No abstract

No abstract

Objective: Study of the possibility of carrying out predictive analysis of the technical condition of locomotive equipment using neural network predictive models enabling to plan the scope of equipment maintenance for routine types of maintenance and repair. Methods: A comparative assessment of the accuracy of forecasts made using a feedforward neural network and a recurrent network with an LSTM layer (Long Short-Term Memory) has been carried out. For training and test-ing of predictive models, we used the results of monitoring the parameters of the lubrication sys-tem of the 2TE116 (2ТЭ116) diesel locomotive by means of on-board diagnostics. Results: The aver-age interval for preventive inspections (TO-3) of locomotives in the existing locomotive mainte-nance system is 25–30 days, and therefore it is this interval that determines the minimum duration of the lead-in period, which the predictive model should provide. We have established that a mod-el based on a feedforward neural network provides sufficient accuracy only for short-term fore-casts with a lead period of no more than 1–3 days. With a further increase in the lead-in period, the error of the model res¬ponse increases to 10–15 %, which prevents it from being effectively used for solving practical problems associated with planning the operation of service locomotive depots. At the same time, the ave¬rage response error of the predictive model based on a recurrent net-work with an LSTM layer does not exceed 3,5–5 % over a 30-day lead-in period, so it can be used to plan the scope and timing of locomotive maintenance procedures. Practical importance: The possi-bility of using time-series analysis methods for predictive analytics of the technical condition of units and systems of a locomotive is shown. Predictive models based on recurrent neural networks with LSTM layers provide prediction accuracy and lead-in period sufficient for solving practical prob-lems that are associated with planning the scope and timing of locomotive maintenance.

Purpose: The purpose of the present work is to test experimentally equalization method for the outputs of multi-cylinder diesel engine cylinders basing on energy minimization of amplitude spectrum low-frequency band for diesel crankshaft instantaneous angular speed signal. Methods: The work uses the methods of digital signal processing, spectral analysis, differential calculus, multidimensional undifferentiated function optimization, experimental results processing. Results: The fulfilled experimental studies at bench diesel generator device as a part of locomotive diesel D50 (6CHN31.8/33) and traction generator MPT 84/39 ensure the possibility to use proposed earlier method for power equalization for the cylinders of multi-cylinder diesel engine basing on energy control of amplitude spectrum low-frequency band of crankshaft instantaneous angular speed signal. Unrvenness index for cylinder power is proposed, its sensitivity to cyclic supply change and the advance angle of fuel supply to diesel engine individual cylinders is estimated. Practical significance: The possibility to apply being proposed equalization method for the powers of multi-cylinder diesel engine cylinders for to adjust fuel equipment after diesel engine repair as with the systems of electronic fuel supply control as well as with hydromechanical fuel equipment has been confirmed experimentally. The use of given method eliminates the need to measure the temperature of exhaust gases and flash pressure in cylinders for fuel equipment adjustment that’s especially important for high-speed diesel engines which cylinder cover construction doesn’t imply indicated valves.

Purpose: Labor intensity reduction and equalization accuracy improvement of cylinder powers of a multi-cylinder diesel locomotive on account of application as a diagnostic parameter, characterizing cylinder power unevenness, — the energy of low-frequency band of spectrum for crankshaft instantaneous angular velocity (CIAV) signal. Methods: Methods of mathematical modeling of diesel crankshaft instantaneous angular velocity, mathematical modeling of diesel engine working process, digital signal processing, spectral analysis, differential calculus and multidimensional non-differentiable function optimization are used in the work. Results: Method for equalization of cylinder powers of a multi-cylinder diesel locomotive that’s based on a continuous monitoring of the spectrum of crankshaft instantaneous angular velocity signal is proposed and justified. Energy of low-frequency band of the signal spectrum of CIAV can be used as a diagnostic parameter, characterizing cylinder load unevenness. The parameter value minimization by means of sequential correction of cyclic fuel supply and fuel advance angle ensures the decrease in the pointed unevenness. Practical importance: The proposed method can be used when adjusting fuel equipment after the repair of a diesel engine as with electronic fuel supply control systems as well as with hydromechanical fuel equipment. The use of the given method excludes the necessity of temperature measurement for exhaust gases and flash pressure measurement in cylinders for fuel equipment adjustment that’s especially important for high-speed diesel engines which cylinder heads’ construction doesn’t provide for indicator cocks. In the engines equipped with electronic fuel control systems, the use of IAV signal spectrum low-frequency band energy as a diagnostic parameter makes it possible to automate fully the equalization process for cylinder load. The result of the proposed method application will be the reduction in the duration and labor intensity of diesel locomotives rheostat tests as well as the rise in diesel locomotives reliability and efficiency at the expense of uniform loading of diesel cylinders.