[Show abstract][Hide abstract] ABSTRACT: The best test of quality of an estimated model is its implementation in real application. However, the use of a bad model is typically too costly. Therefore, model validation is considered as an obligatory step in model learning, and extensive theory has been developed within statistical community. However, the available rules deal almost exclusively with independent data samples. Consequently, they are substantially disqualified for validation of dynamic models. This paper approaches the problem using Bayesian formulation and solution. An algorithm for validation of models estimated within practically important exponential family is presented. Performance of the algorithms is illustrated on simulated example.
[Show abstract][Hide abstract] ABSTRACT: Thermoelectric module is a heat pump device. When DC voltage is applied to the device, the substrate surface becomes cold for absorbing heat energy and another substrate surface becomes hot for releasing the heat energy. The heat sink is attached to the hot side, and the accumulated heat in the heat sink is removed by an AC fan. The main disturbance source of cold plate surface temperature is this accumulated heat; therefore, reduction of the accumulated heat from the heat sink can improve the tracking performance on cold side temperature of thermoelectric plate. The thermoelectric system has two inputs and two outputs. The first input is a voltage applied to power drive to operate thermoelectric and the second input is a PWM signal applied to solid state relay for adjusting the speed of AC fan. This paper proposes model predictive control for controlling cold plate temperature and setting a constraint for bounded estimated heat sink temperature. The heat sink temperature is estimated by using Kalman filter. Simulation and experimental results show that the proposed method handles the accumulated heat in the heat sink better than a model predictive controller without constraint, resulting in significantly more accurate tracking of the cold plate temperature.
The 5th TSME International Conference on Mechanical Engineering, 17-19th December 2014, The Empress, Chiang Mai; 12/2014
[Show abstract][Hide abstract] ABSTRACT: In this paper, we consider a number of technical problems associated with identifica-tion of linear systems using quantized and intermittent information. More specifically, we study asymptotic properties for identification of such systems, including strong consistency, asymptotic normality and asymptotic efficiency, and determine their relationship with quantization errors and the packet losses. Furthermore, we discuss how to design quantizers to improve these asymptotic properties. Some open problems will also be proposed.
Data provided are for informational purposes only. Although carefully collected, accuracy cannot be guaranteed. The impact factor represents a rough estimation of the journal's impact factor and does not reflect the actual current impact factor. Publisher conditions are provided by RoMEO. Differing provisions from the publisher's actual policy or licence agreement may be applicable.