Solar heating forecast in heat system optimal control

Use of solar radiation forecasting in heating system optimization simulation was carried out in this study. Open data from Finnish Meteorological Institute (FMI) data base was used to generate solar and temperature forecast. Temperature forecast was used to estimate the coefficient of performance (COP) for heat pump and generate heat load forecast with heat model. Temperature data from previous year was used to build heat model of case building (district heating connected building in Espoo). The solar forecast calculation was built using Python language, the heat forecast model with APL and the optimization code using Julia language. The aim of the optimization was to minimize operational costs and investigate results using different storage and collector sizes. It was possible to get savings in heating costs with solar collectors and heat pump. Storage can also yield savings both with heat pump and solar collectors. Excess energy from solar collectors can be stored to storage and used later and also heat pump usage shifted to different hours to take advantage on hourly electricity prices. The results indicate that with low electricity price that we have now in Nordic markets it is possible have cheaper heating energy than district heating. However due to the limited scope of the work here, maintenance fees are not investigated. The operation cost can however be up to 50% less without taking account the maintenance. For total profitability analysis, investment cost and longer time period performance simulation would be needed. Target of the study was to investigate integration of solar radiation and heat load forecast models to optimization code. Two models were combined via a bit non-optimal way because both had been previously built on different platforms. It was discussed during the project that combined system could be rewritten with Python for more elegant approach.