A coupled physics informed neural network (CPINN) was used to simulate liquid diffusion controlled drying, an energy intensive process in the food industry. The architecture of the CPINN was designed to permit the prediction of thermo-physical properties and key source and sink terms at the solution boundaries which cause the solution to be highly coupled. The CPINN structure improves upon limitations of using PINNs in low-temperature food drying simulations, most notably allowing multiple and highly coupled variables to be simulated in additional to ensuring dynamic thermo-physical properties updates. The CPINN successfully solved a system 1-D partial differential equations (PDEs), capturing phenomena such as transient moisture diffusion and heat conduction, evaporative and convective heat transfer at the drying surface and moisture loss to the drying air. A benchmark simulation was used to compare the CPINN predicted product temperature, (T) over cap (p), and predicted moisture content, (X) over cap (p,) against a numeric solution. The mean absolute error for the respective comparisons was 0.12 degrees C and 0.0035 kgm(.) kg(s)(-1). Training the CPINN for the first time was the rate limiting step, requiring the greatest time to solve when compared to the numeric solution, with solution times of t(cptnm) = 321 min and t(rk) = 82.7 min, respectively, or a time reduction fraction of t(r) = 3.9, due to generalised initialisation of the CPINN parameters. By utilising a staged transfer learning approach, t(r) was reduced to a range of 0.28-0.027 whilst maintaining solution accuracy, representing a 3 to 37 times faster solution. By saving a library of CPINN models, solutions at key drying conditions of interest can be rapidly evaluated at run time, meaning the saved CPINN effectively acted as a method to compress solutions of PDEs. The techniques used here show how CPINNs can be applied to coupled and multi-scale PDEs using a physics-based approach to problems in the food processing and other sectors.

The integration of battery energy storage systems(BESS)throughout our energy chain poses concerns regarding safety,especially since batteries have high energy density and numerous BESS failure events have occurred.Wider spread adoption will only increase the prevalence of these failure events unless there is a step change in the management and design of BESS.To understand the causes of failure,the main challenges of BESS safety are summarised.BESS consequences and failure events are discussed,including specific focus on the chain of events causing thermal runaway,and a case study of a BESS explo-sion in Surprise Arizona is analysed.Based on the technology and past events,a paradigm shift is required to improve BESS safety.In this review,a holistic approach is proposed.This combines currently adopted approaches including battery cell testing,lumped cell mathematical modelling,and calorimetry,along-side additional measures taken to ensure BESS safety including the requirement for computational fluid dynamics and kinetic modelling,assessment of installation level testing of the full BESS system and not simply a single cell battery test,hazard and layers of protection analysis,gas chromatography,and com-position testing.The holistic approach proposed in this study aims to address challenges of BESS safety and form the basis of a paradigm shift in the safety management and design of these systems.

Commercial fire escape masks (FEMs) use packed bed filters to remove gaseous and vaporous toxic components in the event of building fires. Packed bed filters incur a high pressure drop and commercial masks have no method to remove environmental (fire) or process (reaction and adsorption) heats. Here we derive a computationally efficient numeric model based on a bi-linear driving force (LDF) model to investigate the purification of gas streams in a square channelled monolith filter containing an impregnated activated carbon (AC) section to adsorb and react toxic components, and a section consisting of shape stable phase change materials (SS-PCMs) to absorb heat. The modelled test gas mixture contained an adsorbing component, cyclohexane, and a reacting component, carbon monoxide, permitting the combined effects of heat generation, heat absorption, component reaction and component adsorption to be studied for a novel filter. The biLDF model was validated against a three-dimensional model and provided excellent accuracy at significantly reduced computational time ca. 99.7%. Additionally, the bi-LDF model was used to optimise the dimensions and configuration of the filter, specifically finding an optimal channel diameter, d(ch), to wall thickness, t(w), aspect ratio of d(ch) = 1.3t(w). The optimal configuration consisted of an initial 2.0 cm long impregnated AC section followed by a 2.5 cm SS-PCM section at the outlet, providing 18 min of thermal protection whilst preventing cyclohexane vapour breakthrough for 21 min. Pt/TiO2 was confirmed to be a viable CO oxidation catalyst with a minimum weight fraction within the impregnated monolith of 2.5 wt%. The success of this work represents a step change in FEM design and more widely in air purification devices where heat absorption is important.

The primary function of commercial fire escape masks (FEMs), fitted with granulated activated carbon (AC) packed bed filters, is to provide at least 15 min of respiratory protection by removing toxic gases and particulates from surrounding air in building fires. In this work, the extended functionality of heat entrapment and its impact on inhalation temperature and adsorption performance by using shape-stable phase change material whilst maintaining low pressure drop is reported for the first time. The proposed filter contained an array of monoliths where each monolith consisted of three functional sections, namely the pre-cooler, AC adsorbent section and post-cooler. The pre- and post- coolers consisted of polyethylene glycol 4000/triallyl isocyanurate and were intended to absorb environmental and process heats from the inhaled atmosphere. Numerical models were developed to describe the species and energy transport within the monolith filters and were compared against packed bed filters. The representative challenge conditions were set at an inhalation rate of 50 L min-1, trace amount of butane (1000 ppm) and inlet air temperature of 80 degrees C. The best performing filter contained nine monoliths each with density of 734 channels per square inch, and could protect the user from excessive inhalation temperatures for 22 min and butane breakthrough for approximately 14 min whilst maintaining low pressure drop of 27.4 Pa. In comparison to an equivalent mass packed bed, the monolith provided additional high temperature protection, extended butane breakthrough time by a maximum of 84% and reduced pressure drop by 25%. This work demonstrates promising opportunities to move the FEM industry forward and the possibility for the technology to be used in general industrial respirators in applications such as agriculture, chemical and pharmaceutical industries.