Control scheme formulation for a parabolic trough collector using inverse artificial neural networks and particle swarm optimization

This work shows the results of a new nonlinear control approach for the water outlet temperature control in a parabolic trough collector (PTC). The controller is based on an inverse artificial neural network (ANNi) and the particle swarm optimization (PSO) algorithm. This proposed method is capable of operating at different system reference points due to the ANNi-PSO combination. The ANNi was built from a feedforward ANN with six inputs (rim angle, input temperature, ambient temperature, wind velocity, solar radiation, and input water flow) and one output (outlet water temperature). The ANNi purpose is to obtain some of the ANN input variables considering the desired outlet temperature. In this specific case, the interest variable for the ANNi is the PTC input water flow. To guarantee the optimal water flow supplied to the PTC, the ANNi is solved by the PSO. These control systems are complicated because the environmental conditions are not manipulable (wind speed, solar radiation, ambient temperature) and disturbances. Simulations were carried out considering cloudy days, intense wind speeds, and very low solar radiation to verify the proposed control performance. The control strategy results were satisfactory. The tests were performed for abrupt and smooth reference changes. The results showed that, in some cases, the control reaches saturation due to the climatic conditions to which the PTC is exposed.

Automated summary

This paper presents a new nonlinear control approach for water outlet temperature control in a parabolic trough collector using an inverse artificial neural network and particle swarm optimization algorithm. The method is capable of operating under different system reference points and has been validated through simulations under various environmental conditions.