Quantitative automata model checking of autonomous stochastic hybrid systems

Quantitative model-checking of controlled discrete-time Markov processes

“…Using LTL and automata theory for finite statespace cdt-MP has a long history, part of which can be consulted in [BK08, Section 10.8]. However, extensions to the general state-space case have only appeared recently: [AKM11] has provided an extension to the uncontrolled case (where trivially 10 Note that in cdt-MP the choice of the distribution of the successor state is "labelled" by actions, whereas in gambling models such choice is unlabelled. One may think of this being similar to internal and external non-determinism in probabilistic automata [SL95], however there is no semantic difference between cdt-MP and gambling models, and in both cases non-determinism can be considered both as an internal one or as an external one.…”

Modelling and statistical model checking of a microgrid

“…9 Thermostat value X C where C CW is the thermal capacity of the water circuit and Q cool is the thermal energy removed by chillers. 2 The chilled water temperature controller keeps the chilled water temperature at the set point value T CWSP (another control parameter). The energy requirement is controlled by a feedback loop and hence designed as a PI control scheme:…”

“…Numerical methods to solve model-checking problems of stochastic hybrid systems against some temporal formulas (even reachability) are algorithmically involved and suffer from the curse of dimensionality [1,2,21]. Statistical model checking avoids these problems by resorting to discrete-event simulation (using Monte Carlo techniques).…”

“…For the analysis of the microgrid, we resort to model checking. Recent developments in model checking of stochastic hybrid systems show the possibility to check a rich variety of properties such as probabilistic reachability and invariance [1], or LTL [2]. As the stochastic dynamics and complexity of the microgrid configuration go beyond the scope and feasibility of the above techniques that typically rely on discretization and dynamic programming, we resort here to statistical model checking [4,25].…”

Multi-objective optimal control of stochastic hybrid systems

“…In [7]- [10], the work of [6] was extended to quantify reachability properties in the probabilistic setting. In particular, it was shown that the dynamic programming solutions of [7], [8] constitute the main ingredient needed to verify logical specifications (in probabilistic computational tree logic (PCTL) [10] and linear temporal logic (LTL) [11]) in probability for DTSHS.…”