This Is AuburnElectronic Theses and Dissertations

A Method to Discriminate Between Flicker Sources

Date

2015-07-29

Author

Geiger, Daniel, II

Type of Degree

Dissertation

Department

Electrical Engineering

Abstract

This work is a discussion on discriminating between multiple flicker producing sources that are connected electrically close together in the power system. IEC 61000-4-15 provides a thorough description on how to build a device that measures voltage fluctuations and outputs a number that corresponds to customer irritation to the amount of voltage fluctuation present in the power system. IEEE has adopted IEC 61000-4-15 in IEEE 1453. The input to the flicker measuring device (flickermeter) is time sampled voltage data. It will be shown that it is acceptable to use full cycle RMS values as the input to the flickermeter when the voltage fluctuations are of a relatively low frequency. The RMS values will be held constant on the input of the meter for a full cycle. If the sampling frequency of the voltage is 64 samples per cycle, then the full cycle RMS value would be held constant to the flickermeter input for 64 samples. This allows the flickermeter to operate such that it does not matter to the meter whether the input values are time sampled or RMS values; the flickermeter treats the data as time sampled data. Flicker sources that are connected electrically close together could significantly impact the voltage at the terminals of each other. The flickermeter can determine the amount of flicker that is present in the power system via voltage measurements, but it is not able to determine the amount of flicker that a particular flicker producing source is generating. It will be shown that it is possible to calculate the equivalent voltage at the load of interest which corresponds to the voltage that the load would have if the other flicker producing sources were not connected to the power system. This will be accomplished by measuring the load’s voltage and current to determine the complex power that is being absorbed by the load. A power flow is performed using the load’s complex power to calculate the equivalent load voltage. The calculated voltage is an RMS value, valid for a full cycle, and is used as the input to the flickermeter. By using this equivalent voltage with the flickermeter, it is possible to determine the amount of flicker sensation that is due to a particular flicker producing source while ignoring flicker sensation that is produced by other flicker producing sources that are electrically close. The power flow that is performed to determine the equivalent voltage at the load is a two bus power flow where the source is treated as the slack bus. It is necessary to know what the Thévenin impedance is at the load to be able to calculate a valid voltage for the load using the power flow. If the impedance that is utilized is wrong, then the resulting voltage that is calculated will be incorrect. There is an approximate impedance that is known by the utility company, but this impedance will fluctuate based on the present power system operation. A method for determining the Thévenin impedance will be discussed such that it will be possible to perform the two bus power flow and calculate an accurate voltage. Demonstration of the concepts required at each step of the design process will be provided that validate the approach described in building a device that can determine the amount of voltage flicker being produced by a load when other flicker producing loads are connected electrically close together. Measurement results will also be provided from a device that has been implemented in hardware that utilizes the discussed concepts to measure the amount of flicker sensation that is due to a particular load of interest when there are multiple flicker producing loads connected. The measurement results consist of months of data that was collected at multiple sites that will substantiate the claims associated with the steps involved in building the flicker discriminating device. In summary, a methodology that describes how to determine the amount of flicker sensation that is being created by a flicker producing source that is in parallel with other flicker producing sources will be presented. This will be accomplished by utilizing the flickermeter defined in IEEE 1453 with either a half cycle or full cycle RMS input signal. The RMS values will be calculated from a two bus power flow problem that uses an estimated Thévenin impedance between the source and load. A justification for the acceptability of RMS inputs to the flickermeter in IEEE 1453 is presented. The output of the flickermeter will be the amount of flicker sensation that is being produced by the load of interest, even though there are other flicker producing loads connected electrically close together. A device that implements these design concepts has been built. There will be months of flicker data provided that the device collected at multiple sites. This provided data will validate the concepts that are provided in each stage of the device design.