Nevel's Forssa plant

Increased efficiency through fully digital operations

The Forssa energy production plant in Finland – one of over 150 such sites owned and operated by Nevel – generates 70MW of energy a year using solid bioenergy as its main fuel. In 2017 Nevel began a project to fully digitalise operations at the plant in order to optimise efficiency, reduce emissions, cut OPEX and increase the share of renewable fuel sources. Utilising the Nevel Remote operations and maintenance platform has enabled all these goals to be fully realised.

An optimised combustion process

The first step in the process to fully digitalise the plant’s operations was to initiate remote monitoring. This gave improved access to data that went well beyond what is gained from a normal site visit and covers over 200 items, ranging from residual oxygen levels, and CO₂ and NOx emissions to financial information. Analysing the site data has enabled Nevel to optimise the production process and make better informed decisions. The site is operated around the clock from Nevel’s remote operation centre, which uses machine learning to optimise the site operations. Employees at the site work during normal working hours and focus primarily on preventive maintenance work. 

Improved boiler use and maintenance

The collected data now allows the boiler to be run at the most efficient temperature and minimised residual oxygen content. The data also gives valuable insight into the performance of the heat exchanger, enabling pre-emptive cleaning and better heat dissipation, meaning less energy is needed to produce the same amount of heat.

Sensor data is used to identify acute problems such as imminent component failures and abnormal readings. Machine learning methods analyse the data for correlations and dependencies that can further support maintenance planning. The resulting improvement in predictability reduces downtime and enables service intervals that accurately reflect the needs of the site.

Optimal use of the accumulator

The plant’s accumulator is now run automatically based on daily energy price variations. Using the collected data, the accumulator is charged and discharged depending on the network’s needs, and its operation can be optimised to generate most benefit, for example charged when energy prices are lower and discharged when they are higher. The accumulator also enables even temperatures of district heating during any malfunctions or shutdowns.

Better production forecasting

By integrating information such as historic consumption data, the energy production process optimises energy use with, for example, fuel use, fuel mix and temperature. With production data from all sites being analysed centrally, each plant benefits from an enormous amount of data that is modelled with machine learning to enable features like automated transactions – selling at a predefined price – and highly accurate production forecasting, all of which helps plants to respond more quickly and flexibly to changes in the energy market.