A practical NMPC with robustness of stability applied to distributed solar power plants

Optimal operation of a Solar Membrane Distillation pilot plant via Nonlinear Model Predictive Control

“…For this purpose, a PNMPC strategy has been adopted (Plucenio et al, 2007), which has already been successfully tested in other nonlinear systems (Castilla et al, 2014;Álvarez et al, 2013), being also possible to 290 increase robustness of the approach by filters shown in Andrade et al (2013). PNMPC algorithms are identified by the use of a vectorŶ that includes predictions of the future system outputs in a determined prediction horizon N , as a function of the future movements of the control signal ∆U:…”

“…Thus, to calculate F and G PNMPC for each control mode at each sample time, the algorithm presented in Algorithm 1 and described in Plucenio et al (2007) should be used. Ad-320 ditionally, as pointed out in Plucenio et al (2007); Andrade et al (2013), for the treatment of disturbances, prediction errors and noises, the implementation of the PNMPC algorithm in this work uses an explicit form of generalized predictive control where the integral of the filtered prediction error is added to the predicted system output to correct open loop predictions (not explicitly a The algorithm has been described for the tank fast heating mode case, in which only one input is used. When two inputs are used, as in the normal operating mode case, this procedure must be repeated for each input, applying small increments at one of the input while the other one is kept constant, thus obtaining the formulation presented in Eq.…”

Distributed Collector System: Modelling, Control and Optimal Performance

“…The resulting controllers are believed to present a high degree of robustness when they are appropriately tuned. Lately, robustness of stability against parameters uncertainty and measurement errors in a nonlinear MPC has been taken care of by simply including a candidate Lyapunov function in the objective function and the constraints of the controller [43].…”

“…A linear model is used for the forced response to obtain a set of control actions, while a nonlinear model of the free response is used to handle the effect of the disturbances. A recent application of a nonlinear MPC is presented in [43]. A distributed parameter model is used for the simulated process and a lumped parameter model with time delay is used for prediction.…”

Particle Swarm Optimization Applied to Generalized Predictive Control of a Solar Power Plant

“…The plant dynamics also can be approximated by an affine state-space neural network, of which the complexity is depend on the cardinality of dominant singular values associated with a subspace oblique projection of data-driven Hankel matrix [9]. The use of Nonlinear Model Predictive Controller (NMPC) can manipulate the oil flow rate to maintain the field outlet temperature in the desired reference value and attenuate the disturbances effects [10]. The application of a Nonlinear Model Predictive Controller (NMPC) extended with a dead-time compensator (DTC) of a distributed solar collector field also obtained good performance [11].…”