TOWARDS PERSONALIZED TRAINING OF ELECTRIC POWER

GENERATION OPERATORS

Ricardo Molina

, Guillermo Rodriguez

, Yasmin Hernandez

, Israel Paredes

Gustavo Arroyo

and Liliana Argotte

Instituto de Investigaciones Electricas, Department of information Technologies, Reforma 113,

Cuernavaca Morelos, 62490, Mexico

Comisión Federal de Electricidad (CFE), Subdirección de Generación, Mexico DF., Mexico

Keywords: Personalized Training, e-learning, Labor Competences.

Abstract: This paper presents an approach to personalize the training of operators of power electricity generation.

Although the approach is described in the context of electric power generation, it can be adapted to other

cognitive training environments. The training courses are personalized on the basis of job position and the

operator‘s training history. The personalized training is complemented with the certification of labor

competences.

1 INTRODUCTION

The Web offers the perfect technology and

environment for personalized learning where

learners can be uniquely identified, content can be

specifically presented, and progress can be

individually monitored, supported, and assessed.

(Martinez 2001).

2 COMPLEX DOMAIN

LEARNING

What one regards as simple or complex is somewhat

dependent on the individual making the judgment

[Adelsberger 2008]. Besides individual differences

in terms of prior knowledge and experience,

however, there are some particular characteristics of

complex domain problem and problem-solving skills

associated with the situation. A complex cognitive

skill is one that consists of multiple constituent

skills; some of them involve thoughtful processing.

Multiple measures and reflective treatment can be

found in many work situations and problems in the

operation and maintenance of equipment for

generating electricity.

2.1 Cognitive and Psychomotor

Learning

Currently, electric power generation operators (like

other operators of complex equipment as airplanes,

helicopters, etc.), are trained, first, with theoretical

courses to gain cognitive knowledge. Then, to

acquire psychomotor experience they are sent to

power plant simulators and after that they practice in

the actual power plants supervised by a human tutor,

the trainees become real operators. The psychomotor

training is by definition personalized.

In this paper we describe the personalization of

cognitive knowledge courses for power generation

plant operators.

3 INTELLIGENT

ENVIRONMENT

ARCHITECTURE FOR

PERSONALIZED TRAINING

Our proposed architecture of an intelligent

environment is based on dynamic course generating

systems proposed by [Brusilovsky 2003]. The

intelligent environment is composed of four main

components (see Figure 1): the domain knowledge

module, the tutor, the operator model, and the

Learning Management System (LMS).

372

Molina R., Rodriguez G., Hernandez Y., Paredes I., Arroyo G. and Argotte L..

TOWARDS PERSONALIZED TRAINING OF ELECTRIC POWER GENERATION OPERATORS.

DOI: 10.5220/0003926603720375

In Proceedings of the 4th International Conference on Computer Supported Education (CSEDU-2012), pages 372-375

ISBN: 978-989-8565-06-8

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

Figure 1: Architecture of the intelligent environment.

4 THE DOMAIN KNOWLEDGE

MODULE

The domain knowledge module has three main

components: the concept structure map, the editor

and the repository of learning objects.

4.1 The Concept Structure Map

The concept structure contains the concept/topic

structure of the subject knowledge to be taught. It is

possible to organize the domain concepts/topics into

a set of smaller, possibly interrelated AND/OR

graphs, representing relatively independent sub-

areas of the knowledge, different views, or different

levels of granularity. It is represented as an

AND/OR graph, where nodes represent the concepts

domain or elements of knowledge, such as electrical

topics, components of control board, rules,

procedures and so on; and arcs represent

relationships between concepts, such as a

prerequisite for learning a concept or a sequence.

Every node is associated with a set of teaching and

testing materials labeled as Reusable Learning

Object (RLO), which instantiate different ways to

teach the concept/topic (e.g. introduce, explain, give

an example, and give a simulation, exercise, or test).

For the training of power plant operators, the

concept structure map is made based on the

structural decomposition of the generation process

into unit, structure (boiler, condenser, turbine,

generator, etc.), systems (air-gas, water-steam, fuel-

oil, etc.), equipment (PI Control, etc.) and

component (pump, valve, pipe, thermometer, etc.).

4.2 Editor

The editor contains tools for edition of teaching and

testing materials based on learning objects. Each

material is labeled as Reusable Learning Object

(RLO) or Shared Content Object (SCO) according

with the SCORM (Sharable Content Object

Reference Model) terminology [ADL 2001].

Only for the generation process, CFE has a

collection of more than 400 instructional courses

developed in house during the last 10 years.

Figure 2: SCORM complaint learning objects (SLO).

The editor allows the repository manager to take

learning content, generate a SCORM 2004

compliant label o manifest and to compress all

together in a zip file (see Figure 2). The result is a

SCORM compliant Learning Object (SLO) that can

SCO

Html, WAP, GIF,

JScript, VRML, etc.

Label or

Manifest

(XML)

TOWARDSPERSONALIZEDTRAININGOFELECTRICPOWERGENERATIONOPERATORS

373

be managed by any SCORM compliant LMS.

The SCORM compliant manifest contains

metadata for the classification and recovery of the

SLO. The classification categories have been

extended to satisfy the needs of CFE, some were

inspired from concept structure map.

The SLO is a set of items to structure a course,

workshop, or other aggregation of learning

resources, the organization of the SLO is as shown

in Figure 3.

Figure 3: The organization of a SLO.

4.3 Learning Object Repository

The Domain Knowledge module has a Learning

Object Repository (LOR). The LOR is a central

database in which learning content (SLO) is stored

and managed. The Repository main component is

the database.

The repository is implemented using a relational

database management system and an abstraction

model of the database using a semantic network is

shown in Figure 4.

Figure 4: The semantic network for the database structure.

Other authors [Gascueña 2005] use ontologies to

model the database but in this case a semantic net is

simpler and is a closer representation of the

relational database.

5 THE COURSE PLANNER

MODULE

First we present the basic approach and then the

competences are integrated.

5.1 Basic Course Planner

Figure 5 shows a semantic network that represents

the data used personalized courses.

Figure 5: Data used to personalized courses.

5.2 Integration of Competences

The objective is that the traditional training model

supports to the employees' labor skills certification

without affecting the training contractual rights.

The approach followed to integrate the labor

skills concept with personalized training is centered

in the concept of thematic content that the training

course programs will have to include to support the

employee in the labor skills certification process.

The idea is to establish and achieve thematic

consistency between the elements of the labor skills

norms of the key functions and the specialty courses

of the employee position profiles.

The integration process includes the set of

specialties (SP1, SP2,... SPn) obtained from the

position profiles of the productive organizational

functions and the configuration of specialties based

on skills or competences (c1, c2, … cn) achieved by

correlating the thematic content of the skills norm

with the content of the specialties courses and

adapting them or creating new contents to impact the

competence certification.

In this fashion, a specialty is a group of

competences, Sp1 = C1 + C2 +. + Cn, (see Figure

6). The specialties can be classified in more than one

specialization level, where the highest level contains

the lower levels.

As an example of a competence oriented analysis

the design of a master degree curriculum in power

plant operation is briefly described. Table 1 shows a

5 semester master in engineering curriculum where

the courses in shady background represent

competence oriented courses with thematic contents

CSEDU2012-4thInternationalConferenceonComputerSupportedEducation

374

Figure 6: Competence oriented specialty courses.

matched to the thematic contents of labor skills. The

first generation of 15 employees graduated at the

end of the spring of 2008.

In Figure 7 the dimensions used in CFE to

classify the training levels are shown. The training

levels include the organic positions levels of the

employees, the competence or skills levels and the

levels of specialty to have the appropriate

qualification to his position profile, competences and

specialties.

6 CONCLUSIONS

This paper described an approach to personalize the

training of operators of power electricity generation.

Although the approach is applied in the context of

electric power generation, it can be adapted to other

cognitive training environments. The training

courses are personalized on the basis of job position

and the operator‘s training history. The personalized

training is complemented with the certification of

labor competences.

Table 1: Competence oriented master in power plant

operation specialty courses.

1 2 3 4 5

Thesis

Seminar

Thesis

Seminar

Thesis

Seminar

Project

Thesis

Seminar

Diagnosis

Evaluation

Goal

Negotiatin

Controllrs

Simulatin

Agent

Operation

Recovry

Strategs

Energy

Balance

Tactic

Projects

Performn

Measurm

Thermo

Economic

Performn

Tests

Fuel

Consumpt

Problem

Identificat

Plant

Modificati

Thermo

Fluids

Combusti

Start Up

Maintenan

Planning

Performan

Evaluation

Correctiv

Actions

Communc

Techniqs

Tests

Coordinati

Maintenan

Execution

Informatin

Managem

Preventiv

Actions

Knowlede

Managem

Tests

Planning

Maintenan

Results

Informatin

Systems

Failure

Analysis

Figure 7: The training classifying dimensions.

REFERENCES

Adelsberger, H. H. (ed), “Handbook on Information

Technologies for Education and Training”, Second

edition, 2008, Springer-Verlag Heidelberg

ADL Sharable Content Object Reference Model Version

1.2: The SCORM Overview. Advanced Distributed

Learning (2001), http://www.adlnet.org

Brusilovsky, P., Vassileva, J.: Course sequencing

techniques for large-scale web based education. Int.

Journal Cont. Engineering Education and Lifelong

Learning 13(1/2), 75–94 (2003)

Martinez, M., Foundations for Personalized Web Learning

Environments, Journal of Asychronous Learning

Networks, 4(2), 2001

Gascueña, J. M. et al., “Ontologies for student and domain

models in adaptive and collaborative Learning system”

Advances in Artificial intelligence Theory, Research

on Computing Science 16, 2005, pp. 33-42

TOWARDSPERSONALIZEDTRAININGOFELECTRICPOWERGENERATIONOPERATORS

375