FDD is an automatic process by which faulty operation,
degraded performance, and broken components in a
physical system are detected and the causes identifed.
FDD tools are based on algorithms that process data
coming from the equipment to determine whether it is
experiencing a fault. The tool may be passive, analyzing
the operation of the equipment or system without
altering any of its set-points or control outputs, or active,
automatically initiating changes to produce or simulate
operating conditions that cover a more complete range of
conditions than is likely to be encountered for some time
under normal operation.
The main objective of the FDD system is to provide web-
based real-time rules for proactive monitoring, diagnosis
of faults and performance analysis. The system identifes
and detects faults, analyzes them using information
processing and provides a diagnosis. The Galaxy FDD
framework automatically collects and archives data from
control systems. It analyzes faults and events using
a rule based expert system and can manage defned
performance metrics and KPIs such as production and
consumption, using Gbots for fne tuning and continuous
commissioning based on the analysis.
A rule-based system, consisting of a set of if-then
statements, forms the basis for “expert systems” which
are widely used in many felds. Expert systems aim to
capture the knowledge of an expert, which is encoded
into the rule set. When exposed to the same data, the
expert system artifcial intelligence (AI) will perform in a
similar manner to the expert.
The Galaxy Rule Engine offers a user interface with tools
for confguring rules easily and to manage and control
changes in the rule defnitions. It provides a means of
knowledge representation by translating the human
experts’ approach to identifying a fault into a declarative
programming language. It segregates data and logic, the
data comes from the domain object model of the FDD
system while the Rules are segregated into a central rules
defnition repository.
Users who need a detailed analysis of electricity
consumption can confgure reports to monitor contracts,
available capacity, maximum demand and load factor.
They can include base load reports, league tables based
on a number of different key performance indicators
and profle reports. The energy analysis capabilities are
discussed in more detail in the brochure on the Galaxy
M2M Cloud Services for Energy Optimization.
6.3 Expert Rule Engine for Fault Detection
and Diagnosis
The increasing complexity of systems of all types, as
levels of digital control reach new heights, creates a need
for smarter fault detection and diagnostics. This means
reducing dependence on human input to minimize errors
in analyzing faults and ensure that no abnormal equipment
behavior goes unnoticed. In order to do this the FDD
system must ‘learn’ the analysis of the fault and store it as
a ‘fault signature’ that it will recognize, should the fault
re-occur. Pacifc Controls’ Galaxy GCCC includes an
FDD component designed to do this.
GALAXY OPEN PLATFORM FOR M2M ENTERPRISE CITY MANAGEMENT - 37