The research project CIRCULAR is offering PhD thesis on :
Development of a generic model of degradation by exploiting data to evaluate all characteristics of monitored systems : Application to Li-Ion batteries in circular industrial systems at Grenoble Alpes University, G-SCOP Laboratory
SUBJECT DESCRIPTION :
In the framework of the CDP CIRCULAR (https://www.communaute-univ-grenoble-alpes.fr/circular-732721.htm), a thesis in the field of diagnosis is proposed (WP2). This call for PhD candidature, concerns the modelling and performance evaluation applied to the Lithium Ion batteries case study.
The CIRCULAR project aims at developing reliable circular industrial systems able to transform post-used products into new added-value products to prolong their life and ensure more sustainability . In those systems, adapted diagnostic and prognostic technics are essential to control the product obsolescence. But the technological heterogeneity of the products, with various lives, as well as the dynamics of change, accentuates the difficulty to the establish diagnosis and prognostics for making decision in the new value chain. The aim of proposed thesis is to provide reliable sensor networks as well as information retrieval modules to develop as accurate as possible a diagnosis and a health prognosis of the studied system. Information system, data collection and data integration will be considered for this.
In practice, the required data are not always available to establish such kind of diagnostic / prognostic and their use is often confronted with difficulties related to the following characteristics :
a. Heterogeneity. The data often come from different products life phases and each of those data is likely to add value to the process of diagnosis and prognosis. It is therefore necessary to identify and use each data into the appropriate models, in relation to the different life stages and modes with operating conditions attached to it.
b. Geographical and environmental influences. The behaviour of the monitored systems can change according to their environment and operating conditions, which may change each significantly during the lifetime of the device. This problem may introduce significant perturbations in terms of prognostic and therefore in the economic plan.
c. Quantitative insufficiency. Some data, essential to the establishment of decision models can be rare or censored. Therefore, the effects of the censured data and the disruption of the decision models will be carefully considered.
Developments in this thesis will be structured into four main phases.
The first phase is devoted to the characterization of the requirements and essential metrics for the diagnosis of the studied system. Particular attention will be given to the information system, data collection and data integration which will include sensor networks reliability through proper architecture. Concerning the particular application, it will be necessary to identify the sensitive data allowing evaluating the state of health of the batteries.
The second phase involves the integration of data. We aim to integrate data from different sources (sensors, information from experience feedback (REX)) to develop models of reliability and degradation by taking into account factors related to operating conditions, operating modes and environmental influence parameters.
Construction and modelling knowledge is the third phase. In this context, it is necessary to formalize the reuse of knowledge, which is one of the scientific challenges.
The fourth phase involves the use of various data (sensor, REX) for prognosis and state of health control. This is a fundamental function to planning the reconfiguration for reuse or disassembly.
For further information please download the file