Due to the increasing density of buildings in the city, the occlusion situation is more complex. Generally, there are several buildings sheltered by the sheltered buildings, so the time period for the buildings to get sunshine is often discontinuous, and in some cases, there will be a few minutes or even shorter time period.
I wish the “Ladybug_Sunlight Hours Analysis” could add a “Minimum Time Span filter” (MTS for short) allowing users to set a number in minute to control the minimum continuous sunshine time condition. While this function activated, for each test unit（point or mesh），only the continuous period not less than MTS shall pass the filter condition and be added to the sunlight time.
There are several reasons for not taking these short sunshine time periods into account: first, the sunshine quality of the short sunshine time period is poor; second, due to the limitation of data accuracy and software calculation accuracy, the errors that are easy to appear in the calculation are generally about a few minutes; in addition, in the actual observation, due to the influence of factors such as blurred shadow boundary and surrounding environment, etc It is difficult to judge and detect too short sunshine period. From the comparison of software calculation and actual observation, the deviation between the two is generally within 3.0min-5.0min, so the sunshine period less than 5.0min is uncertain and meaningless. This regulation is mainly from the point of view of convenient calculation. It is also reasonable for some cities to make local regulations on the minimum continuous sunshine time in order to ensure the sunshine quality Appreciate your works
Besides，the time interval of sampling in sunshine calculation is an important factor affecting the calculation error. In order to ensure the accuracy of the sunshine analysis results and control the error within a reasonable range, the shadow sampling time interval should not be greater than 1.0min for the calculation software using the shadow time-sharing algorithm.
For the current version，I guess the sunlightHoursResult value is something relating to the amount of sunlight vectors. So i have to set the timeStep to 60 to increase the sunlight vector amount conversing unit from hour to minitue virtually. After that i need do a division by 60 to the sunlightHoursResult , and make a customed legend par to set the unit back to hour， because the default unit of legendpar of the Ladybug_Sunlight Hours Analysis component is in hour while the value is in minute in fact (time values and unit dont match automatically). All the work is done manually. I wish you could optimize the componet about the way results calculated and the unit translation.
Can you be more specific about the limitation of data accuracy and calculation? What type of errors have you seen using Ladybug?
Do you have a specific case that you can show this error or is it a generic statement that things can be different between observation and calculation because the input models are different from the real buildings. There is nothing random or uncertain about sunlight hours calculation. We can determine the exact sun position at any point of time with a high level of accuracy.
You should be careful in making statements like this. Regulation can differ from country to country and even city or state in the same country.
As you have mentioned in your post you have the option to calculate sunlight hours for each minute but making it a rule as you are trying to do is not how Ladybug Tools is designed. That will make the tool inflexible and just useful for a very certain case.
I also disagree with the statement that interval should not be greater than 1 minutes. It totally depends on the use case of the study.
There is an input in sunlight hours component for timestep. You should connect the same timestep as the one for sunpath to this component. You don’t have to do the post-processing manually.
Dear @mostapha, on the last point, I also encountered the same problem.
If I create the sun vectors for every minute out from SunPath (TimeStep=60), into the sunlightHoursAnlaysis, if I want to keep the result in minutes, I have to set the time step to 1 (See Image 01), because if I use the same value (60 -60), the final calculation does it in hours (See Image 2).
dear @wanaboy, excellent consideration, but if it is not regulated by the norm of your country, you should not worry about it.
Follow the Norm, so no one can object the result of your analysis.
Greetings
Hi @LiamRuvio, I don’t see that as an issue. That’s your choice if you want to keep the results in minutes. You can use recolor mesh component to adjust the legend if that is the issue that you are referring to.
