efficient production processes.
2010. , 68 p.