Design of stochastic PV water pumping systems.

摘要

new method for systematic optimization of photovoltaic water pumping systems has been developed by posing and solving an optimization problem which includes a nonlinear objective function subject to a set of six to eight nonlinear inequality and equality constraints with eight decision variables.;The objective is to minimize both the amount of water in excess of the demand and the net deficit should the demand be greater than the supply of water. The objective function reflects the overall system performance measure and is a nonlinear function of the decision variables: S, the number of photovoltaic modules in series of P strings in parallel, Ke the motor voltage constant, KT the torque constant, D the pump load constant, d, the diameter and H, the height of the tank, and Hg its elevation above the ground.;The constraints include one based on a cost limitation and the others based on the physics of the system. The solution of the program so defined is obtained with a line search algorithm and takes into account the load variability and random autocorrelated irradiation as well as sequential drawdown of the water table.;Subsequently, a computer code that processes irradiation, environmental and economical data, was developed in Matlab. The code evaluates system long term (10 years) as well as short-term performances, in time steps of roughly two minutes. Results are compared with available data from an existing PV pump system at Camp Modibo, Mali.;The use of the optimization algorithm results in an optimized PV motor-pump system capable of delivering quantity of water satisfying the demand there at the cost of