Performance Analysis of FFBP-LM-ANN Based Hourly GHI Prediction Using Environmental Variables: A Case Study in Chennai

Table 2

Performance comparison of study model with similar existing work.

References	Location	Variables	Network	Epochs	RMSE
[25]	Buraydash/Saudi Arabia	DNI, SZA, DHI, hours	4-5-1	-	30.23
[45]	India	Hourly climate data	13-20-1	36	7.591
[27]	Nigeria	Lat, long, day of the year, year	4-20-1	46	3.98
[21]	Mumbai/India	Lat, log and altitude, year and month, mean ambient temperature, mean station level pressure, mean wind speed, mean relative humidity.	6-24-1	1000	3.6461
[23]	Makurdi/Nigeria	Min. Temperature, max. Temperature, relative humidity, wind speed, sunshine hours	5-5-5-4-1	1000	0.98831
[17]	Utter Pradesh/India	Lat, long, altitude, max. Temperature, min. Temperature, humidity, wind speed, sunshine, solar radiation	9-10-1	33	0.1936
[26]	Adane/Turkey	Actual pressure, wind speed, wind direction, relative humidity, average temperature	5-10-30–1	124	0.1543
[24]	New Delhi/India	Lat, long, elevation, the month of the year, day of the month, temperature, atmospheric pressure, humidity, wind speed	9-10-1	1000	0.09441
[30]	Tehran	Temperature, relative humidity, wind speed, particulate matter	5-12-24–1	-	0.0773
Present study	Chennai/India	SZA, RH, T, CC, PW	5-20-1	15	0.07126