Thierry Massart

Structural Damage Analysis of Masonry Walls using Computational Homogenization

This contribution deals with the application of computational homogenization techniques for structural masonry computations, as an alternative to the formulation of complex closed-form macroscopic constitutive laws. The complexity of modeling masonry material stems from the anisotropy evolution and localization induced by mesostructural damage. This phenomenon appears with preferential damage orientations, which are intimately related to the initial periodic structure of the material. The upscaling procedure used here relies on the formulation of mesoscopic constitutive laws at the level of the individual brick and mortar materials. A mesostructural unit cell with its corresponding periodicity requirements is used to deduce the average response of the masonry material through a scale transition. At the macroscopic scale, this averaged material response is used in the frame of an enhanced continuum approach with embedded localization bands, the widths of which are directly deduced from the initial periodicity of the material. The results obtained by the framework are illustrated and discussed by means of a structural computation example, which involves a complex cracking evolution together with fully anisotropic damage development.

Infinite State Model Checking by Abstract Interpretation and Program Specialisation

We illustrate the use of logic programming techniques for finite model checking of CTL formulae. We present a technique for infinite state model checking of safety properties based upon logic program specialisation and analysis techniques. The power of the approach is illustrated on several examples. For that, the efficient tools logen and ecce are used. We discuss how this approach has to be extended to handle more complicated infinite state systems and to handle arbitrary CTL formulae.

Coupled friction and roughness surface effects in shallow spherical nanoindentation

When nanoindentation is used for thin film characterization, usually shallow indents are made to avoid the spurious effect of the substrate. However, surface effects stemming from surface roughness and friction can become important in shallow indentation depths, potentially resulting in the variation of nanoindentation results. A numerical study is conducted aiming for a more complete understanding of the coupled influence of friction and sample surface roughness in nanoindentation of pure nickel, using a slip rate dependent friction law. Two experimentally used post-treatment methods are applied to obtain the elastic properties from the raw numerical data. Results confirm the strong interaction between these two contributions of surface effects, and their cumulative effect leads to significant variations in both the indenter load vs. displacement curves and the evaluated elastic modulus. The resulting dispersion is somewhat higher than the one computed for a slip rate independent Coulomb friction. The velocity-weakening nature of the used friction law, is observed to induce a stick-slip behavior which has a manifestation similar to pop-ins in the load-displacement curves.

Synthesis of communicating controllers for distributed systems

We consider the control of distributed systems composed of subsystems communicating asynchronously; the aim is to build local controllers that restrict the behavior of a distributed system in order to satisfy a global state avoidance property. We model our distributed systems as communicating finite state machines with reliable unbounded FIFO queues between subsystems. Local controllers can only observe their proper local subsystems and do not observe the queues. To refine their control policy, they can use the FIFO queues to communicate by piggybacking extra information to the messages sent by the subsystems. We define synthesis algorithms allowing to compute the local controllers. We explain how we can ensure the termination of this control algorithm by using abstract interpretation techniques, to overapproximate queue contents by regular languages. An implementation of our algorithms provides an empirical evaluation of our method.

dSL: An Environment with Automatic Code Distribution for Industrial Control Systems

We present and motivate the definition and use of the language and environment dSL, an imperative and event driven language designed to program distributed industrial control systems. dSL provides transparent code distribution using simple mechanisms. Its use allows the industrial control systems designer to concentrate on the sequences of control required; the dSL compiler-distributer taking into account the distribution aspects. We show the advantages of our approach compared to others proposed using e.g. shared memory or synchronous languages like Esterel, Lustre or Signal.

On the complexity of partial order trace model checking

The theoretical complexity of CTL*, CTL and LTL model checking over finite partial order traces are studied. CTL* and CTL model checking are PSPACE-complete and that the LTL model checking is coNP-complete. Since CTL is a fragment of CTL*, it implies that the problem for CTL* is also PSPACE-hard. Then, it is showed that, for CTL*, the problem is in PSPACE. Again, since CTL is a fragment of CTL*, it follows that the problem for CTL is also in PSPACE. Those results allow to conclude that for CTL* and CTL, the model checking problem is PSPACE-complete. Several theorems and model checking have been used to prove this complexity.

The formal design of distributed controllers with d SL and Spin

We study the formal verification of programs written in dSL, an extension of the standard ST language used to program industrial controllers. It proposes a trade off between industrial and formal verification worlds. The main advantage of dSL is to provide a transparent code distribution through low level communication mechanisms. The behavior of the synthesized distributed system can therefore be formally modeled, easily monitored and formally verified. The verification of a dSL program, realized with the Spin tool, is eased by the definition of a lattice of models linked with a simulation relation preserving next-free LTL formulae. We show that, although dSL is an industrial programming language, it gives the possibility to verify systems designed with it. We illustrate the benefit of our approach with a simple control system of two canal locks.

Monitoring distributed controllers: when an efficient LTL algorithm on sequences is needed to model-check traces

It is well known that through code instrumentation, a distributed systems finite execution can generate a finite trace as a partially ordered set of events. We motivate the need to use LTL model-checking on sequences and not on traces as defined by Diekert and Gastin, to validate distributed control systems executions, abstracted by such traces, and present an efficient symbolic algorithm to do the job. It uses the standard method proposed by Vardi and Wolper, which from the LTL formula, builds a monitor that accepts all the bad sequences. We show that, given a monitor and a trace, the problem to check that both the monitor and the trace have a common sequence is NP-complete in the number of concurrent processes. Our method explores the possible configurations symbolically, since it handles sets of configurations. Moreover, it uses techniques similar to the partial order reduction, to avoid exploring as many execution interleavings as possible. It works very well in practice, compared to the standard exploration method, with or without partial order reduction (which, in practice, does not work well here).

Synchronizing Objectives for Markov Decision Processes

We introduce synchronizing objectives for Markov decision processes (MDP). Intuitively, a synchronizing objective requires that eventually, at every step there is a state which concentrates almost all the probability mass. In particular, it implies that the probabilistic system behaves in the long run like a deterministic system: eventually, the current state of the MDP can be identified with almost certainty. We study the problem of deciding the existence of a strategy to enforce a synchronizing objective in MDPs. We show that the problem is decidable for general strategies, as well as for blind strategies where the player cannot observe the current state of the MDP. We also show that pure strategies are sufficient, but memory may be necessary.

A Calculus to Define Correct Tranformations of LOTOS Specifications

Abstract We present a basic agent calculus which uses simple actions and contains 6 kinds of operators: the sequence, the choice, the parallel composition, the recursion, the hiding and the relabelling operator. This calculus is used to define correct transformations. As an example, we solve a problem of distribution of an agent into several locations (places): we provide an equality preserving transformation which transforms an agent P into subagents Q1, Q2, …, Qn which synchronize by using a reliable medium M. We also show that, since LOTOS may be mapped on this calculus, it may be used as a theoretical basis for LOTOS developments.