Second-Order Training of Adaptive Critics for Online Process Control

A new algorithm for training of nonlinear optimal neuro-controllers (in the form of the model-free, action-dependent, adaptive critic paradigm). Overcomes problems with existing stochastic backpropagation training: need for data storage, parameter shadowing and poor convergence, offering significant benefits for online applications.

Insight into the key aspects of the regeneration process in the NOx storage reduction (NSR) reaction probed using fast transient kinetics coupled with isotopically labelled (NO)-N-15 over Pt and Rh-containing Ba/Al2O3 catalysts

For the first time, the coupling of fast transient kinetic switching and the use of an isotopically labelled reactant (15NO) has allowed detailed analysis of the evolution of all the products and reactants involved in the regeneration of a NOx storage reduction (NSR) material. Using realistic regeneration times (ca. 1 s) for Pt, Rh and Pt/Rh-containing Ba/Al2O3 catalysts we have revealed an unexpected double peak in the evolution of nitrogen. The first peak occurred immediately on switching from lean to rich conditions, while the second peak started at the point at which the gases switched from rich to lean. The first evolution of nitrogen occurs as a result of the fast reaction between H2 and/or CO and NO on reduced Rh and/or Pt sites. The second N2 peak which occurs upon removal of the rich phase can be explained by reaction of stored ammonia with stored NOx, gas phase NOx or O2. The ammonia can be formed either by hydrolysis of isocyanates or by direct reaction of NO and H2.

The study highlights the importance of the relative rates of regeneration and storage in determining the overall performance of the catalysts. The performance of the monometallic 1.1%Rh/Ba/Al2O3 catalyst at 250 and 350 °C was found to be dependent on the rate of NOx storage, since the rate of regeneration was sufficient to remove the NOx stored in the lean phase. In contrast, for the monometallic 1.6%Pt/Ba/Al2O3 catalyst at 250 °C, the rate of regeneration was the determining factor with the result that the amount of NOx stored on the catalyst deteriorated from cycle to cycle until the amount of NOx stored in the lean phase matched the NOx reduced in the rich phase. On the basis of the ratio of exposed metal surface atoms to total Ba content, the monometallic 1.6%Pt/Ba/Al2O3 catalyst outperformed the Rh-containing catalysts at 250 and 350 °C even when CO was used as a reductant.

The Y-procedure: How to extract the chemical transformation rate from reaction-diffusion data with no assumption on the kinetic model

The paper presents a new method to extract the chemical transformation rate from reaction–diffusion data with no assumption on the kinetic model (“kinetic model-free procedure”). It is a new non-steady-state kinetic characterization procedure for heterogeneous catalysts. The mathematical foundation of the Y-procedure is a Laplace-domain analysis of the two inert zones in a TZTR followed by transposition to the Fourier domain. When combined with time discretization and filtering the Y-procedure leads to an efficient practical method for reconstructing the concentration and reaction rate in the active zone. Using the Y-procedure the concentration and reaction rate of a non-steady state catalytic process can be determined without any pre-assumption regarding the type of kinetic dependence. The Y-procedure is the basis for advanced software for non-steady state kinetic data interpretation. The Y-procedure can be used to relate changes in the catalytic reaction rate and kinetic parameters to changes in the surface composition (storage) of a catalyst.

Making Sense of Leaving Care: The contribution of Bridges Model of Transition to Understanding the Psycho-social process.

This paper is based on research into the transition of young people leaving public care in Romania. Using this specific country example, the paper aims to contribute to present understandings of the psycho-social transition of young people from care to independent living by introducing the use of Bridges (2002) to build on existing theories and literature. The research discussed involved mixed methods design and was implemented in three phases: semi-structured interviews with 34 care leavers, focus groups with 32 professionals, and a professional-service user working group. The overall findings confirmed that young people experience two different, but interconnected transitions - social and psychological - which take place at different paces. A number of theoretical perpectives are explored to make sense of this transition including attachment theory, focal theory and identity. In addition, a new model for understanding the complex process of transitions was adapted from Bridges’ (2002) to capture the clear complexity of transition which the findings demonstrated in terms of their psycho-social transition. The paper concludes with messages for leaving and after care services with an emphasis on managing the psycho-social transition from care to independent living.

A material model for multiaxial stretching and stress relaxation of polypropylene under process conditions

Polypropylene sheets have been stretched at 160 °C to a state of large biaxial strain of extension ratio 3, and the stresses then allowed to relax at constant strain. The state of strain is reached via a path consisting of two sequential planar extensions, the second perpendicular to the first, under plane stress conditions with zero stress acting normal to the sheet. This strain path is highly relevant to solid phase deformation processes such as stretch blow moulding and thermoforming, and also reveals fundamental aspects of the flow rule required in the constitutive behaviour of the material. The rate of decay of stress is rapid, and such as to be highly significant in the modelling of processes that include stages of constant strain. A constitutive equation is developed that includes Eyring processes to model both the stress relaxation and strain rate dependence of the stress. The axial and transverse stresses observed during loading show that the use of a conventional Levy-Mises flow rule is ineffective, and instead a flow rule is used that takes account of the anisotropic state of the material via a power law function of the principal extension ratios. Finally the constitutive model is demonstrated to give quantitatively useful representation of the stresses both in loading and in stress relaxation.

Towards Utilizing Model-Driven Engineering of Composite Applications for Business Performance Analysis

Composite Applications on top of SAPs implementation of SOA (Enterprise SOA) enable the extension of already existing business logic. In this paper we show, based on a case study, how Model-Driven Engineering concepts are applied in the development of such Composite Applications. Our Case Study extends a back-end business process which is required for the specific needs of a demo company selling wine. We use this to describe how the business centric models specifying the modified business behaviour of our case study can be utilized for business performance analysis where most of the actions are performed by humans. In particular, we apply a refined version of Model-Driven Performance Engineering that we proposed in our previous work and motivate which business domain specifics have to be taken into account for business performance analysis. We additionally motivate the need for performance related decision support for domain experts, who generally lack performance related skills. Such a support should offer visual guidance about what should be changed in the design and resource mapping to get improved results with respect to modification constraints and performance objectives, or objectives for time.