Juha Kestilä’s research while affiliated with University of Helsinki and other places

Centrifugal pumps are often used in operating conditions where they can be susceptible to premature failure. The cavitation phenomenon can be considered as a typical cause of pump failure as it erodes the internal parts of the pump and deteriorates the pump performance. The cavitation phenomenon can be detected by means of external measurements, for instance by monitoring the pump vibration. However, the use of additional sensors results in additional costs, which favours the use of sensorless methods in the field of condition monitoring. In this article, a novel method for detecting cavitation in a centrifugal pump with a frequency converter is proposed. The method is based on the frequency converter's estimates for the rotational speed and shaft torque of the pump. The method has been tested in the laboratory with two different pumping systems and the obtained results are presented. Test results show that high flow cavitation can be detected by the proposed method.

Pumps are widely used in industry, and they account for 20% of the industrial electricity consumption. Since the speed variation is often the most energy-efficient method to control the head and flow rate of a centrifugal pump, frequency converters are used with induction motor-driven pumps. Although a frequency converter can estimate the operational state of an induction motor without external measurements, the state of a centrifugal pump or other load machine is not typically considered. The pump is, however, usually controlled on the basis of the required flow rate or output pressure. As the pump operational state can be estimated with a general model having adjustable parameters, external flow rate or pressure measurements are not necessary to determine the pump flow rate or output pressure. Hence, external measurements could be replaced with an adjustable model for the pump that uses estimates of the motor operational state. Besides control purposes, modelling the pump operation can provide useful information for energy auditing and optimization purposes.In this paper, two model-based methods for pump operation estimation are presented. Factors affecting the accuracy of the estimation methods are analyzed. The applicability of the methods is verified by laboratory measurements and tests in two pilot installations. Test results indicate that the estimation methods can be applied to the analysis and control of pump operation. The accuracy of the methods is sufficient for auditing purposes, and the methods can inform the user if the pump is driven inefficiently.

Pumps are widely used in industry. Nowadays, a frequency converter can be used to control the operation of a pump drive. This improves the operational flexibility of the pump drive compared with a throttle control method. In addition, modern frequency converters provide information on the pump operation for monitoring purposes. However, in some cases also speed variation can be an inadvisable flow control method. To optimise the pump drive operation, there should be limits for the recommendable operating region of the pump drive. General guidelines for fixed-speed pumps have been given, but there is no mention about variable-speed-driven pumps. Since limit values are pump-specific and rotational speed of the pump affects them, recommendation limits for variable-speed-driven pumps are difficult to determine. In this article, factors affecting the recommendable operating region of variable-speeddriven centrifugal pump are discussed. Operating region recommendations are then given for a laboratory pump drive. The validity of the recommendations is analysed by test measurements.

Pump drives account for nearly a quarter of the electric energy consumption in the European industry (Almeida and Fonseca, 1997). Furthermore, their reliability has a strong impact on the production in paper industry. Hence, the optimization of energy consumption and maintenance can lead to substantial cost savings. By applying pump drive life cycle cost analysis (LCCA), the most critical factors affecting life cycle costs can be determined. Nowadays, an inverter can be used to monitor the power consumption and condition of a pump drive. With these abilities, the inverter provides a way to optimize pump drive life cycle costs. In this article, life cycle costs for three different pump drives are estimated. In addition, the increase in electric energy price in the future is taken into account in the LCCA. Based on the analysis, the most important operations that should be performed for life cycle cost optimization are determined.