Energy is today the key to economic growth. Manufacturing activities or operation of complex products (train, plane etc.) may involve significant energy consumption. Energy resources are however limited and become more and more costly. Energy optimization of fixed installations (continuous industrial processes, manufacturing, computer data centers and massive storage ...) and mobile systems (mobility, transport, weapons systems, vehicles ...) is therefore an important issue to their economic competitiveness, their environmental impact as well as to control of energy resources (electricity, gas, water, oil, hydrogen, biomass ...). This should be reflected primarily by improving energy efficiency, i.e. reduce the amount of energy required to provide products and services. Indeed energy efficiency is considered as one main key lever to deploy sustainability, industrial ecology, and circular economy.
To face these issues, Europe has set ambitious goals to promote the development of new methodologies, new technologies or disruptive technologies that can improve the energy efficiency and reduce energy costs up to 60% in the most energy-intensive industrial sectors (manufacture of glass, cement, steel, refining ...). To this end, one of the most important solutions is to develop efficient maintenance strategies which cannot only avoid the failure of the system but also help to anticipate the growing up of global energy consumptions by replacing preventively and optimally high energy consumption components by low consumption ones, e.g. high technologies components.
The aim of this thesis is firstly to model energy efficiency behaviors of a component/machine based on different indicators such as its characteristics, energy consumption, physical deterioration behavior, etc. The energy efficiency of a component will be used as a main decision indicator for preventive maintenance of the component. To estimate the energy efficiency evolution, prognostic techniques will be developed and implemented. The second objective of this PhD work is to propose predictive dynamic grouping maintenance strategies for complex structure systems. The proposed maintenance strategies can help to select optimally components to be preventively maintained with lowest global energy consumptions and lowest maintenance costs in dynamic contexts. Moreover, in the framework of multi-component systems, dependencies between components often exist and they may have significant influence on the deterioration behaviour, energy consumption of components. These dependencies may impact indirectly to logistic support policies, maintenance strategies as well as production planning. In this PhD work, three different kinds of dependencies (structural, economic and stochastic dependencies) will be investigated.
Maintenance optimization: Maintenance involves preventive and corrective actions carried out to retain a system in or restore it to a correct/specified condition. Optimal maintenance policies aim to provide optimum system performance metrics (reliability/availability, safety, energy consumption, etc) at lowest possible maintenance costs.
Complex structure systems: system structures become more and more complex with a large number of components and very complex inter-connections between components which could be the mix of basic connections (e.g. series connections, parallel connections, bridge connections, etc).
Dependencies: Structural dependence applies when maintenance of a failed component implies maintenance of other components. Stochastic dependence occurs when the state of a component influences the lifetime distribution of other components or when components are subjected to common cause failures. Finally, economic dependence relies on the fact that the maintenance cost of a group of components does not equal the total cost of individual maintenance of these components.
Prognostic techniques/technologies: They provide the capability to predict the remaining useful life (RUL) of a machine before a failure occurs (or, one or more faults) given the current machine condition, the future and the past operation profiles.
Condition based maintenance (CBM) and CBM+: for which preventive maintenance decision is based on the observed system condition (e.g. deterioration level), has been introduced. Thanks to the rapid development of monitoring equipment which provides accurate information about the system condition over time, CBM becomes nowadays more and more popular approach in industrial application. Starting from the CBM policy, new trend of maintenance strategy has led to anticipate the failure. Hence, the degradation monitoring is followed by a prognostic step. Among such strategies, one finds CBM+, PHM (Prognostics and Health Management), proactive maintenance. The use of prognostic is dedicated to the forecast of the remaining useful life (RUL) before the failure occurs. Moreover, it is usually performed during the use of the material/system in order to adapt the maintenance policy. Indeed, the prognostic algorithms are fed with monitored data in order to predict the RUL in the current condition and the future operation profile.
Dynamic grouping maintenance: The main idea is to jointly perform several preventive maintenance activities in order to save maintenance costs and/or increase the system availability. Dynamic grouping maintenance becomes nowadays a powerful tool in the maintenance optimization of multi-component systems. Moreover, they can help to online update maintenance planning by taking into account short-term information which may occur over time.
• State-of-the-art on energy consumption, efficiency and prognostic approaches
• Proposition of energy efficiency models and related prognostic approaches
• State-of-the-art on dynamic grouping maintenance
• Proposition of dynamic grouping maintenance strategies for which energy efficiency is considered as a main decision indicator
• Numerical results: analytic and estimation calculation (Matlab)