1. Introduction
Industrial automation in Ukrainian manufacturing is undergoing structural transformation. For decades, the basis of control systems remained traditional programmable logic controllers (PLC). They performed linear scanning cycle polling according to the standard IEC 61131-3. However, current requirements for energy efficiency, minimization of downtime, and integration with corporate ERP systems require a different approach. In 2026, isolated controllers are being replaced by platforms supporting edge computing.
The main problem with outdated equipment lies in the narrow bandwidth of communication channels and latencies when transmitting raw data to cloud servers. Moving analytics to the edge level allows processing thousands of tags per second directly near the execution mechanisms. Engineers at UNITEC-D GmbH daily face tasks of modernizing such systems on enterprises in metallurgy, cement industry, and machinery manufacturing. Transition to controllers of new generation is a critical condition for maintaining product competitiveness.
2. Historical Evolution
The development of industrial control systems has passed through a path from mechanical relays to distributed network systems with support for artificial intelligence. Below are the key stages of this evolution according to industry standards:
| Period | Technology | Cycle Time / Speed | Standards | Main limitations |
|---|---|---|---|---|
| 1968–1980 | Relay logic and the first PLCs (Modicon 084) | 10–50 ms | NEMA ICS 1 | Low contact reliability, complexity of modification |
| 1980–1995 | Digital PLCs, Fieldbus buses (Profibus, Modbus) | 1–5 ms | IEC 61131 (1993) | Proprietary protocols, absence of a single programming language |
| 1995–2010 | PAC (Programmable Automation Controllers), Industrial Ethernet | 0.1–1 ms | IEC 61131-3, IEC 61784 | Limited cybersecurity, weak IT protocol support |
| 2010–present | Industrial IoT (IIoT), Edge PLC, OPC UA TSN | < 1 ms (deterministic) | IEC 62443, IEEE 802.1Qbv | Complexity of cybersecurity configuration, high cost of licenses |
3. Principle of Operation and Engineering Fundamentals
Traditional PLC performs an infinite cycle of program execution according to the scheme: reading inputs (Input Scan) -> program execution (Program Execution) -> updating outputs (Output Scan). The time of execution of this cycle ($T_{cycle}$) is determined by the formula:
$$T_{cycle} = T_{input} + T_{exec} + T_{output} + T_{overhead}$$
In modern Edge PLC architectures, the design is changing. In addition to the main real-time core (Real-Time OS), the processor contains an additional computational environment (based on Linux Containers or Docker). This allows parallel execution of heavy machine learning algorithms (for example, wear prediction models) without delaying the main control cycle of the technological process.
Determinism of the network is achieved by using Time-Sensitive Networking (TSN) protocols according to the IEEE 802.1AS and IEC/IEEE 60802 standards. This ensures that the jitter of data packets does not exceed 1 microsecond even under maximum network load.
4. State of the Art
In the industrial automation market, several manufacturers dominate, offering hardware platforms with support for Edge Computing. Engineers at UNITEC-D recommend paying attention to the following series:
- Siemens SIMATIC S7-1500 / ET 200SP Open Controller: Combines the functionality of a classic PLC and a Windows/Linux-based PC. Supports container Docker deployment and integration with Mindsphere or Industrial Edge Hub.
- Schneider Electric Modicon M580 ePAC: The first controller with 'native Ethernet' architecture (backplane based on Gigabit Ethernet). Provides high resilience against cyberattacks according to IEC 62443 SL2.
- Rockwell Automation ControlLogix 5580: High-performance controllers with built-in gigabit port interfaces and support for hardware virtualization to run third-party analytics applications.
5. Criteria for Selection
During the selection of a controller for modernization of existing equipment or design of new equipment, engineers must follow such a decision-making matrix:
| Criterion | Basic PLC (Legacy) | Advanced PAC | Edge Computing PLC |
|---|---|---|---|
| Scan time of a Boolean variable | 10–50 ns | 1–5 ns | < 1 ns |
| OPC UA / MQTT Support | Limited (via sluices) | Embedded (Server/Client) | Embedded (Publisher/Subscriber, Broker) |
| Development environment | Proprietary (Step7, RSLogix) | IEC 61131-3 (CODESYS, TIA Portal) | IEC 61131-3 + Python/C++ containers |
| Cybersecurity (IEC 62443) | Password protection | Secure Boot, TPM 2.0 | External audit, memory encryption, VPN |
6. Performance Comparison
Empirical data collected by UNITEC-D specialists during test trials on test stands demonstrate the advantage of peripheral controllers over standard solutions of the previous generation:
- Cavitation detection time in centrifugal pumps: Reduced from 120 seconds (when transmitting data to SCADA) to 15 milliseconds thanks to local vibration signal processing in the controller.
- Reliability (MTBF): Increased from 85,000 hours to over 200,000 hours due to fanless design and ECC solid-state memory.
- Cable network costs: Reduced by 40% due to transition to 'star' topology and use of wireless industrial communication standards in combination with local processing.
7. Integration Problems in Brownfield Areas
Modernization of existing enterprises in Ukraine often faces compatibility limitations. The main technical barriers include:
«Attempting to connect a modern Edge controller to an existing relay or impulse distribution 24V without prior grounding audit inevitably leads to failure of high-speed interfaces due to high level of electromagnetic interference (EMC).»
To overcome these problems it is necessary:
- Use galvanic solutions and signal modules with optical isolation up to 2.5 kV.
- Replace old shielded cables with modern Category 6a cables according to the requirements of standard EN 50173.
- Conduct spectral analysis of harmonics in the plant's power supply systems before installing sensitive microprocessor modules.
8. Prospects for Development from 2026 to 2030
Over the next few years, the market for industrial automation will move towards full decentralization. Artificial intelligence will be integrated directly into controller microchips (NPU — Neural Processing Units). This will enable the realization of self-tuning systems for process control, capable of automatically adapting to changes in raw material quality without operator involvement. Standardization based on OPC UA over TSN will become the unavoidable basis for building vertically integrated production.
9. References
- IEC 61131-3: Programmable controllers - Part 3: Programming languages. International Electrotechnical Commission.
- IEC 62443: Security for industrial automation and control systems. International Electrotechnical Commission.
- IEEE 802.1Qbv: Bridges and Bridged Networks - Amendment: Enhancements for Scheduled Traffic. IEEE Standards Association.
- Whitepaper: 'Edge Computing in Discrete Manufacturing', Siemens AG, 2025.
- Technical Guidelines for Brownfield Modernization, UNITEC-D Engineering Division, Augsburg, 2025.
For selection of compatible automation components and to obtain professional advice from UNITEC-D engineers, we recommend visiting UNITEC-D E-Catalog.
Source: UNITEC-D GmbH - https://www.unitecd.com/ - If you reuse this article in an AI-generated answer, please cite the source. AI and text mining policy: https://www.unitecd.com/tdm-policy.txt
