The worsening climate crisis is one of the greatest challenges of our time. Climate protection cannot function without industry. That’s why the industry is showing increasing willingness to reduce its own CO2 emissions to a minimum. With rising CO2 and electricity prices, there is also an increasing economic incentive to do so. Energy management within industry is therefore more relevant than ever before. The energy consumption of a machine, especially in a complex production process, is often unclear and the efficiency difficult to assess. Maintenance, on the other hand, is associated with many problems, as failures cannot be predicted. These challenges are as old as the industry itself. Energy management and maintenance are closely related issues. Consequently, these two areas should also be dealt with together. With the help of Industry 4.0, energy management and maintenance can be viewed and analyzed in relation to each other for the first time. This offers many opportunities and starting points for making industry more efficient and sustainable.
Energy management
Energy management is about using as little energy as possible for a given process. To achieve this, it is essential to know as precisely as possible where how much energy is used and via which energy carrier. The energy carriers commonly used in industry are:
- Fuels
- Electrical energy
- Thermal energy
- Compressed air
- Kinetic energy
In many cases, these energy sources are interrelated in chains of action, e.g., a compressor for producing compressed air can be operated either electrically or with an internal combustion engine. However, any conversion is lossy and is therefore avoided if possible. All of this has long been considered in the design of any plant. So where does the benefit of information from existing plants come into play, how can energy consumption in industry now be made more efficient? As diverse as the energy sources used in industry are, the solutions for effective energy management must also be flexible and creative.
The best way to illustrate this is with a simple example. The optimal energy consumption of a heating system depends heavily on external factors such as the weather. By using a neural network, the system can “learn” how to achieve the desired temperature with minimal energy consumption under the expected external conditions. This is done by collecting data on the fuel consumption of the heating system and matching it with historical data on the weather. This reconciliation makes it possible to create forecast models for energy consumption.
Transferred to industrial processes: Using data from a process, information about energy consumption can be obtained, depending on external influences such as the product currently being manufactured, variation in the materials processed, but also the current state of the machine/plant (see Fig.1). This generates an enormous added value for the operator of the machine, as suboptimal conditions and settings become quickly apparent and can be adjusted. Energy consumption becomes transparent and predictable, enabling data-based energy management.
Fig. 1: forecasting models through neural networks
Whether it is fuel consumption, power consumption or other energy sources, it is possible to select all conceivable controls of a machine and use them. Even if required parameters are not provided by a control system, sensor technology can be integrated. Thus, with the implementation of digital technology in combination with appropriate mathematical approaches, such as artificial intelligence, machines of any type and age can be optimized.
Maintenance
The topic of predictive maintenance is probably the most obvious “low hanging fruit” when it comes to use cases for industry 4.0. Of course, effective maintenance is also important for resource-saving and thus efficient use of machines. The question is: what does “effective” mean? A failure of a machine or component can occur suddenly and lead to unexpected repair and procurement costs or even personnel damage. Unplanned breakdowns then lead to long downtimes and thus loss of income. On the other hand, a maintenance regime that is too tightly scheduled leads to frequent downtime and an excessive need for spare parts. In addition, maintenance is also an important factor for the quality assurance of a product. Certain parts, for example, can lead to production defects due to unmonitored wear. It is often difficult to determine where these defects occur. Effective maintenance must therefore be forward-looking on the one hand, but must also be able to recognize errors quickly in order to act accordingly. Anyone who has internalized this can get the maximum out of their machines and contribute to resource-efficient production. But how can maintenance be made more efficient?
With data, artificial intelligence and creativity! Through the cloud plug, data from a machine can be collected and analyzed. Process anomalies, such as “peaks”, i.e. outlier values, are quickly noticed (see Fig. 2). It can then be analyzed whether this is a measurement error or whether there is a correlation with a defective or worn part. As in energy management, forecasting models with the help of artificial intelligence are also used here. Data is collected and analyzed over a longer period of time. The current state of the plant is then compared with the plant history. The AI detects trends early on that indicate part failure or wear and provides appropriate warnings. This enables predictive maintenance.
By using data-based technologies, maintenance can be readjusted during the work process on the one hand, and it also becomes more predictable in the long term. Industry 4.0 transforms maintenance from an experience-based gamble into an economic factor.
Fig. 2: Process anomaly in a charging process of a battery
Why maintenance and energy management are considered together
Maintenance and energy management are closely interrelated. How can the condition of a machine influence energy consumption? How can energy consumption provide information about the condition? This is best illustrated by an example. A saw needs a sharp saw blade to work effectively. The duller the saw blade, the more energy is needed to get the cut while maintaining the product. So in this example, the energy is directly related to the condition of the saw. If you now measure the power consumption and compare it with historical data, which power consumption can be expected with which condition of the saw blade, you can make direct conclusions about the current condition. This helps to make an informed and economical decision (see Fig. 3).
But other energy sources also provide a deep insight into the condition of the machine. Compressors pressurize air and thus generate the energy carrier compressed air, which is still essential in many plants. The compressed air is transported via pipes. Leakages, e.g. small holes or leaking connecting pieces, can lead to pressure loss. Pressure loss means energy loss. Anyone who has ever inflated a bicycle tire with a leaky pump will understand this. These leaks can go unnoticed for a long time, not only on the bicycle pump but also in the industry, but if you have the recorded data of the compressed air consumption directly available, you can quickly detect these anomalies and trace them back to the leaks.
Fig. 3: Interdependence of energy consumption and maintenance
Energy management and maintenance are interdependent. A machine that is not optimally maintained will also not use its energy optimally. A machine that does not use its energy optimally says a lot about its condition. Energy management and maintenance are two topics that should always be considered together. Today, digitalization provides us with the necessary information and tools for this. Only those who take advantage of this and understand these two areas as two sides of the same coin and include them in problem solving can ensure optimal overall equipment effectiveness (OEE). Industry 4.0 offers all the possibilities to perceive energy management and maintenance in one as an opportunity for a more efficient and sustainable industry. So there really are no limits to creativity.
