1、 g49g50g3g38g50g51g60g44g49g42g3g58g44g55g43g50g56g55g3g37g54g44g3g51g40g53g48g44g54g54g44g50g49g3g40g59g38g40g51g55g3g36g54g3g51g40g53g48g44g55g55g40g39g3g37g60g3g38g50g51g60g53g44g42g43g55g3g47g36g58processing, communication and presentation Part 2: Data processing ICS 17.160; 35.240.99Condition m
2、onitoring and diagnostics of machines Data BRITISH STANDARDBS ISO 13374-2:2007BS ISO 13374-2:2007This British Standard was published under the authority of the Standards Policy and Strategy Committee on 31 October 2007 BSI 2007ISBN 978 0 580 55407 0Amendments issued since publicationAmd. No. Date Co
3、mmentscontract. Users are responsible for its correct application.Compliance with a British Standard cannot confer immunity from legal obligations.National forewordThis British Standard is the UK implementation of ISO 13374-2:2007.The UK participation in its preparation was entrusted by Technical Co
4、mmittee GME/21, Mechanical vibration, shock and condition monitoring, to Subcommittee GME/21/7, Condition monitoring.A list of organizations represented on this committee can be obtained on request to its secretary.This publication does not purport to include all the necessary provisions of a Refere
5、nce numberISO 13374-2:2007(E)INTERNATIONAL STANDARD ISO13374-2First edition2007-07-15Condition monitoring and diagnostics of machines Data processing, communication and presentation Part 2: Data processing Surveillance et diagnostic dtat des machines Traitement, change et prsentation des donnes Part
6、ie 2: Traitement des donnes BS ISO 13374-2:2007ii iiiContents Page Foreword iv Introduction v 1 Scope . 1 2 Normative references . 1 3 CM time-order/time-reference data, normally referenced with UTC and local time zone; data quality indicator (e.g. “bad”, “good”, “unknown”, “under review”, etc.). Ex
7、amples of digitized data include: floating point values for scalar data; magnitude and time series for dynamic data; thermal radiation data with digitized image for thermographic data; sample test results for lubricating fluid/air/water sample data. BS ISO 13374-2:200774.3 Data Manipulation (DM) blo
8、cks As detailed in Figure 4, the DM block processes the digital data from the DA block to convert it to a desired form which characterizes specific descriptors (features) of interest in the machine condition monitoring and diagnostic process. Often the functionality within this layer consists of som
9、e signal processing algorithms. Figure 4 Data Manipulation block This block may contain speciality processing functions such as Fast Fourier Transforms, wavelets or simple average values over a time interval. Examples of the descriptor outputs of the DM block include: extracted feature; conversion f
10、rom time domain to frequency domain and vice versa; calculated, non-interpretative values; virtual sensor (differential pressure from inlet and outlet pressures); integrating acceleration to velocity/double integration to displacement; filtering; normalization; time series including sample rate. BS
11、ISO 13374-2:20078 4.4 State Detection (SD) blocks As shown in Figure 5, the primary function of the SD block (sometimes referred to as “state awareness”) is to compare DM and/or DA outputs against expected baseline profile values or operational limits, in order to generate enumerated state indicator
12、s with respective boundary exceedances. The SD block generates indicators which may be utilized by the Health Assessment block to generate alerts and alarms. When appropriate data are available, the SD block should generate assessments based on operational context, sensitive to the current operation
13、al state or operational environment. Figure 5 State Detection block Typically, this block of processing provides data which will contribute to a diagnosis in the health assessment block. The SD block may make use of current and historical DA and DM outputs to evaluate the current state. It may provi
14、de data manipulation and sensor module control signals, such as acquisition scheduling commands, data triggers and processing instructions. Examples of outputs of the SD block include: enumerated state indicator; threshold boundary alerts; severity of threshold boundary deviation above/below; rate o
15、f change alert; BS ISO 13374-2:20079 degree of abnormality; statistical analysis using parametric and non-parametric approaches, e.g. Weibull and Gaussian distribution. 4.5 Health Assessment (HA) blocks As shown in Figure 6, the HA block is an information block which utilizes expertise from a human
16、or automated agent to determine the current health of the equipment and to diagnose existing fault conditions. It determines the state of health and potential failures by fusing the outputs of the DA, DM, SD and other HA blocks. Figure 6 Health Assessment block An output of this block includes the c
17、omponent/systems current health grade and diagnosed faults and failures with associated likelihood probability. A calculation of the current risk priority number (RPN) may also be performed. Modelling of ambiguity groups and multiple hypotheses may be included in the output data structures. The HA b
18、lock may also output an explanation detailing the evidence for a diagnosis or health grade. BS ISO 13374-2:200710 4.6 Prognostic Assessment (PA) blocks As shown in Figure 7, the primary function of the PA block is to project the future state of the monitored equipment using a combination of prognost
19、ic models and their algorithms, including future operational usage model(s). This block determines the future state of health and failure modes by combining the relevant outputs of the DA, DM, SD, HA and other PA blocks and applying a prognostic algorithm or model based on supplied projected operati
20、onal utilization. To aid the algorithm or model, the HA block may also retrieve account historical failure data and operational history, along with projected failure rates related to operational utilization. The prognostics layer may report health grade at a future time or may estimate the remaining
21、 life of an asset given its projected usage profile. Assessments of future health or remaining life may also have an associated prognosis of the projected fault condition. A calculation of the future risk priority number (RPN) may also be performed. An output of this block includes the component/sys
22、tems future health grade and future failure events with associated likelihood probability. Modelling of ambiguity groups and multiple hypotheses may be included in the output data structures. The PA block may also output an explanation detailing the evidence for a proposed failure event or health gr
23、ade. Figure 7 Prognostic Assessment block BS ISO 13374-2:2007114.7 Advisory Generation (AG) blocks As detailed in Figure 8, the primary function of the AG block is to integrate information from DA, DM, SD, HA, PA and other AG blocks and external constraints (safety, environmental, budgetary, etc.),
24、and to provide optimized recommended actions and alternatives to applicable personnel or external systems. Recommendations may include prioritized operational and maintenance actions and capability forecast assessments or modifying operational profiles to allow mission completion. The decision suppo
25、rt module needs to take into account the operational history (including usage and maintenance), current and future mission profiles, high-level unit objectives and resource constraints. Maintenance advisories from this block should detail future maintenance work required, which may include the verif
26、ication of monitoring data or the performance of additional monitoring. The structure of these advisories should be put into a “work request” format for external maintenance work management systems. Based on this request, maintenance work management systems can schedule work in advance and locate sp
27、are parts and tools required for these jobs. Operational advisories from this block can be immediate in nature, such as the current notification of operators of alerts and resulting action steps. Other production-related advisories can be more strategic, such as sending a notice to a production plan
28、ning system about the high risk of failure on a production line due to a soon-to-fail critical piece of equipment. Capability forecast assessments from this block provide the results for requests about the likelihood of accomplishing a specific mission or production run. These assessments are critic
29、al to production forecasting systems when evaluating whether or not to accept certain missions/orders and where to assign the work, based on asset optimization principles. Figure 8 Advisory Generation block BS ISO 13374-2:200712 4.8 Block configuration Each data processing block requires configurati
30、on information, some of which may be static data, and other parameters may be changed dynamically by the system during operation. As an example, the following is a sample of the configuration of the Data Acquisition block: a) measurement location description (measurement location table) 1) orientati
31、on and relative position, 2) location description; b) monitoring intervals dynamic vs. static 1) on-line continuous, 2) on-line polled, default polling rate, default parameters; c) triggered vs. non-triggered 1) set points, 2) deadband; d) asynchronous vs. synchronous; e) transducer information 1) r
32、esponse curve, 2) measurement confidence, 3) transducer electronic data sheet (TEDS) information; f) calibration; g) channels 1) single or multiple channel collection. 4.9 External systems Retrieval of previous work histories from the maintenance system and previous operational data (starts/stops/lo
33、ads) from a process data historian is important in the assessment of machinery health. After a health assessment is made, the maintenance action to be taken can range from increasing the frequency of inspection, to repair or replacement of the damaged machinery or component. The effect on operations
34、 may be an adjustment of operating procedures or a request to shutdown the equipment immediately. This need for rapid communication to maintenance and operational systems requires software interfaces to maintenance management systems and operational control systems. These interfaces are useful in or
35、der to communicate recommended actions in the form of maintenance work requests and operational change requests. BS ISO 13374-2:2007134.10 Data archiving Data archiving is an important feature during all processing of a machine condition monitoring program. Previous data trends can be analysed for s
36、tatistical relevance. The data archiving system should provide rules for the archiving rate and amount of data stored. Previous advisories should be audited for accuracy and root cause information added upon its discovery. 4.11 Technical displays Relevant technical displays showing data from each bl
37、ock are necessary to facilitate analysis by qualified personnel. These displays should provide the analyst with the data required to identify, confirm or understand an abnormal state. 4.12 Information presentation Information from the HA, PA and AG blocks is displayed by this processing block. It is
38、 important that the data be converted to a form that clearly represents the information necessary to make corrective action decisions. In some cases, the user will need the ability to drill down into the SD, DM and DA technical displays when abnormalities are reported. 4.13 Compliant specifications
39、An open CM no interfaces are therefore defined in this part of ISO 13374. The first step was defining an object-oriented data model in Unified Modeling Language (UML) for each layer that was then converted into an abstract interface specification. The abstract specification can then be converted to
40、the desired middleware language for a specific interface definition. The UML object model defines interfaces only. For a given layer of the architecture, the data model does not prescribe the object classes that would be required for a software implementation. The focus is on describing the structur
41、e of the information that might be of interest to clients of that layer. OSA-CBM does not impose any requirements on the internal structure of compliant software modules. The architectural constraints are applied to the structure of the public interface and to the behaviour of the modules. This appr
42、oach allows complete encapsulation of proprietary algorithms and software design approaches within the software module. BS ISO 13374-2:200722 Annex B (informative) References to UML, XML and Middleware B.1 Purpose This annex provides a simple reference of definitions and glossary of terms on Unified
43、 Modelling Language (UML), eXtensible Markup Language (XML) and middleware services. In addition, a number of useful references have been provided along with pointers for simple tutorials and detailed tutorials on these topics. B.2 Unified Modeling Language (UML) B.2.1 Definition of UML The Unified
44、Modeling Language (UML) is a graphical language for visualizing, specifying, constructing and documenting the artifacts of a software system. UML offers a standard way to write a systems blueprints including conceptual entities, such as business processes and system functions, as well as concrete en
45、tities, such as programming language statements, database schemas and reusable software components27. B.2.2 Glossary of Terms Terms and their definitions have come from various sources, in particular from Reference 20. Activity A step or action within an Activity Diagram, which represents an action
46、takenby the system or by an actor. Activity Diagram A glorified flowchart that shows the steps and decisions and paralleloperations within a process, such as an algorithm or a business process. Actor A person or an external computer system that interacts with the softwareunder design. Association A
47、connection between two elements of a Model. This might represent amember variable in code, or the association between a personnel record andthe person it represents, or a relation between two categories of workers, orany similar relationship. By default, both elements in an Association are equal, an
48、d are aware of each other through the Association. An Associationcan also be a Navigable Association, meaning that the source end of theassociation is aware of the target end, but not vice versa. Association Class A Class that represents and adds information to the Association betweentwo other classes. Attribute A data field or property that represents information about a Classifier. Base Class A Class which defines Attributes and Operations that are inherited by a Subclass via a Generalization relationship. BS ISO 13374-2:2007
