Installation av en ackumulatoratank på Sandvik Coromant i Gimo

Sammanfattning: Sandvik Coromant in Gimo wants to implement an accumulator tank to reduce their usage of fossil fuelsand therefore lessen their climate impact. The system implemented at the moment is dimensioned fortemperatures down to -14◦C, but at lower temperature the system cannot meet the demands. That is whenthe industry needs to use oil as backup. This report studies the possibilities to install an accumulator tankthat will cover the demand when the temperatures are below -14◦C. This report constructs and takes twomodels into consideration: one is an approximative model whereas the other is an adequate model. Theapproximative model is divided into two scenarios. The first scenario studies the heat storage needs for theaccumulator tank for the longest period when the ambient temperature is -14◦C and decreases to -20◦Cthen increases back to -14◦C. The second scenario uses the same method as the previously mentionedscenario, but the interval is down to -17◦C. Results from the first scenario gave the accumulator tanka volume of 244.8 m3 whilst the second scenario, the accumulator tank had a volume of 92.5 m3. Theapproximative model gave no credible results since the data used and available at the time was inadequate,but also because of other sources of errors.The adequate model is based on a code written in Python, one that Sandvik Coromant can use andapply the needed data themselves. The model takes the ambient temperature and heat income as inputsand will with the help of the input dimension an accumulator tank but also calculate its heat losses.Since there was no adequate data available, no result could be obtained from the model. Instead, themodel’s robustness and reliability were tested with the help of a sensitivity analysis but also through avalidation. The sensitivity analysis showed that the result from the model is mostly impacted when theambient temperature varies. To validate the model’s ability to correctly calculate the dimensions and thevolume of an accumulator tank, a master thesis was used that studied and dimensioned two accumulatortanks. When the master thesis values were used as inputs, the model gave a deviation of 0 % and 0.4%. The model’s ability to calculate the heat storage need was validated through using hourly data fromFebruary and April given from Nevel as inputs. The results from the model were compared to Nevels owncalculation done in excel. The results from February gave a 23.5 % deviation and for April the deviationwas 11 %. These deviations can seem big but the heat demand for Sandvik Coromant was calculatedthrough different equations in the model compared to Nevel’s own calculations. When Nevel’s equationwas used for the heat demand in the model the deviation was 0 % and 2.3 % for February respectiveApril. This shows that the model is capable of calculating the heat storage needs and the volume for theaccumulator tank and is therefore reliable.Furthermore, the results obtained was discussed and sources of errors were analyzed. From the discussionsome conclusions could be drawn. Some of the conclusions drawn are that the tank should be placed inconnection with the current oil tanks and based on the size of the accumulator tank two smaller tankscould be constructed instead. Based on the sensitivity analysis and the validation, it can be concludedthat the model will work without significant errors.