Commissioning is a critical phase in the lifecycle of an industrial drive system, significantly affecting overall performance and reliability. It involves configuring drives and motors for specific applications—such as mixing, pumping, or operating conveyors—and often requires managing complex interdependencies. Traditionally, commissioning is performed manually by domain experts through multiple steps, requiring them to navigate various resources ranging from drive datasheets to vendor-specific tools for drive parameter adjustment. This manual process is time-consuming, and errors can cause significant delays, sometimes lasting several days. In this paper, we propose DriveAIAgent, a novel system architecture and methodology that integrates generative AI-powered multi-agent systems to automate and enhance the commissioning and troubleshooting of industrial drives. Our agentic system utilizes expert tools to interact with and respond to external environments, incorporating a human-in-the-loop approach. The evaluation shows that our system achieves 100% accuracy in identifying encoder parameters during the initial commissioning phase, as well as very high solution relevance (94.73%) and correctness (86.29%) for troubleshooting across different drive types. Overall, DriveAIAgent can streamline complex interactions and significantly reduce the time required for commissioning and troubleshooting industrial drive systems while maintaining the same level of quality as a human expert.

Automation engineering in process industries involves mapping IO list specifications for sensor and actuator signals to control function blocks from vendor libraries. Due to the non-standardized format and content of IO lists, it is challenging to automate this mapping as it involves interpreting unstructured data (e.g., signal descriptions). Previous approaches proposed mapping of IO list contents to common data schemas, such as ontologies or UML models, but did not provide means to automate the initial data import from heterogeneous input formats. We propose IO-AutoMapper, an LLM-supported method to automatically process spreadsheet IO lists and map their entries to control library function blocks. Experiments with five IO lists with 300 signals each showed more than 98% correct mappings and low LLM processing costs. While LLM hallucinations could not yet be fully eliminated, the method has the potential to reduce human labor for mapping IO list entries by more than 90%.

Testing PLC and DCS control logic in industrial automation is laborious and challenging since appropriate test cases are often complex and difficult to formulate. Researchers have previously proposed several automated test case generation approaches for PLC software applying symbolic execution and search-based techniques. Often requiring formal specifications and performing a mechanical analysis of programs, these approaches may uncover specific programming errors but some-times suffer from state space explosion and cannot process rather informal specifications. We proposed a novel approach for the automatic generation of PLC test cases that queries a Large Language Model (LLM) to synthesize test cases for code provided in a prompt. Experiments with ten open-source function blocks from the OSCAT automation library showed that the approach is fast, easy to use, and can yield test cases with high statement coverage for low-to-medium complex programs. However, we also found that LLM-generated test cases suffer from erroneous assertions in many cases, which still require manual adaption.

Large language models (LLMs) providing generative AI have become popular to support software engineers in creating, summarizing, optimizing, and documenting source code. It is still unknown how LLMs can support control engineers using typical control programming languages in programming tasks. Researchers have explored GitHub CoPilot or DeepMind AlphaCode for source code generation but did not yet tackle control logic programming. A key contribution of this paper is an exploratory study, for which we created 100 LLM prompts in 10 representative categories to analyze control logic generation for of PLCs and DCS from natural language. We tested the prompts by generating answers with ChatGPT using the GPT-4 LLM. It generated syntactically correct IEC 61131-3 Structured Text code in many cases and demonstrated useful reasoning skills that could boost control engineer productivity. Our prompt collection is the basis for a more formal LLM benchmark to test and compare such models for control logic generation.

Modern industrial automation applications require a flexible and modular software architecture that can be easily adapted to meet dynamic market requirements. The IEC 61499-based programming model, due to its distributed and modular nature, could be a potential tool in this regard. In our previous works, we proposed the concept of a PLC-Service bus, a service-oriented architecture-based control application development paradigm for the Programmable Logic Controller (PLC). We have also demonstrated its applicability using the IEC 61499 modelling language. In this paper, we propose an OPC UA-based configuration approach for software components of the PLC-Service bus.

To address current challenges of production systems such as increased number of product variants, batch size one, and shorter product life cycle, they need to be modular, loosely coupled, and more adaptable than ever. A PLC-Service bus architecture has been proposed to improve the modularity and adaptability of a PLC-software. The architecture enables software components of PLC-software to interact via middleware in a loosely coupled manner. Previous work identified a catalog of message integration patterns for PLC-software, and IEC 61499 was proposed to implement the architecture. The current version of the IEC 61499 standard does not fully support all message integration patterns from the catalog.Therefore, this paper first identifies suitable middleware technologies for IEC 61499 and, later, based on the requirements of the PLC-Service bus architecture, evaluates them to propose the suitable middleware technologies.

Modern production systems are constantly getting more complex and moving towards smaller lot sizes and more customizable products. This poses new challenges on the control software, which is growing in size and complexity. To deal with this increasingly large and complex control software, design guidelines are needed. We are working on such design guidelines for IEC 61499, a modeling language for distributed control systems. In our previous work, we proposed a distributed and highly reusable application design based on skills. In this work, we will present concepts for extending this design beyond normal operation handling.

This work analyzes the problems that arise in IEC 61499 development due to the fact that there are no ways of structuring type libraries. As a result of this analysis we derived requirements for integrating modules into the IEC 61499 specification. The modules that are portable among several runtime providers and vendors can be imported into IEC 61499 systems. Every module contains a set of function blocks that can be used inside the application. The proposed IEC 61499 module has its own version and namespace. It is also capable of defining dependencies on each other. Reusability from types can be increased by dividing the type library into several parts with the help of the proposed modules.

The software architecture of PLC-software has a direct influence on its adaptability, modularity, and flexibility. The software component (Program organization unit, POU) is the smallest unit of PLC-software that is responsible for providing certain machine functionality in an overall production system. The way these software components interact with each other has a significant role in the overall PLC-software architecture. To address current market demands, software components need to be atomic, modular, and loosely coupled. PLC-Service bus-based architecture enables such PLC-software. We are using this architecture as a basis. In this paper, we are evaluating available messaging interaction patterns from the enterprise IT domain. Our evaluation is based on requirements defined by the PLC-Service bus architecture. Therewith, we identified a list of interaction patterns that can be used to design interactions between software components in a PLC-Service bus environment. We use a reference production cell as a running example. To evaluate our findings, we first explain what the PLC-Service bus concept would look like for our running example. Later, we used four practical process scenarios from our running example to evaluate the applicability of identified messaging integration patterns for the PLC-Service bus.

IEC 61499 emerged as a language for modeling distributed control systems. An Application is a platform-independent model that is comprised of modular components. High reusability of these components and high reconfigurability of the underlying system configuration can only be achieved with advanced design patterns. We show two example implementations of such designs and extended a pattern to foster reuse. All designs are compared based on a set of evaluation criteria. We show that our proposed design provides improved understandability and adaptability, while maintaining a slim library with highly reusable Function Block types.

How flexible and adaptive a production system can be, unquestionably depends on its software architecture and the level of modularity, flexibility, and adaptability of its software components. Programmable Logic Controllers (PLCs) are responsible for the functionalities of machines in the production system. Therefore PLC-software architecture and its software components play a vital role in a flexible, modular, and adaptive production system. In the current scenario, PLC software components use global variables to interact with each other to deliver certain machine functionality. Having global variables in PLC-software, increase hidden dependencies, reduces flexibility thus results in a less modular and less adaptable production system. In this paper, we are evaluating a service bus concept for modular and adaptable PLC-software that controls modular machines and production cells. As a result, a list of requirements for realizing such a concept is proposed.