The safety and performance of engines such as Diesel, gas or even wind turbines depends on the quality and condition of the lubricant oil. Assessment of engine oil condition is done based on more than twenty variables that have, individually, variations that depend on the engines’ behaviour, type and other factors. The present paper describes a model to automatically classify the oil condition, using Artificial Neural Networks and Principal Component Analysis. The study was done using data obtained from two passenger bus companies in a country of Southern Europe. The results show the importance of each variable monitored for determining the ideal time to change oil. In many cases, it may be possible to enlarge intervals between maintenance interventions, while in other cases the oil passed the ideal change point.
31 March 2022 in Eksploatacja i Niezawodność – Maintenance and Reliability
Predictive maintenance is very important for effective prevention of failures in an industry. The present paper describes a case study where a wood chip pump system was analyzed, and a predictive model was proposed. An Ishikawa diagram and FMECA are used to identify possible causes for system failure. The Chip Wood has several sensors installed to monitor the working conditions and system state. The authors propose a variation of exponential smoothing technique for short time forecasting and an artificial neural network for long time forecasting. The algorithms were integrated into a dashboard for online condition monitoring, where the users are alerted when a variable is determined or predicted to get out of the expected range. Experimental results show prediction errors in general less than 10 %. The proposed technique may be of help in monitoring and maintenance of the asset, aiming at greater availability
Forecasting has extreme importance in industry due to the numerous competitive advantages that it provides, allowing to foresee what might happen and adjust management decisions accordingly. Industries increasingly use sensors, which allow for large-scale data collection. Big datasets enable training, testing and application of complex predictive algorithms based on machine learning models. The present paper focuses on predicting values from sensors installed on a pulp paper press, using data collected over three years. The variables analyzed are electric current, pressure, temperature, torque, oil level and velocity. The results of XGBoost and artificial neural networks, with different feature vectors, are compared. They show that it is possible to predict sensor data in the long term and thus predict the asset’s behaviour several days in advance.
Accurate precipitation data are essential for effective water resource management. This study aimed to correct precipitation values from the ERA5_Ag reanalysis dataset using observational data from 20 meteorological stations located in the Tensift basin, Morocco. Five machine learning models were evaluated: MLP, XGBoost, CatBoost, LightGBM, and Random Forest. Model performance was assessed using RMSE, MAE, R2, and bias metrics, enabling the selection of the best−performing model to apply the correction. The results showed significant improvements in the accuracy of precipitation estimates, with R2 ranging between 0.80 and 0.90 in most stations. The best model was subsequently used to correct and generate raster maps of corrected precipitation over 42 years, providing a spatially detailed tool of great value for water resource management. This study is particularly important in semi−arid regions such as the Tensift basin, where water scarcity demands more accurate and informed decision−making.
22 January 2021 in WSEAS TRANSACTIONS ON SYSTEMS AND CONTROL
The importance given to the maintenance in the industrial world has grown over time, with new methods, new procedures and new challenges, due to the availability of new technologies. This paper focus on a global survey about predictive maintenance tools that support predictive maintenance, from the time series and decision trees until Artificial Intelligence. The approach of the several tools that can help the prediction is holistic, because new tools do not eliminate the importance of the old ones: they are complimentary and each new tool that is developed add potential for a better prediction. Additionally, it must be emphasized that some tools, that seem new are, in practice, old tools with new and powerful computational devices, assuming a new and strategic importance nowadays.
The maintenance paradigm has evolved over the last few years and companies that want to remain competitive in the market need to provide condition-based maintenance (CBM). The diagnosis and prognosis of the health status of equipment, predictive maintenance (PdM), are fundamental strategies to perform informed maintenance, increasing the company’s profit. This article aims to present a diagnosis and prognosis methodology using a hidden Markov model (HMM) classifier to recognise the equipment status in real time and a deep neural network (DNN), specifically a gated recurrent unit (GRU), to determine this same status in a future of one week. The data collected by the sensors go through several phases, starting by cleaning them. After that, temporal windows are created in order to generate statistical features of the time domain to better understand the equipment’s behaviour. These features go through a normalisation to produce inputs for a feature extraction process, via a principal component analysis (PCA). After the dimensional reduction and obtaining new features with more information, a clustering is performed by the K-means algorithm, in order to group similar data. These clusters enter the HMM classifier as observable states. After training using the Baum–Welch algorithm, the Viterbi algorithm is used to find the best path of hidden states that represent the diagnosis of the equipment, containing three states: state 1—“State of Good Operation”; state 2—“Warning State”; state 3—“Failure State”. Once the equipment diagnosis is complete, the GRU model is used to predict the future, both of the observable states as well as the hidden states coming out from the HMM. Thus, through this network, it is possible to directly obtain the health states 7 days ahead, without the necessity to run the whole methodology from scratch.