Remaining Useful Life: Predictive Maintenance
Remaining Useful Life (RUL) states the duration of a component to reach its failure [1]. By taking RUL into account, engineers can schedule predictive maintenance, optimize operating efficiency, and avoid unplanned downtime. For this reason, estimating RUL is a top priority in the predictive maintenance program. In this article, I would tell such an example in implementing RUL for predictive maintenance.
What is Battery?
Battery is a device used to store electrical energy in the form of chemical energy by utilizing electrochemical reactions [1]. Based on its reactions, the battery is divided into two kinds:
1. Primary Battery
A battery that can only be used once and cannot be recharged. This matters because electrochemical reactions that occur are irreversible [1]. Example: an alkaline battery.
Figure 1: Primary Battery (www.sparkfun.com & www.id.aliexpress.com)
2. Secondary Battery
Batteries that can be used and recharged several times in which the cell reactions are reversible [1]. Example: Li-ion batteries.
Figure 2: Secondary Battery (www.amazon.com & www.apple.com)
Battery Parameters
Before we discuss it more, let me explain the two common parameters of the battery.
1. State of Charge
State of Charge (SOC) is defined as a ratio of current battery capacity to battery capacity before discharge [1]. Mathematically, SOC battery can be obtained as in equation (1) with SOC states the battery charge condition, Cr states the current battery capacity in Ah, and Co states the battery capacity before discharge in Ah.
SOC = Cr/Co …………………………………………………………..…..(1)
2. State of Health
State of Health (SOH) of the battery is the ratio between the maximum capacity of the battery before discharge to the nominal capacity. The nominal capacity of the battery is given by the battery manufacturer that shows the maximum capacity of the battery. SOH indicates the degree of degeneration of a battery [1]. SOH battery can be obtained as in equation (2), where Co expresses the maximum capacity of the battery before discharge in Ah and Cn denotes the nominal capacity of the battery in Ah.
SOH = Co/Cn ………………………………………………………………(2)
In general, the difference between SOC and SOH battery is SOC battery explained short-term changes in battery capacity, while the SOH battery explained long-term changes in battery capacity [1]. For more details, the difference between both of them can be seen in Figure 3 below.
Figure 3: The Difference Between SOC and SOH [1]
Remaining Useful Life of The Battery
In the battery field, RUL is approached with the remaining useful cycle which states how many cycles can occur in the battery from the current conditions to the end of its failure [1]. One cycle states the change in the charging state until the next charging state according to the scenario of the Battery Management System (BMS). Illustration of the RUL battery can be seen in figure 4 below.
Figure 4: Remaining Useful Life Prediction [1]
There are several prognostic prediction methods used for determining the RUL [2].
Figure 5 : RUL Degradation Methodology
1. Model-Based
RUL prediction is applicable to Statistics and Computational Intelligence (CI) approaches. These models are derived from the usage, configuration, and historical ‘run-to-failure’ data and applicable to maintenance decision making. Such components that are analyzed and documented in the literature include gear plates and bearings from manufacturing industries. The model-based methodology is often used to predict RUL thereby.
2. Analytical-Based
The analytical-based RUL prediction approach represents the physical failure technique. The analytical-based model refers to an understanding of techniques that aid reliability estimates of the physics-based model attributed to Physics-of-Failure (PoF), the physical science of components and generated experimental equations. Failure events such as corrosion of components and crack by fatigue are used to predict the RUL.
3. Knowledge-Based
This model is a combination of Computational Intelligence (CI) and experience. The knowledge-based approach relates to the collection of stored information from Subject Matter Experts (SME). It can be seen as a service performance system for service delivery based upon the principles of service feedback for analysis. Parameters of reliability are estimated using an experience-based approach to information gathered from understanding the asset.
4. Hybrid-Based
The hybrid model uses several techniques for RUL prediction to improve accuracy. The hybrid model uses parametric and non-parametric data to perform RUL prediction and to improve accuracy. It predicts RUL individually and through methods based on probability theory facilitates the fusion of two or more RUL prediction results to achieve a new RUL.
Conclusion
The effective prediction of RUL for manufactured products within the Industrial Product-Service System is a key factor in sustainable service delivery. This one is similar with PTI’s vision . . .
By applying predictive maintenance, it will reduce regular maintenance costs and improve operational efficiency. Thank you for reading this article and have a nice weekend :)
Source:
[1] Beny, M.A., Fajar, M.N.R. (2019). The Development of Remaining Useful Life Prediction System on VRLA Battery using Support Vector Regression. Bandung: Institute Technology Bandung
[2] Okoh, C., Roy, J. (2014). Overview of Remaining Useful Life Prediction
Techniques in Through-Life Engineering Services. United Kingdom:
Elsevier.
