Ing. Ondřej Zoubek, Ph.D.

This paper discusses the methods of optimal IIR filter FPGA implementation. The methods are focused on the reduction of occupied resources and increasing data throughput. Higher demands on an internal controller complexity are successfully solved by utilizing programmable microcode controller. The novelty of SOS core and its capabilities are presented and different variants of SOS core are assessed. The workflow of IIR filter design using MATLAB considering rounded coefficient method is demonstrated.

Wheel speed of heavy freight train locomotive under non-ideal traction conditions was measured with very high resolution and fine precision. Many authors published papers in this area over past century, but only recent advance in the information technology enabled measurement in such extent and resolution at the same time. Most modern locomotives have the speed sensor coupled with the traction motor. Unlike many other authors doing research in the area of wheel-slip, in the presented case the wheel speed was measured directly on the wheel, not indirectly on the traction motor. This approach made it possible to observe the wheel speed directly, without the requirement to implement motor – wheel model, which can introduce inaccuracies. Thanks to an independent method of train speed determination using a sensitive GPS receiver, wheel slip speeds were identified accurately regardless of synchronous slip phenomenon. Despite the adhesion theory, three types of operation mode were identified: the stable traction with negligible slip without oscillations, the semi-stable traction with evident slip and oscillations, and the prominent wheel slip. As the wheel speed was measured directly on the wheel, much larger oscillations than ever published were revealed. The source of the oscillations is discussed in the paper. A simple MatLab model of rotational masses was constructed and tuned to match the measurements. Eigen values and eigenvectors of the model are discussed, too. The model of rotational masses is completed with adhesion model to perform simulation. The simulation results are presented and compared with the results of measurement.

Important task of railway traction vehicles is to achieve maximal traction effort at any conditions if required. For this purpose the vehicles have slip controllers. Some types of this controllers need the vehicle longitudinal velocity value as an input. The vehicle longitudinal velocity is difficult to measure directly because the vehicle longitudinal velocity could not be directly determined from wheel velocity. When the wheel is driven its velocity is higher than the vehicle longitudinal velocity because of the slip phenomenon. The paper describes method that provides possibility to estimate the vehicle velocity on base of measured wheel velocity. Estimation of the vehicle longitudinal velocity is enabled by Kalman filtration. The method is verified by MatLab simulation using real measured data.

For a locomotive wheel slip velocity detection and slip control are typically used measurement sensors that are mounted on the locomotive. The sensors are typically intended for another purpose e.g. wheel speed measurement or tractive effort determination. When the slip control is based on indirect methods that needed to measure some specific locomotive behavior that occurs during high slip velocity it is difficult to use the locomotive measurement sensors. The paper presents an additional experimental measurement device that is mounted on locomotive wheelset to measure the wheelset dynamic motions that typically occur when the wheel has large value of slip velocity. The measurement device is equipped with accelerometers. Measured data are filtered to get signal that can be used for locomotive wheel slip velocity detection and control.

Instantaneous speed measuring algorithm for incremental rotary encoder (IRC) is presented in this paper. Overview of implementation of the algorithm in a Field Programmable Gate Array (FPGA) is presented. The proposed method was used to achieve advanced motor control using single FPGA without standalone Digital Signal Processor (DSP) using an internal soft processor. Using soft processors inside FPGA is a highly progressive method which helps to increase system reliability and decrease the system cost. The results from experimental implementation are presented.

System overview and design principles of power electronics design with soft processor, implemented in a Field Programmable Gate Array (FPGA), are presented in this paper. Typical power electronics system consists of a Digital Signal Processor (DSP) and a FPGA attached to this processor. In the presented system the entire control is achieved using only one complex digital part. The entire system is working on a single chip, eliminating separate standalone DSP (System on Chip or SoC). The FPGA contains all necessary digital circuits including dual processor core implemented inside FPGA structure. Using soft processors inside FPGA is a highly progressive method which helps to increase system reliability and decrease the system cost. The results from experimental implementation are presented

Instantaneous speed measuring algorithm for incremental rotary encoder (IRC) is presented in this paper. Overview of implementation of the algorithm in a Field Programmable Gate Array (FPGA) is presented. The proposed method was used to achieve advanced motor control using single FPGA without standalone Digital Signal Processor (DSP) using an internal soft processor. Using soft processors inside FPGA is a highly progressive method which helps to increase system reliability and decrease the system cost. The results from experimental implementation are presented.