Soft Sensing of LPG Processes Using Deep Learning.

This study investigates the integration of soft sensors and deep learning in the oil-refinery industry to improve monitoring efficiency and predictive accuracy in complex industrial processes, particularly de-ethanization and debutanization. Soft sensor models were developed to estimate critical variables such as the C2 and C5 contents in liquefied petroleum gas (LPG) after distillation and the energy consumption of distillation columns. The refinery's LPG purification process relies on periodic sampling and laboratory analysis to maintain product specifications. The models were tested using data from actual refinery operations, addressing challenges such as scalability and handling dirty data. Two deep learning models, an artificial neural network (ANN) soft sensor model and an ensemble random forest regressor (RFR) model, were developed. This study emphasizes model interpretability and the potential for real-time updating or online learning. The study also proposes a comprehensive, iterative solution for predicting and optimizing component concentrations within a dual-column distillation system, highlighting its high applicability and potential for replication in similar industrial scenarios.

Data for the project management, life cycle inventory, costings and energy production of a ground-mounted photovoltaic farm in Greece.

Data was collected using standard communication equipment and invoices provided by an established civil construction and renewable energy development and operation company. Data referring to the construction, costings, operation and environmental impacts of a photovoltaic farm were recorded into four distinct Excel files namely: i) Project Management Data, ii) Life Cycle Inventory (LCI), iii) Electricity Generation Data and iv) Operational Cost Data. For the project management, the given quantities of the resources used in each activity could be further combined with the costs from different geographical and time regions to estimate overall project implementation costs for similar projects. The LCI data for the materials and transportation used can set the basis for life cycle assessment modelling of ground-mounted photovoltaic farms of that size and type. The electricity generation data along with meteorological parameters and location coordinates can be further enhanced to predict and manage energy generation and cashflow of expectations installations of this type and size over time. Finally, the data referring to a number of cost categories('maintenance costs', 'operational costs', 'insurance costs' and 'any other costs'), especially combined with the previously mentioned types of data could support a holistic technoeconomic and environmental assessment of comparable commercial photovoltaic installations. In addition, these data can be used for a comparative multi-disciplinary evaluation between photovoltaics and among various renewable electricity generation alternatives and traditional fossil fuel-based options as well.