As part of a program to provide an infrastructure for transmission line and substation monitoring, EPRI in the U.S. has developed remote sensors to monitor the condition of important assets. This INMR article from 2015, contributed by Andrew J. Phillips and Chris S. Engelbrecht of EPRI USA and M. Major of SWRI reviewed this program and its results.
The main goal behind this program was to obtain a cost effective platform that enables installation of numerous sensors over a transmission system. These sensors would provide for increased condition awareness, resulting in improved reliability as well as reduced maintenance costs. All sensors are communicated to with the same radio frequency (RF) system that collects data from sensors and integrates it with Utility Asset Health Systems. An overview of the sensor package demonstrated at over a dozen utilities is shown in Fig. 1.
Sensors under development and commercialization include:
• Conductor sensors to monitor temperature, conductor movement & vibrations;
• Structure motion & vibration sensors;
• A ground wire sensor to detect lightning & fault current;
• A sensor to monitor geo-magnetically induced currents;
• A radio frequency interference sensor to detect arcing or other discharges at substations or on lines;
• An arrester condition monitoring sensor;
• A disconnect sensor that monitors its temperature, current & inclination of switch arm;
• Physical security sensor;
• A bushing sensor;
• Underground cable sensors (5 different types).
The sensor suite also includes an insulator leakage current sensor, designed as a robust, low-cost device to allow widespread application on lines and at substations to monitor insulator performance under polluted conditions.
Monitoring Leakage Current
Insulator leakage current measurement can be done with various degrees of sophistication, including:
• Capturing detailed oscillograms of leakage current surges, or bursts, and/or the supply voltage (as in Fig. 2);
• A date and time stamped array of peak leakage current amplitude measured over a given time period (as per Fig. 3);
• Binned leakage current peak values over an extended time period (as shown in Fig. 4). It should be noted that it is not possible to determine exactly when the highest leakage current pulse occurred with this methodology unless data retrieval is very frequent. Retrieval rate is every 5 months in the example shown.
• Time stamped binned leakage current peak values. In this case the binned current peak values are retrieved from the leakage current monitor at relatively short time intervals (e.g. 10 minutes). This way, both time and severity information is obtained about leakage current events, which can also be correlated with available weather data. An example of such time stamped binned leakage current data is presented in Fig. 5.
Evaluation of Results
Results need to be evaluated in order to obtain an indication of site severity or to estimate flashover risk. Level of possible data analysis depends on how many parameters are captured. For example, multivariate analysis techniques can be used to better understand the relationship between environmental parameters and leakage current when time-stamped leakage current and weather parameters are logged. However, there are limitations in such an analysis since level of leakage current is not only a function of the environment at that time instant but also of weather conditions preceding the event. For example, contamination accumulating over a long dry period before a wetting event will result in higher levels of leakage current than when the time of measuring is preceded by a rainy period. Nevertheless, leakage current as a fundamental parameter to characterize contamination performance of insulators can still provide useful information for:
1. Pollution site severity measurement
Information gathered enables selection of insulator designs and dimensions with respect to contamination performance. For this purpose, leakage current activity on the insulator should be correlated with that during artificial contamination tests under known pollution severity.
2. Initiator for insulator maintenance
Increased leakage current activity on insulators can be used to trigger insulator maintenance before critical conditions arise.
3. Insulator characterization
Leakage current can be used to compare performance of different types of insulators installed at the same testing site and determine the most appropriate design and dimensions for those conditions.
4. Condition of RTV coatings & ageing of polymeric insulators
Measuring Site Pollution Severity
Determining site severity through leakage current measurement is described in the latest version of IEC 60815-1 (illustrated in Fig. 6) and is comprised of the following steps:
1. The relationship of leakage current to contamination severity is determined through laboratory tests on the insulator of interest. This has traditionally been done with the salt-fog test, considered representative of coastal conditions across Europe;
2. On site, leakage current is monitored across the insulator type for which this leakage current relationship has been determined;
3. Monitoring of leakage current continues for a considerable time period, i.e. at least a year;
4. Maximum current measured over the monitoring period is converted to equivalent site severity with the relationship determined in the first step.
Monitoring leakage current can also be an early warning system to initiate maintenance on ceramic or glass insulators, e.g. cleaning. Leakage current activity across an insulator increases as it becomes more contaminated and higher contamination levels mean greater flashover risk. As such, increasing leakage current activity indicates higher risk of flashover, as illustrated in Fig. 7.
The curve showing probability of flashover for any given leakage current can also be derived from laboratory test results. In this EPRI implementation, both peak leakage current and also more complex algorithms based on duration and magnitude of the event are used to generate alarms.
Leakage current measurements are often used as an objective way to quantify and compare discharge activity on different types of insulators exposed to the same contaminated conditions. Comparing absolute levels of leakage current peaks allows an impression of the relative flashover performance of insulators whereas other current parameters, e.g. cumulative charge, are often associated with erosion stress a polymeric material is subjected to (i.e. ageing performance).
EPRI has developed a robust, low cost sensor for monitoring leakage currents across insulators. These self-contained units report measured data and status via radio frequency link to a local base station where data is integrated into an asset health system. The sensor has the following features:
• Current measurement is done using a ferrous current transformer surrounding a straight current conductor as primary winding. The saturation characteristics of the magnetic circuit and low impedance of the primary winding protects the internal electronics of the sensor from overvoltages;
• The device stores the statistical parameters of leakage current peaks that occur. Leakage current is monitored continuously and the peak value over a 1-minute period is binned and stored. Amplitude of the highest peak current occurring during the last measuring cycle is also reported. The device has 6 bins for which leakage current threshold can be changed wirelessly at any time;
• It keeps track of the time since the last reset, thereby limiting the data message sent by the device to a bare minimum. It also limits internal circuit complexity and power consumption for the device.
• The communication system of the sensors is based on the IEEE 805.15.4 architecture but with a customized communication protocol. The communication protocol allows two-way communications enabling the user to reset the bins or change bin levels remotely;
• The electronics are embedded in potting compound and housed in a metallic enclosure in order to protect internal components from environmental as well as electromagnetic influences;
• Power is supplied by on-board, long life batteries. Based on battery characteristic and low power consumption of the unit, battery life is estimated at over 12 years.
EPRI has two standard settings for leakage current threshold for the bins. These bin ranges are programmable and can be adjusted (within limits) remotely while the sensor is in service
EPRI sensors utilize a current pick-up (guard electrode) on the insulator being monitored that is isolated from ground to facilitate measurement. For disk insulators, the guard electrode is in the form of a special clip-on arrangement while, on substation or post insulators, a simple metal band is used. Typical installations are illustrated in Fig. 8.
There are already over 150 sensors installed at a range of sites worldwide. Some have been operational for 5 years and continue to provide data to act upon. Examples of installation locations are shown in Fig. 9.
Sensor Operations Overview
There are two basic applications for obtaining data captured by the leakage current monitors:
1. Data is downloaded during periodic rounds inspection through a portable interrogator with an antenna. Inspection period is not pre-determined and could vary according to need or opportunity. Sensors and communication are designed so that data can be collected via fly-by or drive-by inspections at up to 100 km/h.
2. Sensors are continuously interrogated via a dedicated sensor base station. In this mode, it is possible to resolve the times when leakage current events occur. With this operation strategy, automatic alarms can be triggered to indicate need for insulator cleaning or times when there is significant risk for flashover.
Basic architecture of the dedicated leakage current system is shown in Fig. 9. There are 3 basic components to the system for data collection, storage and presentation:
1. The wireless sensors networking protocol uses the IEEE 802.14.5 standard; operating in the 2.45 GHz frequency band. Each sensor in the wireless network can be configured to broadcast at set intervals ranging from 5 to 120 seconds. All data packets transmitted by the sensors are collected by the ZAP (Wireless Access Point), which reassembles and stores the sensor data in memory. Bin levels on the sensors can also be wirelessly reset and calibrated.
2. Sensor Base Station: The base station is the main on-site data collection point. Each base station is equipped with a ZAP, data logger, cell modem, battery/solar panel, and peripheral sensors. The base station consists of the following components with the functions described:
• ZAP(Wireless Access Point): Collect and store data from EPRI wireless sensors.
• The ZAP can be integrated with utility asset health systems directly, using Modbus or DNP3, or integrated into the EPRI Research Sensor Network, as described below.
• Data logger: Interrogates the ZAP for wireless sensor data (via RS232 modbus), and collects data from peripheral sensors such as wind, temperature/humidity, and door alarm. All collected data are stored in a circular buffer.
• Cell Modem: Allows the EPRI VDV server to pull data from the data logger- Battery/Solar Panel: Main power source of the base station. Without sunlight, the base station will remain operational for up to 7 days.
3. EPRI VDV Server: For the EPRI Research Sensor Network all base stations in the field report back to a central data server. This server is configured to pull data from the field at fixed intervals throughout the day (if necessary it also perform post processing on the data). Data collected are then stored in a MySQL database. A web interface is also available for accessing and viewing the collected data. Additional password protection and access control are available to only allow certain users to access the data stored on the server. Email alarms can be generated and sent to both utility and EPRI personnel.
Data is made accessible to users via an automatically generated web page, shown in Fig. 10. Leakage current activity is shown in 3 formats: 1. development of bin counts over time (top graph); 2. highest current measured during latest monitoring interval (middle graph), and 3. algorithms that have been developed to trigger alarms (bottom graph). A summary of sensor locations and visible alarms with important weather parameters are shown on a dashboard (as in Fig. 11).
EPRI’s approach to wireless leakage current monitors sees leakage current on insulators monitored using a robust, low-installed-cost device that allows widespread applications to monitor insulator performance under polluted conditions. The sensor has been deployed at a number of sites to demonstrate its utility and robustness. For utilities, the goals are obtaining information on when to conduct insulator washing and collecting information for dimensioning insulation with respect to pollution. A key aspect of the EPRI system is that the leakage current sensor is just one of a range of sensors that have been developed to address a range of transmission assets and issues so as to form a suite. New sensors for the suite are continually being developed and commercialized based on industry needs. Basic development of the EPRI leakage current sensor concept is now final and sensors are now commercially available.