The month of July was just crazy. There were three very different "voltage behaviours" during the month. First, and until 17 July, the voltage was well-behaved with an average very near the nominal voltage of 120 VAC. This was good. Then, around 3:30 p.m. on 17 July the voltage was suddenly increased by about 5 VAC, from 117 to 122 VAC, and with the voltage never going below 120 VAC. Then, superimposed on this new, higher voltage, there were six days when the voltage appeared to be out of control, rising at times above 130 VAC and as high as 132.6 VAC.
Above is a graph of the voltages on the six days with crazy voltage behaviour. The only pattern that I see here is that the voltage was highest during the early morning hours, similar to most days, except that these voltages were especially, even dangerously, high. One wonders what is going on at Hydro One and their generating stations. Can someone explain, please??
So, are the generally high temperatures causing these wild voltage variations? It seems possible. Although the data sample is small (only 14 days), the chart above does show that when the daily maximum temperature is 27° C or less, the voltage is well-behaved. When the maximum temperature is above 27° C, there is a good chance (6 times out of 11) that the voltage will misbehave and go to 130 VAC or above. My guess is that there is some errant computer algorithm that makes inconsistently bad decisions when the outside temperature is 27° C or above. Or, maybe it is just some nervous supervisor at Hydro One? It is curious that the "bad" days seem to have come in pairs.
The histogram above shows the statistical distribution of voltage measurements and the predominance of voltages above the supposed nominal value of 120 VAC (too much voltage). The voltage should be above and below the nominal voltage 50% of the time each. Clearly it is not.
Looking at the Average Voltage at Time of Day diagram above shows the general trend of voltage being higher at night and lower during the daylight hours of each day. But this chart is almost meaningless because it is comprised of data from before 17 July when the voltage was well-behaved, data afterwards when the voltage was generally higher by 5 volts, and data from the six days when the voltage was out of control.
The diagram above shows all of the voltage measurements according to the time of day when they were taken during July. You can easily see how confused the data is, as described above.
See voltage measurements for other months during 2019.