Abstract

The present electric power scenario in Brazil and abroad has led companies in the industry to operate in a highly competitive context, forcing those companies to constantly strive for better efficiency, higher quality supply and lower costs. Following this general trend, there has been a migration from preventive  to predictive maintenance, which is only possible through online monitoring systems.

In these circumstances, circuit breakers, the equipment responsible for electric  triggering of transmission lines, transformers, reactors and other equipment in substations are also being monitored.

The circuit breaker monitoring system allows, either locally or remotely, and in real time, providing accurate information to maintenance personnel about circuit breaker status, as well as minimizing failure risks and maintenance costs.

A practical analysis of the system performance is reported in this paper, which explains diagnosis types, their goals and the types of failures they can prevent, as well as how they can help to diagnose the cause of sudden failure in case it occurs. Therefore, future preventive measures can be used. Following this reasoning, monitoring can show a new range of approaches to problems which were unknown before, or were taxonomically treated as the same type of failure.

For a long time, equipment integrity at the substations and plants depended on preventive maintenance. Currently energy companies and industries coexist with reduced maintenance engineering teams, and it becomes necessary to develop tools to aid in the decision-making process for an accurate intervention.

One of these tools is the implementation of a predictive maintenance program in high voltage equipment based on these equipments’ operating condition (1). In this paradigm, circuit breakers are very important, since they are responsible for the safety of the electrical system if a failure occurs. In addition to that, they are important as maneuvering devices.

The currently adopted maintenance practices are based on fixed time intervals or on the number of performed maneuvers, indicated through the operation counters (2) . Maintenance, formerly based in time, now takes into consideration the equipment status. The emergence of systems that allow the organization, storage, visualization and interpretation of data which are important for maintenance made this paradigm shift possible. These systems allow safe equipment operation and provide the necessary elements for the current maintenance demands based on status.

Table 1 provides a brief comparison between maintenance philosophies based on time and status.

Comparison between monitoring philosophies

Just as an example, some signals that can be monitored to detect malfunction of circuit breaker components, such as opening and closing times, SF6 gas pressure, pump operation time, were cited in the literature  (3). By analyzing these signals and comparing the range of values specified by the manufacturers, one can detect incipient faults in the circuit breaker and classify them as mechanical, located at the main electric circuit or control and auxiliary circuit.

This is the idea behind the diagnostic system based on rules.

Circuit breakers are electrical switching devices which should be able to conduct, stop and establish currents from the electric power grids, including high fault current (short circuit)  (4) .

They consist of interrupting units, also called extinction chambers, which is where the voltaic arc created during the current interruption, and can be of various technologies. They have an operating or driving system which is the subset consisting of components that allow us to store the energy needed for the circuit breaker mechanical operation, as well as the release of energy through appropriate mechanisms, when a command is sent to open and close it; and last, but not least, they are a command unit, made of the circuit breaker command, control and supervision elements.

With the purpose of identifying the most common causes of failures in circuit breakers, some research was conducted by CIGRE  (5) . At first, it was conducted between the years 1974 and 1977 using a database from observations of 20,000 circuit breakers with a voltage above 63 kV.

The results of this survey showed that 70% of the most frequent failures in circuit breakers are caused by mechanical factors, 19% are connected to control and auxiliary circuits and 11% are due to main circuit electric factors.

The summary results of the second survey conducted by CIGRE in the period from 1988 to 1991 at 18,000 SF6 circuit breakers for voltages above 63 kV are presented in Table 2, published in the IEEE Std.C37.10 IEEE-1995 guide  (6) and transcribed here.

It is a procedure to investigate diagnoses and failures in circuit breakers, giving an overview of the components in the circuit breakers which are frequently responsible for failures.

Statistics of circuit breaker failure causes (6)

The results obtained in this research allows us to conclude that most circuit breaker failures are associated to the components of the circuit breaker operating mechanism; the second most frequent cause are high-voltage components and the third most frequent cause is connected to control circuits  (7) .

The goal of online monitoringn is to prevent issues arising from points listed in Table 2 .

The Circuit Breaker Monitoring System installed in the Rio Verde Substation is a tool for engineering maintenance that acquires, stores and processes measurements taken from the monitored equipment, generating diagnosis and prognosis of potential problems that could lead to interruptions in electricity supply  (8).

Among the many findings of this monitoring system, we cite those considered most important in this project:

• Rapid diagnosis of the current state of the equipment in order to provide input for decision-making regarding the operation or not of the circuit breaker, thus helping to preserve not only the circuit breaker but also the equipment it is powering.
• The early prognosis of fault conditions before their operation, in order to increase system availability and allow equipment downtime only when needed for corrective action or not include it in the system to prevent damaging it.
• Monitoring equipment operating conditions along its whole life so that equipment is available for operation, in addition to controlling equipment at the end of its service life.
• Quick access to status information from the equipment through the Furnas intranet thus facilitating the access to all information on the equipment.
• Possibility of integration with other corporate monitoring and maintenance management equipment.
• Integration of maintenance management and cost reduction.
• Monitoring of several types of circuit breakers.
• Investment management and insurance premium negotiation.
• Corporate image preservation.

The architecture of the sensor network is shown in Figure 1 :

The main variables collected by the monitoring system sensors are Table 3 :

Key variables obtained by the monitoring system (8)

As can be seen in Figure 1, the entire system, to be functional, requires successive stages in the acquisition of the data so that they can be used in monitoring rules.

• Step 1: Field sensing is responsible for acquiring all variables in the circuit breaker monitoring system, during opening and closing operations, spring loading, motor and variables that may be obtained without the operation, for instance, SF₆ gas pressure.
• Step 2: The data obtained are transferred to the software located on the monitoring server where all calculations are made so that this information is displayed to the user. All these data are recorded in SQL Server installed in the monitoring server after this step.
• Step 3: This OLM Explorer SQL Server database is read by the circuit breaker monitoring system, each time a new record is included, thus making all comparisons and calculations needed so the result of monitoring rules is displayed to the users.

After this step, if an alarm is detected, the information will be displayed on the monitoring system website and an e-mail alert will be received by the maintenance team.

The rule system encoded into the monitoring software allows information obtained by the sensors to be transmitted into automatic diagnoses, indicating the kind of problem to the maintenance team, as well as its severity and the maximum time for intervention, if this situation applies.

In Table 4 : Main diagnostic rules used by the monitoring system  (8) , the the diagnostic rules and their main function are displayed: