Hi @Ajay
My guts tell me that in one of the three fuzzy inputs {Temperature, Humidity, and Wind Speed}, the designed MFs did not cover the entire the universe of discourse. For example, say 0°C – 40°C is LowTemp MF, and 60°C – 100°C is HighTemp MF, but no MF is assigned for the mid range 30°C – 70°C. If the measured temperature is around 50°C, then "no rules fired for Output 1".
Can you show or post an image from the ruleview(fis) command? It would help to find the root cause of your case.
Edit: By viewing the inference process for your fuzzy system, I guess I have found the root cause of the warning.
Fig. 1: The inference process for your fuzzy system.
See my example of the Mamdani FIS below. If the values from both input are in the range 
, then it guarantees at least 3 rules will be triggered. In Fig. 2, a maximum of 8 out of 9 rules are triggered and the output value is 0.5 (from the centroid defuzzification method).
, then it guarantees at least 3 rules will be triggered. In Fig. 2, a maximum of 8 out of 9 rules are triggered and the output value is 0.5 (from the centroid defuzzification method).
Fig. 2: 8 out of 9 rules are triggered and the output value is 0.5.
In Fig. 3, Input #1 is 0 and Input #2 is 1, but the output value is also 0.5. This happens because none of the rules is triggered, and the internal algorithm automatically set the defuzzified output value to its mean range value that is 0.5.
Fig. 3: No rules fired for output1 and the defuzzified output value set to its mean range value.
For the technical explanation, this happens because all designed MFs only cover the range , excluding 
and 
. Thus, since you used the OR operator in your rules, if any of the input value is either 0 or 1, then, no rules fired.
, excluding and . Thus, since you used the OR operator in your rules, if any of the input value is either 0 or 1, then, no rules fired. Suggested solution:
To prevent this, make sure the designed MFs cover the entire universe of discourse, as shown in Fig. 4.
Fig. 4: the designed MFs in the fuzzy inputs cover the entire universe of discourse, including the extremities {0} and {1}.
