Dynamic Control Chart Research Articles

A novel RUL prediction method for rolling bearings based on dynamic control chart and adaptive incremental filtering

Abstract Degradation of rolling bearings typically consists of two stages: a stable stage (Stage I) characterized by stable fluctuations in the health indicator (HI), and a degradation stage (Stage II) where early damage leads to HI degradation, eventually reaching the failure threshold. Therefore, to achieve online RUL prediction for bearings, three aspects should be studied: 1) Degradation modeling; 2) Inter stage change point identification; 3) Online degradation state updating. Firstly, a two-stage degradation model is constructed by simultaneously considering inherent randomness, individual differences, and measurement errors. Then, a dynamic statistical process control (SPC) method is proposed to identify the change point from Stage I to Stage II. The SPC is designed to dynamically control limits based on the bearing's condition monitoring (CM) data to prevent false alarms. An adaptive incremental filtering (AIF) is proposed to update the degradation states by simultaneously considering the state increment and the dynamics of the system noise and measurement noise. The effectiveness of the proposed method is validated on 16004 bearing test data and XJTU-SY bearing data. Results show that the proposed method can accuracy identify the change point and improve the accuracy of the prediction result during stage II.

Metric Based Dynamic Control Charts for Edge Anomaly Detection in Factory Logistics

The optimization of transport logistics in production environments is a holistic task for the factory of the future. Autonomous guided vehicles that perform transport jobs in factories are facing this challenge and have to detect, react and prepare to unforeseen changes and anomalies in the production system. Due to data protection concerns, details like production plans are often not available for an external transportation system. Hence the anomaly detection has to be based on self-collected and observed data of the transport system like occurred transport needs or the evolution of internal metrics. In this paper we infused a production system with manufacturing process anomalies and demonstrate a detection based on the observation of transport needs to overcome the gab caused by restricted information. For that detection we extended classic control charts to work with expected values based on learned dynamic production characteristics. The system sets a tolerance field as narrow as possible around dynamically determined values, resulting in an average precision of 95% for detection unusual number of transport jobs.

Study on the Dynamic Control Chart of Pumping Wells

In order to improve the pump efficiency of pumping wells, by means of the basic formula of pump efficiency, the mathematical model and boundary conditions of dynamic control chart is obtained, and the pump efficiency of pumping wells is drawn. Analysis shown that the pump efficiency is sensitive to the water content and pump leak rate. The higher the water content is, the higher pump efficiency is. The pump efficiency will be reduced with the increasing of pump leak rate. The dynamic control chart of class II block of A oilfield is given in the paper. After taking the measures, the pump efficiency of pumping wells is obviously improved. The dynamic control chart drawn by this method can provide a scientific basis for improving the pump efficiency of pumping wells in oilfield.

Integrating Dynamic Control Charts and Maintenance Policies

Statistical process control and maintenance optimization have an intuitive relationship, but the literature has largely treated them as separate entities. We formulate a partially observable Markov decision process (POMDP) that integrates these two fields by jointly optimizing the conditioned-based maintenance policy and dynamic control chart parameters used to observe the system. We present an insightful numerical example that demonstrates the flexibility of the policy as well it promising structural properties and practical implementation.

Dynamic control charts for finite production runs

A dynamic programming model is developed for the optimization of statistical process control for finite production runs. The case of one-sided control charts for variables is examined first. A dynamic chart allows all three parameters, namely the sampling interval, sample size and control limit location, to change during production, as sample information becomes available and the state of the process is updated using Bayesian methods. The economic performance of the optimal dynamic chart is compared to that of the optimal static control chart through a number of numerical examples. It is concluded that substantial improvements may result from the adoption of dynamic charts. The benefits of direct observability of assignable causes are also estimated. Then, a generalized theoretical formulation is provided for two-sided dynamic control charts. The increased size of the state and decision spaces leads to a discussion of practical implementation issues. An example of computing transition probabilities between states of different stages is given. The major conclusion of this research is that significant cost savings may be realized through the application of dynamic control charts and even greater benefits may be reaped by investing in process understanding and improvements.