Search

Neural network models become increasingly popular as dynamic modeling tools in the control community. They have many appealing features including nonlinear structures, being able to approximate any functions. While most researchers hold optimistic attitudes towards such models, this paper questions the capability of (deep) neural networks for the modeling of dynamic systems using input-output data. For the identification of linear time-invariant (LTI) dynamic systems, two representative neural network models, Long Short-Term Memory (LSTM) and Cascade Foward Neural Network (CFNN) are compared to the standard Prediction Error Method (PEM) of system identification. In the comparison, four essential aspects of system identification are considered, then several possible defects and neglected issues of neural network based modeling are pointed out. Detailed simulation studies are performed to verify these defects: for the LTI system, both LSTM and CFNN fail to deliver consistent models even in noise-free cases; and they give worse results than PEM in noisy cases., Comment: The failure of NARX model modeling a noiseless LTI system is mainly due to some initilization issues with the current Matlab SYSID Toolbox. If this procedure is done purely in the Neural Network Toolbox, the situation can be improved for great extent

Kalman filter is widely used for residual generation in fault detection. It leads to optimality in fault detection using some performance indices and also leads to statistically sound residual evaluation and threshold setting. This paper shows that these nice features do not necessarily imply an optimal fault detection performance. Based on a performance index related to fault detection rate and false alarm rate, several occasions where Kalman filter should not be used are pointed out; further the residual evaluation and threshold setting are discussed, in which it is pointed out that in stochastic setting an optimal statistical test of Kamlan filter is not related to optimality of commonly used detection performance indicators. The theoretical analysis is verified through Monte Carlo simulations and Tennessee Eastman process (TEP) dataset.

Multi-energy systems have been leaping forward for its various benefits, e.g., energy conservation and emission reduction. Coupling components are capable of transmitting energy from one time scale system to another time scale system, so the multi-energy system exhibits multi-time scale characteristic and broad bandwidth, thereby causing difficulties in dynamic modeling. In this work, two-time scale system identification is studied. A method is developed to solve the problem, which is uses signal pre-filtering and subtraction. The high and low frequency parts of the two-time scale system are identified separately and then combined to form the incorporated in parallel structure. The consistency of the method is proved and case studies are used to verify the effectiveness of the method.