Industrious bees demonstrate to us that each bee can autonomously complete small tasks. Through ingenious, minimally communicative methods, they can organize themselves and compensate for errors within the entire organization. Is this distributed intelligence an example of interconnected motor techniques, or is it the centralized nervous system of our organisms, where all information is concentrated at a single point, processed, and leads to action? We attempt to clarify this question from the perspective of an electrical technology manufacturer.
At first glance, the situation with electric motors seems clear: a servo motor simply obeys commands. It drives at a specified speed, meeting predetermined torque and position specifications. The higher-level control system, acting as the central nervous system, takes care of everything. Disturbances are identified and compensated for through centralized control.
Many interconnected systems, such as production machinery, currently operate on the principles of centralized systems. The advantages are obvious: programming is done through a centralized system, and debugging is also done in a centralized location. Furthermore, when higher-level entities (such as ERP systems) need the latest production data or need to transmit information, there is a centralized information exchange venue.
Mirroring this to the central nervous system of organisms, such as mice, means that organs are controlled by the central nervous system; muscles that move are activated; and eyes, ears, and skin monitor the environment. The brain handles secondary tasks, such as breathing and heartbeat, but also more complex tasks, such as escaping enemies. A beehive cannot function on this principle. 100 different decisions, such as "Should I fly to the red flower field or the yellow flower field?" or "Can our 20 bees drive away the bumblebees?" require decisions to be made from a distant brain. Therefore, the communication lines are too long and the tasks too complex to be relayed to the bees using simple communication tools.
Have we found the answer to determining whether distributed or centralized systems are superior? Therefore, the following will apply to: Fixed systems – centralized or multi-purpose systems – distributed?
This is worth examining closely. If bees had a 5G mobile connection to a centralized "brain," they could make decisions without requiring as much intelligence. Conversely, a mouse's heart, stomach, lungs, and muscles also possess their own intelligence. Therefore, the central nervous system can handle the stress of simple tasks. The nervous system can then focus on more complex tasks, such as finding food.
Whether in nature or in electrical engineering, the following applies: centralized intelligence and efficient communication between components reduce the intelligence required by the components themselves. However, if the components are intelligent, communication becomes easier, and centralized intelligence is mitigated or can be completely ignored.
If you look at the automotive technology market, you'll find two options: robust centralized systems that communicate effectively with motor components, and intelligent and powerful distributed drive systems. Both solutions are easy to implement and deploy. When deciding between a centralized or distributed system, the following guidelines are important:
Task complexity
Low-cost processor power consumption enables simple distributed drive systems to become intelligent systems, allowing a variety of small to medium-complex tasks to be completed without superior control. They read analog and digital data from sensors and communicate with each other. Tasks are distributed among motors. Setup is done via a human-machine interface (HMI). Many packaging machines, testing equipment, and specialized machines operate simultaneously without superior control. Each component of the machine can monitor itself. Communication can be reduced to a minimum. Economical classic fieldbuses are sufficient for communication between them.
For more complex tasks, such as those where many mechanical parts are interconnected, the need for synchronization is high—in most cases, centralized mechanical control from inventory is worthwhile. However, in these cases, it also makes sense to implement decentralized components for autonomously executed processes. One example is solar thermal machinery: permanent single-mirror adjustments can be performed entirely decentralized.
What are we using now?
Whether a mechanical structure is centralized or decentralized often depends on the inherent limitations of that structure. A system is well-known, and its design is geared towards existing systems. Therefore, the construction of new systems tends to be similar. Understandably, engineers dislike studying the communication types and system architectures of every new machine concept. Especially when orders are full, companies prioritize rapid implementation over perfect solutions. This is clear. However, it is worthwhile because the total cost is not always readily apparent.
System cost
System costs are difficult to define because some costs are invisible, or economical solutions may lead to high subsequent costs. Here are two less obvious but relevant cost factors:
Electrical control box: How much space is needed inside the electrical control box? What dimensions are required? Here, a decentralized solution may be appropriate. In a decentralized solution, heat loss is also dispersed, so forced cooling is not necessary.
What seems like the most reasonable solution at first glance could become a bottomless pit if the entire system isn't considered from the outset. Are all components of the planned system available? Can't the required components be directly implemented into the network? Depending on this, it might require acquiring expensive gateways or complex, specialized programming.
How did Dunkermotoren, as a manufacturer of electric motor technology, prepare?
Prepare two parts
Regarding connectivity, Dunkermotoren has prepared both components. Dunkermotoren motors can be implemented as simple slaves in a centralized architecture or perform distributed tasks. In both cases, monitoring the motor's functionality and protection against permanent damage are implemented in each motor. This makes sense. Each motor knows its own characteristics and how to withstand overloads. Hardware and software protection mechanisms directly mounted on the motor allow it to operate normally without damage, enabling it to reach its physical capacity.
Dunkermotors uses several of the most important communication languages, currently including CANopen, Profibus, Profinet, and EtherCAT. Like most component manufacturers, Dunkermotors faces a complex mix of fieldbus and industrial Ethernet communications. To integrate into a centralized communication system, Dunkermotors needs to implement their respective interfaces. Like many other component manufacturers, Dunkermotors anticipates a manufacturer-independent standard.
Future Interconnectivity
Dunkermotoren asked himself how motors would need to be interconnected in future distributed and centralized solutions. Simply saying that all common fieldbuses, industrial Ethernet, and all wireless standards need to be covered is not enough. The question is, what data must be transmitted to where, and at what speed? Topics such as predictive maintenance, cloud-based application analytics, remote service, and pay-as-you-go will be crucial. In these cases, data needs to find a path from the machine to the cloud or directly to the manufacturer. If Ethernet-based systems don't have the accompanying security risks, or mobile communication-based systems don't have the associated network coverage and cost issues, it remains exciting. In fact, if these issues can be resolved, the customer benefits of the so-called "Internet of Things" (IoT) characteristics are very high.
Interconnectivity and intelligence
Nearly 20 years ago, with the launch of the BG65CI, Dunkermotoren laid the foundation for today's extensive portfolio of distributed motors. Since then, not only the product portfolio but also the market for distributed solutions has grown significantly. The market demands increasingly complete drives that not only have gearboxes, brakes, and high-resolution encoders , but ideally, integrated intelligence. It should be able to recreate the main processes or perform tasks completely autonomously. Distributed motors, with lower cost and smaller per-processor power, can meet these expectations of greater intelligence. This trend will continue. Distributed motors can not only automate tasks but also collect data from applications, analyze it directly, or forward it to external analytics tools stored at the end customer, OEM, or manufacturer. Currently, data collection has a negative connotation in the media. However, for motor data, end customers, OEMs, and component manufacturers can leverage this because the application becomes transparent due to its use. Therefore, processes can be optimized, new machines can be designed better, and errors can be detected more quickly.
Whether implementing a centralized or distributed system, interconnectivity and intelligence should not be viewed as isolated concepts. More powerful communication systems and smarter components open up more possibilities for the future. This will drive the development of centralized, distributed, and hybrid solutions.
Disclaimer: This article is a reprint. If it involves copyright issues, please contact us promptly for deletion (QQ: 2737591964). We apologize for any inconvenience.
