Diagnostic method and device for evaluating and forecasting the technical condition of farm machinery in operation

The paper discusses methods and ways to diagnose the technical condition of agricultural machines and harvesters, existing practices, and approaches to get reliable data on the current health of the machinery used. The device for assessing and predicting machines’ technical condition includes software and technical means developed with virtual technologies to measure diagnostic parameters of the machinery. The main device elements are digital sensors with physical modifiers (pressure, temperature, medium composition and motion sensors, a-d converters with signal amplifiers), software to configure data gathering, and output to conduct analyses and produce recommendations. The core of the present approach is the technology of virtual prediction of breakdowns by changes in the technical condition parameters. It is based on modular devices, software with an interface that collects and processes data and provides a complete set of failure diagnostics and forecasting. The given method based on a device operating in the information and communication network increases farm machinery’s performance. Furthermore, it reduces operating costs due to the prevention of expensive breakdowns, individual forecasting, and scheduled maintenance of machines in operation. The approach under consideration was applied in the laboratory of digital engineering technologies of the Bashkir State Agrarian University Republic of Bashkortostan of the Russian Federation. The given work is aimed to boost the efficiency of the farm machinery diagnostics and maintenance system by applying a virtual breakdown prediction technology to conduct an automated evaluation, registration, and analysis of a machine’s condition. It can be achieved by developing software and technical means to register data and their structure systematization.

Vibration Signal Parameter Estimation in Variable Speed: Algorithms and performance bounds

Vibration signal parameter estimation for rotating machinery diagnostics operating under variable speedconditions is considered. At first, we provide a brief survey of existing methods for Quadratic Phase Signal (QPS)parameter estimation. Then, we introduce improved solutions for the general QPS case and the Order QPS (O-QPS)case, respectively. For all considered cases (namely the QPS, O-QPS with tachometer and O-QPS without tachometer),we develop the Cramer Rao Bounds to assess and compare the estimation performance limits for each model. Finally, wecompare the performance of all considered methods and highlight, in particular, the gain of the proposed solutions.

Automatic analysis methods of vibration parameters for diagnosis of marine rotating machines

Автоматическое диагностирование ответственных роторных машин по вибрации является одним из основных способов обеспечения их надежности и безопасности эксплуатации. Известные методы автоматической обработки вибрационных параметров в диагностических целях обладают ограниченной эффективностью в судовых условиях из-за нестабильной виброактивности машин в установившихся режимах их работы, вынуждающей завышать пороги опасности, и существенного взаимовлияния близко расположенных узлов и агрегатов, приводящего к ошибочным диагнозам. Для решения первой указанной проблемы предложен метод адаптации пороговых значений, позволяющий своевременно обнаружить и прогнозировать ухудшение технического состояния судовых роторных машин. Для решения второй проблемы предложен инвариантный к типу объекта контроля метод автоматического определения причин ухудшения технического состояния судовых роторных машин, который позволяет конфигурировать диагностические правила в табличном виде с возможностью учета влияния дефектов на вибрацию разнесенных в пространстве точек. Рассмотренные методы успешно используются в системах диагностирования роторного оборудования по вибрации. Automatic diagnosis of important rotating equipment using vibration signal is one of the main ways to ensure their reliability and operational safety. Known methods for automatically processing vibration parameters for machinery diagnostics have insufficient effectiveness on shipboard. The reason for this is unstable vibration activity in steady operating modes, which requires increasing thresholds, and the mutual influence of neighboring mechanical components and machines, which leads to erroneous diagnoses. The article provides methods to solve these problems. The first threshold adaptation method allows timely detection and reasonable prediction of marine machinery condition deterioration. The second automatic diagnosis method allows determining causes of this condition deterioration. The diagnosis method does not depend on the type of machine and uses the configuration of diagnostic rules in table form. In addition, this method allows to use defects influence on vibration at spaced control points. Declared methods are successfully applied in diagnostics systems of rotating machines.

A Data-Independent Genetic Algorithm Framework for Fault-Type Classification and Remaining Useful Life Prediction

Machinery diagnostics and prognostics usually involve the prediction process of fault-types and remaining useful life (RUL) of a machine, respectively. The process of developing a data-driven diagnostics and prognostics method involves some fundamental subtasks such as data rebalancing, feature extraction, dimension reduction, and machine learning. In general, the best performing algorithm and the optimal hyper-parameters suitable for each subtask are varied across the characteristics of datasets. Therefore, it is challenging to develop a general diagnostic/prognostic framework that can automatically identify the best subtask algorithms and the optimal involved parameters for a given dataset. To resolve this problem, we propose a new framework based on an ensemble of genetic algorithms (GAs) that can be used for both the fault-type classification and RUL prediction. Our GA is combined with a specific machine-learning method and then tries to select the best algorithm and optimize the involved parameter values in each subtask. In addition, our method constructs an ensemble of various prediction models found by the GAs. Our method was compared to a traditional grid-search over three benchmark datasets of the fault-type classification and the RUL prediction problems and showed a significantly better performance than the latter. Taken together, our framework can be an effective approach for the fault-type and RUL prediction of various machinery systems.