3. Consider the 1 MW wind turbine whose power-wind speed characteristic Pu (vw) (in MW) is found in (1), and an historic probability distribution of the hub-height wind speeds observed in the vicinity of that wind turbine in (2). P. (0) 0, if v. 2 identical wind turbines as in (1) are forming a wind farm. Each turbine has a wind speed probability distribution indepen- dent, identically distributed (i.i.d.) as in (2). Demonstrate the positive effect of "wind farming with more than one turbine in terms of annual energy yields and its relative level of certainty. Comment on how, in reality, the independence assumption is not adequate and what it could affect your conclusions. Hint: You can estimate the energy yield probability distributions by drawing a suffi- ciently large sample of wind speeds and passing them through the speed-power char- acteristic of the wind turbine (i.e., perform this calculation by means of a "Monte Carlo Simulation"). Use Matlab, R or Python to generate your random samples and obtain the necessary statistics. 3. Consider the 1 MW wind turbine whose power-wind speed characteristic Pu (vw) (in MW) is found in (1), and an historic probability distribution of the hub-height wind speeds observed in the vicinity of that wind turbine in (2). P. (0) 0, if v. 2 identical wind turbines as in (1) are forming a wind farm. Each turbine has a wind speed probability distribution indepen- dent, identically distributed (i.i.d.) as in (2). Demonstrate the positive effect of "wind farming with more than one turbine in terms of annual energy yields and its relative level of certainty. Comment on how, in reality, the independence assumption is not adequate and what it could affect your conclusions. Hint: You can estimate the energy yield probability distributions by drawing a suffi- ciently large sample of wind speeds and passing them through the speed-power char- acteristic of the wind turbine (i.e., perform this calculation by means of a "Monte Carlo Simulation"). Use Matlab, R or Python to generate your random samples and obtain the necessary statistics