Journal: Polymers

Article Title: Short-Beam Shear Strength of New-Generation Glass Fiber-Reinforced Polymer Bars Under Harsh Environment: Experimental Study and Artificial Neural Network Prediction Model.

doi: 10.3390/polym16233358

Figure Lengend Snippet: Figure 21. Structure of ANN used in MATLAB (w = weight; b = bias).

Article Snippet: In the final testing phase, a separate test subset is used to objecti l of t e fully trained model. For this study, the neural network toolbox in MATLAB (“nftool”) was employ d t develop a feed-forward back-propagation model using the Bay sian regularization algorithm.

Techniques:

Journal: Materials

Article Title: Achieving Lower District Heating Network Temperatures Using Feed-Forward MPC

doi: 10.3390/ma12152465

Figure Lengend Snippet: Overview of FFMPC implementation and feed-forward signal: Scenario 1—historical load and Scenario 2—prediction model.

Article Snippet: This first order model was initially utilized as the MPC’s internal prediction model. A feed-forward model predictive controller (FFMPC) was then designed and developed in Matlab-Simulink by implementing a fixed six hour input-delay on the end-user heat demand where the physical model was connected to the MPC by using the Dymola-Simulink interface.

Techniques:

Journal: Materials

Article Title: Achieving Lower District Heating Network Temperatures Using Feed-Forward MPC

doi: 10.3390/ma12152465

Figure Lengend Snippet: Implementation of FFMPC results and comparison between actual and controlled signals, Scenario 1. ( a ) supply temperature from CHP, ( b ) supply temperature to the end-user, ( c ) return temperature to CHP from the end-user, and ( d ) mass flow rate.

Article Snippet: This first order model was initially utilized as the MPC’s internal prediction model. A feed-forward model predictive controller (FFMPC) was then designed and developed in Matlab-Simulink by implementing a fixed six hour input-delay on the end-user heat demand where the physical model was connected to the MPC by using the Dymola-Simulink interface.

Techniques: Comparison