Assessment of Noise Signature for a Cavitating Centrifugal Pump

In recent days, sophisticated instruments have emerged to obtain an online measurement of performance parameters from centrifugal pump of different kinds and the signals can be directed to the hands of pump user through mobile application. With this in mind, a centrifugal pump of low specific speed was chosen for cavitation studies from 80% to 120% of nominal flow rate and for three different speeds. An assessment was carried out for cavitation noise signature from those operating condition of that pump. The result of cavitation noise based on peak magnitude as well as average revealed a nature in relation to cavitation coefficient and it greatly depends on the flow rate with respect to nominal flow rate. The noise envelope for the flow rate at best efficiency and above was having similar trend whereas at flows less than the nominal, it was totally different. So the criteria for finding the deviation in noise cannot be uniform for all flow rates. In this paper, the method adapted was to impose a trend line to the measured cavitation noise information and to find out the deviation with respect to normal operating condition. It was concluded that detection of abnormality in pumps due to cavitation effects requires the current operating condition to be diagnosed first and then proper criteria for deviation in noise has to be imposed.

Reducing Uncertainty in the Onset of Combustion Instabilities Using Dynamic Pressure Information and Bayesian Neural Networks

Modern, low emission combustion systems with improved fuel-air mixing are more prone to combustion instabilities and therefore use advanced control methods to balance minimum NOx emissions and and the presence of thermoacoustic combustion instabilities. The exact operating conditions at which the system encounters an instability is uncertain because of sources of stochasticity, such as turbulent combustion, and the influence of hidden variables, such as un-measured wall temperatures or differences in machine geometry within manufacturing tolerances. Practical systems tend to be more elaborate than laboratory systems and tend to have less instrumentation, meaning that they suffer more from uncertainty induced by hidden variables. In many commercial systems, the only direct measurement of the combustor comes from a dynamic pressure sensor. In this study we train a Bayesain Neural Network (BNN) to predict the probability of onset of thermoacoustic instability at various times in the future, using only dynamic pressure measurements and the current operating condition. We show that, on a practical system, the error in the onset time predicted by the BNNs is 45% lower than the error when using the operating condition alone and more informative than the warning provided by commonly used precursor detection methods. This is demonstrated on two systems: (i) a premixed hydrogen/methane annular combustor, where the hidden variables are wall temperatures that depend on the rate of change of operating condition, and (ii) full scale prototype combustion system, where the hidden variables arise from differences between the systems.

Reducing Uncertainty in the Onset of Combustion Instabilities using Dynamic Pressure Information and Bayesian Neural Networks

Modern, low emission combustion systems with improved fuel-air mixing are more prone to combustion instabilities and therefore use advanced control methods to balance minimum NOx emissions and and the presence of thermoacoustic combustion instabilities. The exact operating conditions at which the system becomes encounters an instability is uncertain because of sources of stochasticity, such as turbulent combustion, and the influence of hidden variables, such as un-measured wall temperatures or differences in machine geometry within manufacturing tolerances. Practical systems tend to be more elaborate than laboratory systems and tend to have less instrumentation, meaning that they suffer more from uncertainty induced by hidden variables. In many commercial systems, the only direct measurement of the combustor comes from a dynamic pressure sensor. In this study we train a Bayesain Neural Network (BNN) to predict the probability of onset of thermoacoustic instability at various times in the future, using only dynamic pressure measurements and the current operating condition. We show that, on a practical system, the error in the onset time predicted by the BNNs is 45% lower than the error when using the operating condition alone and more informative than the warning provided by commonly used precursor detection methods. This is demonstrated on two systems: (i) a premixed hydrogen/methane annular combustor, where the hidden variables are wall temperatures that depend on the rate of change of operating condition, and (ii) full scale prototype combustion system, where the hidden variables arise from differences between the systems.

Development of Track Support Stiffness Measurement and Evaluation System for Slab Tracks

The track support stiffness measurement and evaluation system for slab tracks proposed in this study enables the calculation of the load–displacement diagram at the various measurement positions. Therefore, it is possible to evaluate the track support stiffness directly on-site without evaluating the spring stiffness of the elastic material through sampling or in situ testing, also enabling the evaluation of the deterioration of the elastic material. In addition, the performance evaluation data for elastic materials obtained through field tests using measurement equipment and software to track support stiffness are integrated and managed on the administrator′s computer. Therefore, the replacement plan is established, and the maintenance history is managed by identifying the replacement time and location of elastic materials. It is possible to evaluate the performance and condition of the elastic material at various points during track inspection and the track support stiffness and durability of the elastic material (spring stiffness variation rate, replacement periods, among others) at the current operating condition.

Artificial Intelligence Mechanisms on Interactive Modified Simplex Method with Desirability Function for Optimising Surface Lapping Process

A study has been made to optimise the influential parameters of surface lapping process. Lapping time, lapping speed, downward pressure, and charging pressure were chosen from the preliminary studies as parameters to determine process performances in terms of material removal, lap width, and clamp force. The desirability functions of the-nominal-the-best were used to compromise multiple responses into the overall desirability function level orDresponse. The conventional modified simplex or Nelder-Mead simplex method and the interactive desirability function are performed to optimise online the parameter levels in order to maximise theDresponse. In order to determine the lapping process parameters effectively, this research then applies two powerful artificial intelligence optimisation mechanisms from harmony search and firefly algorithms. The recommended condition of (lapping time, lapping speed, downward pressure, and charging pressure) at (33, 35, 6.0, and 5.0) has been verified by performing confirmation experiments. It showed that theDresponse level increased to 0.96. When compared with the current operating condition, there is a decrease of the material removal and lap width with the improved process performance indices of 2.01 and 1.14, respectively. Similarly, there is an increase of the clamp force with the improved process performance index of 1.58.

A Thermoeconomic Based Tool for the Real-Time Management of a Cogenerative Combined Cycle

In this paper a tool for the management of large cogeneration plants based on thermoeconomic analysis is presented and applied to a combined cycle. The tool is constituted of two thermodynamic models of the plant and a thermoeconomic model. The first model is a plant simulator whose characteristic parameters are obtained from measured data or information available in literature. Inputs are the electric and the thermal loads and the ambient conditions. This model determines the corresponding fuel consumption, mass and energy fluxes exchanged between the components and the thermodynamic state of all the mass flows. It can be used to determine how variations in the loads, in the ambient conditions, in the setpoints affect the plant operating condition as well as to generate a proper reference condition for the analysis. The second model performs data reconciliation. This is directly based on measured data corresponding with the current operating condition. The unavailable quantities are obtained by applying the balance equations together with some proper hypotheses. The value assumed by the characteristic performance parameters in the current condition constitutes an output of this model. The thermoeconomic model provides the costs associated with the plant products (heat supplied to the district heating network and electricity) as well as some evaluation parameters for diagnosis purposes. Future operating conditions can also be forecasted. In this case, the information provided by diagnosing the status of the components is introduced as an additional input in the simulator. The results can be used for planning the plant production, in particular for the day after, and for taking decisions about maintenance.