Project title: Predictive maintenance for energy efficiency
Financing: the state budget
Name of the Program in PN III: Program 3 – European and International Cooperation
Subprogramme name: Subprogramme 3.5 – Other European and international initiatives and programs – EUREKATraditional projects, EUREKA-Cluster, Eurostars
Project type: EUREKA Traditional
Total value of the Contract: 2,041,880.00 lei
Source 1 – from the state budget: 1,354,000.00 lei
Source 2 – from other attracted sources (own contribution) 1: 687,880.00 lei
Contract duration: 24 months

Contracting Authority: Executive Unit for the Financing of Higher Education, Research, Development and Innovation
The PREVENTION project aims to develop a platform that will integrate Digital Twin technology and predictive maintenance to increase energy efficiency and provide contextual information for the design, operation and reconfiguration of the production line. The novelty will be presented as a platform that will provide procedures for increasing energy efficiency in the field of production, based on advanced maintenance. The new model will replace the production line with products, as in most current approaches to energy efficiency.
This goal will be achieved through an innovative combination of artificial intelligence techniques with advanced control algorithms, through 2 main approaches:

  • Holistic approach to optimizing the energy consumption of the entire production flow life cycle, based on intelligent prediction algorithms. We will develop methods and tools to increase energy efficiency, starting from a production flow, analyzed from the design phase (with an innovative character compared to typical product-focused approaches) and continuing with methods of optimized process control and production planning. .
  • Monitoring the health of resources for real-time anomaly detection and energy impact assessment. Whenever such behavior is identified, this part of the system is isolated and the platform strategy for analysis or maintenance procedures is started.
    The main result is that the platform that will be developed throughout the project will integrate Digital Twin (DT) technology. DT will provide the ability to respond to stakeholder needs by developing predictive maintenance strategies and diagnostic algorithms that can be proposed to the maintenance team after real-time validation. The architecture of the platform will be based on a service approach, including data collection, fault identification, diagnosis, isolation and selection of maintenance methods. An innovative result is also the simulation and testing method, including the verification of the results of the maintenance intervention.

Based on the needs of the PREVENTION platform, such as continuous monitoring of asset health, process characterization and decision making, early detection of issues that may affect equipment or process operation, verification of adequacy of Digital Twin intervention solutions, enforcement proactively correct, production owners will have sufficient arguments about the benefits of operating their services.
The main result of the project is the modules that will be added to the existing platform to provide new functions for increasing energy efficiency, through an approach based on predictive maintenance, to prevent suboptimal operation of process equipment. This functionality will be achieved by integrating DT technology to create a state model of process consumption. DT will offer the possibility to respond to the needs of the beneficiaries by developing predictive maintenance strategies and diagnostic algorithms that have been previously validated on the real-time model.
To ensure the replication of the solution to other processes, the project will develop a methodology for the design and selection of maintenance instrumentation based on cost and performance criteria.

Modules will be developed for data collection (from system equipment and process data), modules for data analysis, and procedures and functions for active synchronization between planned and executed actions.
The proposed strategy will use predictive algorithms and context models, correlated with risk and uncertainty assessment. The platform will analyze the process data and compare it with the results obtained from the Digital Twin model, to increase the efficiency of the processes.

In stage 1 of the Prevention project, a laboratory assembly was analyzed in detail consisting of a flexible assembly line, built of several individual stations, acting as interoperable and reconfigurable line modules. The lab includes sensors, motors, conveyors and robots. SIS will handle data acquisition, retrieval and analysis to generate a preventive and predictive maintenance plan. Following the analysis, the requirements and architecture of the platform were defined, as well as the instrumentation and data acquisition structures.