|
||||||||||
Many static rating parameter sets are based on historical weather observations of ambient temperature and wind speed. A typical procedure includes selection of a "hot" temperature and a "slow" wind speed, a solar angle approximating maximum solar heating and an angle of wind impingement (often chosen perpendicular to the conductor). The resulting static thermal ratings are deemed conservative, although often without data to support the "conservative" hypothesis.
Two key factors have made the evaluation and justification of line rating parameters a more tractable problem than in the past:
- Improved data resolution, quality and availability
- Increased computational capability
- Weather parameter correlation and variability
- Rating verification/justification
Data Resolution, Quality and Availability
Throughout much of the 1980s and 1990s, the best and most complete data source for climatological data were National Weather Service (NWS) datasets that provided hourly data including temperature, wind speed, and wind direction. In the late 1990s, deployment of a variety of weather observation stations, often under the supervision of state climatological offices or agricultural extension services, led to the availability of additional weather datasets. The datasets arising from these systems are often richer than those previously available, often including multiple measurements within each hour, and including data that was previously unavailable. Of course, these datasets must be analyzed with caution, as they are not necessarily subject to the same degree of verification as NWS measurements.
Data Resolution: Data resolution improved dramatically with the availability of multiple observations within each hour. This increased resolution provides key information regarding the short-term variability of weather conditions; in particular, variation of wind speed and direction at a low wind speeds. For example, calm conditions experienced for a period of an hour may be of concern; however, calm conditions persisting for only one 15-minute interval within an hour may be of substantially lesser concern.
Data Quality: Advances in measurement technology have improved data quality by eliminating starting speed issues associated with mechanical anemometers used in earlier weather stations. Data from mechanical anemometers showed a disproportionate number of observations at very low wind speeds and required adjustment to achieve a reasonable distribution of observations. Ultrasonic wind speed and direction sensors that are commonly used in modern weather stations detect wind movement at any rate greater than virtually zero.
Data Availability: Measured solar radiation values support assessment and justification of the solar radiation values used for ampacity calculations. Typical weather parameter assumptions utilize a value approximating maximum solar radiation, which may not be coincident with temperatures and wind speeds used in the calculation of conductor ampacities.
Computational Capability
IEEE 738 provides a detailed method for computation of conductor ampacities for a given set of ambient weather conditions, maximum conductor temperature and physical conductor parameters (resistance, thermal emissivity, solar absorbtivity). The standard also provided a BASIC computer program as a convenience. Although convenient, the program handles only a single ampacity computation for a single conductor under a single set of ambient conditions. As such, the program is not suited for batch processing of the large datasets available today.
Computing ampacities for large numbers of weather observations provides insights into conductor ampacities that are not evident from individual weather parameters. The BASIC program provided with IEEE 738-1993 also utilizes the solar angle as determined by the latitude and azimuth of the sun to determine the incident solar radiation. While this approach is useful when a direct measurement of the incident solar energy is not available, it becomes unnecessary when direct measurements of incident solar radiation are available as is the case with many modern weather stations.
Batch processing large numbers of observations allows the annual characteristics of available conductor ampacities to be determined using a minimum number of assumed variables. The figure below shows a distribution of ampacities for 954 ACSR Cardinal conductor with absorbtivity and emissivity of 0.5 and a wind angle perpendicular to the conductor.
Batch processing a large number of observations allows identification of the risk associated with parameter selections. For example, the figure above shows a less than two-percent risk that the available ampacity of 954 ACSR Cardinal will be less than 1200A under the assumption that wind is perpendicular to the conductor.
Weather Parameter Correlation and Variability
Unfortunately, the various climatological measures are correlated, often in unexpected ways. These correlations can adversely impact the quality of the assumptions underlying static thermal ampacities. One of the more surprising observations is that ampacities computed for weather observations during darkness often result in lower values than those computed for daylight hour observations. This is a result of the correlation between solar heating and temperature during the day and wind speed. The result: winds are generally calmer in darkness which can counteract the benefits of these reduced solar heating and lower temperatures experienced at night.
The figure below shows the percentage of observations for which calm conditions (wind speeds less than 0.5 miles per hour) were recorded over a 10-year period in central Florida. The figure shows calm observation percentages for the entire year and for summer (June-August). The figure also shows the observation percentages from these periods with screening for daylight (solar radiation greater than 1 W/m2) observations only.
This particular data set indicates that calm conditions are much less prevalent during summer daylight than when considering the year as a whole. This has two conflicting impacts: First, that during summer daylight periods (presumably including the warmest temperatures) the potential for forced convective cooling is greatest, potentially offsetting the higher ambient temperatures; Second, and conversely, calm conditions are more likely to occur during non-daylight periods (when temperatures can generally be expected to be lower, even in summer).
Of course, simply looking at correlations between weather parameters does not generally provide sufficient insight into the non-linear heat transfer impacts of variations in weather conditions. Such insights will generally require ampacity calculations across a wide range of observed weather conditions.
Verification and Justification
Processing of large data sets allows the applicability of static ampacity parameter assumptions to be assessed versus the potential that the static ampacity will be unavailable. This approach allows assessment of seasonal and temporal variations that can provide justification for current static ampacity parameter sets or impetus for reassessing parameters if the risk is too great or excessively small. The figure below shows a graphical interpretation of the risk associated with a given ampacity when assessed versus ten years of weather data taken at 15-minute intervals.
Of course, the risk represented in the above figure should not be interpreted as an absolute risk as some parameters used may be inherently conservative or optimistic (e.g., wind angle, emissivity and absorbtivity). Additional risk mitigation is provided by the fact that conductor loading near the static ampacity rating would need to be coincident with several consecutive observations of adverse weather conditions to result in adverse conductor heating.
Rich weather datasets and computational capability that allow assessment of parameter correlation and provide a basis for verification of rating parameters will not end debate over appropriate parameters for use in determining static line ratings. However, these items can provide quantification of the risk associated with weather parameters to support intelligent system planning and operation.
