A systematic approach for assessing steam balances for kraft pulp mills-model development and case study

Paper i proceeding, 2009

When estimating aggregated potential for energy efficiency and the introduction of new, efficient, technology solutions within the pulp and paper industry it is important to consider the existing energy system of the individual mills, since the present energy situation at each mill has a significant effect on the potential for future changes. Recognising the need to assess the existing system, one main problem is that detailed steam data is not publicly available or easily accessible, neither on the mill level nor the national level. In the study presented in this paper, a model is proposed for assessing the steam balance of a kraft pulp mill, assuming only a limited amount of public and easily accessible data. The model was developed and validated using detailed data from both model mills and real mills. From the validation of the model it was concluded that the model provides a good estimate of the total steam production and consumption, with an accuracy of +/-8%, and a fair estimate of the steam use at different pressure levels. The model was also applied in a benchmark study of the Swedish kraft pulp industry. In the benchmark study the potential for increased electricity production was analysed, assuming the steam balance for each mill given by the model. The results showed that all the kraft pulp mills in Sweden could increase their power production substantially by investing purely in new, larger, back-pressure turbines to avoid throttling any steam. If at the same time the mills invest in steam-saving measures, saving LP steam, and a condensing turbine, electricity production could increase even further, making all mills self-sufficient, and most mills even net exporters, of electric power. However, to analyse the potential for steam-saving measures further as well as the introduction of technology solutions other than increased electricity production, such as black liquor gasification, the model needs to be developed further.