Page 53: of Maritime Reporter Magazine (December 2001)

Ship's Store

Lemag Premet

The Lehmann & Michels permanently mounted LEMAG PREMET Online mea- sures the cylinder pressure continuously, from each cylinder at the same time. Once installed, you receive continually indicated power, max, cylinder pressure, main trend etc. on your PC.

Circle No. 122 www. maritimere- porterinfo.com

Sure Seal

Sure Seal connectors are rugged, low-cost and environmentally sealed. The free guide is a complete design resource providing applica- tions overview, technical specifications, detailed test data, dimensions, and assembly instructions. If you need an inexpensive sealed connector Sure

Seal is your solution.

Circle No. Ill www.maritimere- porterinfo.com

Dickow Pump Company-ARI

Valve Corporation

Dickow Pump Company has manufactured centrifugal pumps for more than 75 years, always with an emphasis on precision, longevity in service, and hydraulic efficiency.

Today we offer the broadest and most technologically advanced range of magnet drive pumps, and have solid in-plant experience. Dickow

Pumps are engineered products for optimum performance in your specific application.

Circle No. 112 www.maritimereporterinfo.com

Your M.irinc U OffsflOMV

Signage Experts.; ;

Maritime Associates

Maritime Associates is your marine and off- shore signage expert, manufacturing a com- plete range of IMO/SOLAS safety signs, posters and LLL pathway systems. Our vast capa- bilities extend beyond conventional safety sign requests. We can create and manufacture any sign and sign system required, utilizing an array of materials includ- ing our New product lines and unique base materials, mounting and installation methods all cost effectively manufactured in the USA.

Circle No. 105 www.maritimereporterinfo.com 3M Air-Mate 3M Occupational Health & Environmental

Safety Division introduced the Air-Mate

Combination Escape Self-Contained Breath- ing Apparatus (ESCBA)/Supplied Air Respira- tor (SAR). The NIOSH-approved system fea- tures a lightweight design and interchange- able five- and 15- minute cylinders that allow for flexibility in different applications.

An airline connection enables users to remain in hazardous environ- ments for extended peri- ods of time.

Circle No. 197 www.maritimereporterinfo.com

Lafarge POUnd FOf ihe product is a flow- PrtlinH able, water-retentive ^H^.Q^g substance consisting of inorganic, non-toxic, 15 BOTlOi granular and ground fines blended with water • and appears as a flow- able mortar during installation. However, the comparison with

Portland cement based products is limited to the common usage of mixing plants, delivery vehicles and pumping equipment. Once installed it firms up to a semi-solid mass which although stiff it has little (0.5 mpa) to no compressive strength. This feature means that the material can be removed for repair or modification. Circle No. 114 www.maritimereporterinfo.com

Airmar

Airmar Technology Corp. introduced a line of

American-made bronze through-hull trans- ducers/sensors with a long-stem design appro- priate for cored fiberglass hulls, thick-hulled wooden boats, steep deadrise hulls and other vessels. Airmar's new B124 transducer fea- tures a low-pro- file design that extends only 3/16 in. outside the hull, minimizing drag and providing a smooth surface for water to flow across the acoustic element, improving both boat and sounder performance at speed.

Circle No. 196 www.maritimereporterinfo.com .

Sigma Coatings

SigmaPrime is a high quality epoxy primer specifically designed to fit shipyard building practices while offering excellent long-term corro- sion protection.

SigmaPrime is the latest in a series of ground-breaking products from

Sigma Coatings Marine who are continuously developing and improving the way coating systems build-in effective vessel protection.

Circle No. 135 www.maritimereporterinfo.com , (Continued from page 47) by one of DEXTER's diagnostic agents.

When specifying symptoms in BRAINS, a list of available tag names and their descriptions are obtained directly through the interface between DEXTER and a user's automation sys- tem. For example, if the automation software defines a

Process Database containing all measured sensor inputs in the plant, DEXTER interrogates the automation software for this list. A user can then simply pick from a drop-down list of data points when building a diagnostic. This makes it very easy for a user to integrate DEXTER into plant automation.

The various software agents within DEXTER are "knowl- edgebase-centric". This means that each agent is linked to a specific knowledgebase. The knowledgebase defines both the data source and the specific data points that an agent will monitor. Because DEXTER agents are knowledgebase-cen- tric, the amount of setup information that you must specify to configure an agent is kept to a minimum. A user simply select a knowledgebase to be used by your agent and it then knows exactly which set of data points to monitor.

Before a user builds any knowledgebases, a user must first configure DEXTER to work with a specific a real-time data source. DEXTER is designed for interoperability with most of the major process control software packages on the market, such as Intellution FIX, Wonderware, Rockwell Automation,

National Instruments, etc. All of these software packages have a mechanism for storing real-time sensor data in a data- base. BRAINS will automatically extract a list of all data points defined in the process control software database. The user will then be able to select data points from this list when the faults and symptoms for a knowledgebase are entered.

A user can create multiple knowledgebases using BRAINS, storing each one under a different name. Each knowledgebase can pertain to a separate machinery plant, specific system within a plant, or even an individual piece of equipment. A

December, 2001 user has complete flexibility in how knowledgebases are defined and used. A user should consider what types and how many agents are desired when creating knowledgebases.

Transforming Knowledge into Artificial Intelligence

A major concern in deploying software agents for diagnos- tics and prognostics is the robustness of their artificial rea- soning with respect to correctly identifying real problems when they occur. Missing, noisy, or corrupted sensor data, which are all common real-world occurrences, must be toler- ated and not mistaken as equipment faults. Faulty sensor data introduces uncertainty into the diagnostic inferencing process. The reasoning technique should handle such uncer- tainties in some statistically valid way. The diagnostic agent's robustness can directly impact maintenance and repair costs.

Robustness can be quantified by the accuracy of the diagnos- tic call. An incorrect diagnosis is declaring a fault different from the one actually present. A missed diagnosis is declar- ing that nothing is wrong, when, in fact, one or more faults exist. A false alarm involves declaring a fault when there is none. Each of these diagnostic conditions can lead to unnec- essary expenditures of maintenance resources and/or reduc- tions in plant reliability, not to mention loss of faith in the diagnostic system. DEXTER's goal is to minimize the proba- bility of each of these cases and to maximize the probability of a correct diagnosis.

DEXTER uses probabilistic neural networks for its diag- nostic and prognostic reasoning about machinery faults.

DEXTER's neural networks automatically learn to associate patterns of alarm conditions with the machinery faults you enter into your knowledgebases. DEXTER's neural network learning occurs instantaneously, as compared to other neural network techniques, allowing you to rapidly build, modify, and deploy diagnostic agents on the factory floor. This allows you to immediately put agents to work, without any pro- gramming. DEXTER agent characters are driven by

Microsoft Agent software.

The four main types of intelligent software agents avail- able in DEXTER are:

Alarm Detection Agent - These agents simply perform an alarm monitoring function for all data points associat- ed with a knowledgebase. Alarms are detected when a monitored data value exceeds it alarms thresholds. The agent character will appear on your screen to alert you and provide you with a list of alarms.

Trend Detection Agent - These agents perform an automatic trending analysis of historical data pertaining to all data points in a knowledgebase. A regression analy- sis is performed to detect any statistically significant trends developing in your machinery plant performance.

Developing equipment problems can often be uncovered by degrading performance trends shown in the mea- sured data. This can help avert unexpected failures. The agent character will appear on your computer screen when it detects any significant trends. It will give you a list of trends and allow you to view a trend graph of the data histories.

Diagnostic Agent - These agents perform an alarm detection function similar to the Alarm Detection Agent, but take the analysis a step further by diagnosing possi- ble machinery faults based on the detected alarms. Diag- nostic reasoning is based on a neural network that has been trained from the knowledgebase to which the agent is attached. Once trained, the neural network is able to recognize alarm patterns and their associated machinery faults. The agent compares the alarms it detects from the plant automation to those it has been trained to rec- ognize and produces a set of diagnostics.

Prognostic Agent - These agents perform a trend detection function similar to the Trend Detection Agent, but take the analysis a step further by predicting future machinery faults. The trending analysis performed by this agent will identify any statistically significant trends in machinery performance. These valid trends are extrapo- lated into the future to predict future alarm conditions. If the performance trends continue, i.e. no corrective main- tenance actions are taken to alter degrading perfor- mance trends, then the agent will generate a set of pre- dicted alarm conditions. The agent will then use these predicted alarms as input to its prognostic reasoning.

Prognostics are also based on the neural network asso- ciated with the knowledgebase to which the agent is attached. The agent character will appear on your screen if it predicts any future machinery faults above a given probability level, which you specify when you configure the agent.