Stable memory in substation automation: a case study

Abstract

In high voltage substations, electromagnetic interference results in transient faults. Typically, hardware stable storage devices are used to stabilise data, to store status information of the controller and to protect the application against these faults. A case study of electric substation automation introduced a controller distributed on a parallel architecture and substituted the stable storage device by a fault-tolerant software implementation of stable memory. This allows the application to stabilise data to ensure its integrity. Based on the combination of temporal and spatial redundancy, it tolerates permanent faults in memory and transient faults affecting computation, input and memory devices. Transient faults lead to extra cycles before data is stabilised; permanent faults can be masked or lead to reconfiguration. The flexibility of the software architecture allows to fine-tune parameters according to application's needs. This enhances reusability in wide automation areas. Evaluation confirms the suitability of the software solution during typical disturbances.

Application-Layer Fault-Tolerance Protocols

“…Most of these tools have been designed and developed by the TIRAN partners and have been especially conceived in order to meet hard real-time requirements. A deeper description of some of these tools can be found in (Deconinck, Botti, Cassinari, De Florio, & Lauwereins, 1998;Deconinck et al, 1999). As a side effect of using these tools, the TIRAN Backbone (see Sect.…”

System Fault Tolerance Specification: Proposal of a Method Combining Semi-formal and Formal Approaches

“…by using mechanisms which maintain the current state (judged correct) till a confirmation of correctness of the next state (e.g. by using a Stable Memory technique [Deco98]) c) the transitions between two subsequent acceptable states must be non interruptible and without uncertainties -e.g. by an atomic action d) evolution must be coherent among the different system components -e.g.…”

Integrating recovery strategies into a primary substation automation system

“…allocate functions to tasks and nodes; set parameters) and to express recovery strategies -i.e., to describe diagnosis, containment and recovery actions to be executed when an error is detected [5], [8]. As such, it is possible to start a standby task, to reset a node or link, to move a task to another node, to generate synchronization signals for reconfiguration, etc.…”

“…The THEN (ELSE) keyword marks the beginning of a list of fault tolerance actions to be executed when the guard is evaluated as true (false). Three sets of software modules build up the middleware architecture [4], [5].…”

“…The DepAuDE architecture deploys a set of middleware modules to provide fault tolerance by exploiting the embedded systems' distributed hardware and by separating functional behavior from the recovery strategy, i.e., the set of actions to be executed when an error is detected [4], [5].…”