Towards a smart world
To prevent breakdowns is good but the goal must be to avoid machine failures from the beginning. This has been the aim of the DALOG GmbH for almost two decades. Ever since their online condition monitoring systems helped to maximise the availability of machines by making precise analysis and supporting the plant team remotely with failure predictions, maintenance preparation support, and even advice for process improvement. Yet, the outcome on the production can only be assured by an interconnected monitoring solution that covers the full value chain of the cement production process.
Process correlated condition monitoring
Creating save islands in the plants by monitoring the most critical assets with acceleration and temperature sensors allows to predict failures on gears and bearings before they end in a catastrophic breakdown. A tooth failure on a gear can be detected before it breaks off and turns into a loose cannon within the gearbox. A bearing with upcoming pitting can be ordered in time. Statistical models even allow to predict the remaining lifetime of a machine component. In general, failures are detected at an early stage, which gives time to act and plan. Plants benefit from less unplanned stoppages and a minimisation of secondary damages.
The next logical goal is to increase the time between failures, by optimising operational stability. The torque measurement on rotating shafts has shown to precisely measure the real load on the machine and a high frequency measurement shows peaks not visible in the motor power. Consequently, it is a valuable tool to evaluate operational stability and to detect dangerous overloads on the machine. This is especially important for machines with a high dynamic process like vertical roller mills and roller presses, where operational instability can often be related to issues in the grinding elements or the process, like an improper feed. For a root cause analysis in the process, the online condition monitoring system needs to be able to communicate with the PLC to receive parameters like feed rate, pressures and motor power; sending alarms and trends to the DCS, helps to integrate the operators in the project, which gives them the necessary information to take appropriate measures to prevent machine failures or instabilities.
One of the big challenges for the cement industry towards a successful transition to the digital factory, will be the standardisation of communication protocols. OEM proprietary protocols dominate the cement plants and given that this situation will not change soon, the DALOG condition monitoring system supports most of the fieldbus protocols, to establish a bi-directional communication with the PLC. The key elements to prevent failure in the future are real load measurements with a torque sensor, the data exchange (process and condition) between the condition monitoring system and machine control, and the integration to the operator.
Online data acquisition for all machines
The industry has made great progress in the last years with its maintenance strategies. The FMEA (Failure Mode and Effects Analysis) today is a standard tool evaluate risks and thus even today many of the most critical machines, like mills and kilns, are equipped with an online condition monitoring system and respectively covered by the predictive maintenance strategy. Nevertheless, less complex and smaller equipment, like fans, conveyors, or bucket elevators, often fall through the net, besides its clear impact on the production. The maintenance strategy for those machines in today`s plants are either "run to breakdown", preventive, or condition based with temporal measurements on the equipment. They all have in common that the initial investment is very low, but the ongoing operational costs are soaring. The run to failure strategy bears a lot of risks and will cause unplanned stoppages and production loss.
Preventive maintenance reduces the risk, but relies solely on life-time estimations or manual inspections to schedule maintenance intervals. This either results in changing machine parts too early or too late, one ends in higher spare part investment, the other in production loss. Condition based maintenance intervals can be achieved by mobile measurements. Yet, the quality and reliability of an analysis with mobile equipment has a high variance. They are time consuming and must be carried out by trained personal that needs to fulfill a rigid schedule taking those measurements. Risks involve selecting a non-optimal measurement position, not considering different machine production situations at the time of the reading, or in the worst case, not taking the measurement after all. Trend analysis and process correlation is in the best case suboptimal.
The cement plant of the future will solely rely on a predictive and proactive maintenance strategy, which bases its decisions on data from online condition monitoring systems. The advantage lies on the hand: while reducing time consuming field work, the plant can count on more precise measurements and a higher data density, thus a higher reliability of their predictions and consequently a higher availability of the plant.
New challenges for maintenance
As the data density increases and analysis algorithms improve, what about the typical tasks for the maintenance team? While it is possible to measure the impacts of the tropical rain fall on the feed of a vertical roller mill in a cement plant in the Philippines, or to detect the effects of sand (silica) on the operational stability in a roller press in the Mid-West, it is yet a great task to integrate all the external influences into an algorithm that could replace the judgement of a trained expert. On the other hand, as reliability increases and field work decreases, more time and effort will be spent on performance optimisation and the collected data will build the foundation to plan, carry out and measure the outcomes of such initiatives. Experienced vibration analysis, maintenance, and process experts from DALOG, guide already today many plants to interpret the data and optimise their performance.
Online condition monitoring
How can online condition monitoring go even further and support a company-wide strategy? Wikipedia defines strategy as "a high-level plan to achieve one or more goals under conditions of uncertainty". It is clear how online condition monitoring continuously provides information to maintenance to eliminate uncertainty. But why stop there? The DALOG online condition monitoring solution acquires data in high resolution to evaluate machine condition, operational stability and process parameters that will help to evaluate the overall performance. The latest software generation DALOG Busy Bee calculates KPIs and statistics that are visualised in customisable dashboards. It is possible to display statistics about machine availability, energy consumption and operational stability, to name just a few. Surely, the access to the data is not limited locally to plant facilities. An encrypted connection to the cloud permits to access the data worldwide from a laptop, smartphone or tablet. This allows regional managers and headquarters to gain insight about the performance and condition of cement plants of a region, or country, thus any production strategy within the organisation will be exposed to less uncertainty.
A look back and the possibilities we have today
DALOG was founded in 1998. Back then, it was state-of-the-art to equip the monitoring system with a GSM module to dial up and establish a 9.6 kBit/s connection. Few people back then could have imagined that today most plants have access to the internet 1,00,000 times faster (1 Gbit/s). I would not dare to make a prediction for the technology we will have in 20 years, but I can say what is possible since today.
The DALOG Plant Protection Concept is a bottom up approach where on the lowest layer data is collected and stored on the cloud.
The massive amount of data is processed to be useful for different stakeholders within the same organisation. The maintenance team can work with precise failure analyses and can plan maintenance intervals based on machine condition and proactively work on increasing the time between machine failures.
The production team gains statistics and models for production planning. The plant manager can evaluate overall plant performance comfortably from his smartphone. In the end, online condition monitoring will build the foundation, for a seamless integration of data-driven decision making within the whole organisation.
Courtesy: DALOG, Diagnosesysteme GmbH, Muehlbachstr. 21