
Due to the wear of machine components, performances degrade over time. Fidia aims to monitor this degradation and prevent it, to avoid any changes in the performances of the machines.
Early diagnosis and prevention of machine component failures are key points and can become a significant advantage to improve quality and competitiveness. As known in the literature and practice in the field, most common failures are related to the movement components and sub-systems of the machine, such as spindle, racks and pinions, balls screw, bearings, gears, shafts, belts, guideways, etc.
In some case, a component suffers a major and unexpected failure, resulting in an abrupt termination of machining. Other non-blocking malfunctions could occur with either performance degradation or out of tolerance or scrap parts. All these aspects influence final quality and costs of products. A Zero Defects Strategy is mandatory to achieve and maintain production targets.
FIDIA empowered its numerical controls with a monitoring system allowing not only to monitor and analyse data coming from the CNC, but also from the PLC and sensors of the machines.
The installed systems allow to analyse and monitor any shift in the performances of the machines over time and enable the users in understanding the causes and the contributing factors.
zMachineAnalytics
zMachineAnalytics acts as the back-end, and allows the user to define different analytics and work in conjunction with the models developed with several ZDMP components. The models or algorithms to define the alarms are developed using the Prediction and Optimization Runtime, and them uploaded to the platform through AI Analytics Runtime, which creates the corresponding API functions in the API Gateway, which are the ones then exploited by zMachineAnalytics. All the results are finally published in the Message Bus through MQTT protocol.
zMachineMonitor
zMachineMonitor acts as the front-end and presents two main functionalities. The first one is to show in a user-friendly way the results in real time of the alarms analyzed with the zMachineMonitor, allowing the operator to have a clear and updated view of all the problems that may have arisen in the machines. The second functionality has the aim of helping the operator even further, allowing the visualization of the parameters related with each alarm. This is achieved thanks to the integration of the Digital Twin ZDMP component, which makes a contextualization of all the machines in the plant and registers the historical data of the desired sensors.