In recent years, explainable artificial intelligence (XAI) techniques have been developed to improve the explainability, trust and transparency of machine learning models. This work presents a method that explains the outputs of an air-handling unit (AHU) faults classifier using a modified XAI technique, such that non-AI expert end-users who require justification for the diagnosis output can easily understand the reasoning behind the decision. The method operates as follows: First, an XGBoost algorithm is used to detect and classify potential faults in the heating and cooling coil valves, sensors, and the heat recovery of an air-handling unit. Second, an XAI-based SHAP technique is used to provide explanations, with a focus on the end-users, who are HVAC engineers. Then, relevant features are chosen based on user-selected feature sets and features with high attribution scores. Finally, a sliding window system is used to visualize the short history of these relevant features and provide explanations for the diagnosed faults in the observed time period. This study aimed to provide information not only about what occurs at the time of fault appearance, but also about how the fault occurred. Finally, the resulting explanations are evaluated by seven HVAC expert engineers. The proposed approach is validated using real data collected from a shopping mall.

An investigation of the pyrolysis kinetics of American, Chinese and Estonian oil shales was conducted applying a non-isothermal thermo­gravi­metric analysis (TGA). TGA weight loss curves clearly indicate that the pyrolysis of all oil shales tested is independent of their geographic origin, and is mainly taking place in the temperature range of 300 to 500 ºC. As expected, at temperatures above 700 ºC mass loss due to the decomposition of oil shale carbonates was detected, except for the Kentucky and Chinese 2 oil shale samples. The kinetic decomposition rate parameters such as activa­tion energy and pre-exponential factor were calculated applying the Coats-Redfern integral and direct Arrhenius methods. Independent of the oil shale kerogen origin, pyrolysis occurs in two consecutive temperature zones with slightly dissimilar kinetic parameter values. The activation energy values obtained were in the range of 14 to 31 kJ/mol for the low and 70 to 149 kJ/mol for the high temperature zone.

Carbon dioxide emissions are considered a major environmental threat. To enable power production from carbon-containing fuels, carbon capture is required. Oxy-fuel combustion technology facilitates carbon capture by increasing the carbon dioxide concentration in flue gas. This study reports the results of calcium rich oil shale combustion in a 60 kW_th circulating fluidized bed (CFB) combustor. The focus was on the composition of the formed flue gas and ash during air and oxy-fuel combustion. The fuel was typical Estonian oil shale characterized by high volatile and ash contents. No additional bed material was used in the CFB; the formed ash was enough for the purpose. Two modes of oxy-fuel combustion were investigated and compared with combustion in air. When N₂ in the oxidizer was replaced with CO₂, the CFB temperatures decreased by up to 100 °C. When oil shale was fired in the CFB with increased O₂ content in CO₂, the temperatures in the furnace were similar to combustion in air. In air mode, the emissions of SO₂ and NO_x were low (<14 and 141 mg/Nm³ @ 6% O₂, respectively). Pollutant concentrations in the flue gas during oxy-fuel operations remained low (for OXY30 SO₂ < 14 and NO_x 130 mg/Nm³ @ 6% O₂ and for OXY21 SO₂ 23 and NO_x 156 mg/Nm³ @ 6% O₂). Analyses of the collected ash samples showed a decreased extent of carbonate minerals decomposition during both oxy-fuel experiments. This results in decreased carbon dioxide emissions. The outcomes show that oxy-fuel CFB combustion of the oil shale ensures sulfur binding and decreases CO₂ production.