A Novel Architecture for Testing and Evaluating Programmable Logic Controller (PLC) Based Control Systems Using Software-in-the-Loop (SIL): A Naval Hull, Mechanical, and Electrical (HM&E) Subsystem Case Study

Despite rigorous and exhaustive testing in a ship representative test site prior to going underway for sea trials, programmable logic controller (PLC) based control systems frequently have undiscovered software bugs—such as incorrect input/output (IO) signal usage, mismatched signal names, erroneous setpoint values, and incorrect logic. These bugs require the vessel to return to port for software corrections and updates before additional testing can commence. The expense and inflexibility of a full ship hardware representative set of material, coupled with the time and monetary cost associated with the sea trial interruptions, is an obstacle to agile software deployment. Motivated by increasing efficiency and decreasing costs, in this paper, we present a novel architecture to verify and validate (V&V) PLC control systems before they are deployed and tested on a vessel. Our agile framework loads a real-world Siemens 1500 Series PLC project file into a virtual simulation environment that runs custom test scenarios and evaluates the control system. The environment consists of virtualized Siemens 1500 Series PLCs—using Siemens PLCSIM Advanced—and loaded plant models to achieve software-in-the-loop (SIL) simulation testing. A python based software suite orchestrates the simulation by interacting with and facilitating the data flow between the virtualized PLCs and plant models, logging data, running the scenario scripts, and generating test reports. These reports detail the controller performance throughout the various scenarios and identify the parts that failed. Utilizing the test reports, control software bugs and incorrect logic is detected and resolved before platform testing. The all-virtual architecture eliminates the need to have a full ship hardware representative set of material. The framework also allows the plant simulation models to be V&V’d separately from the simulation environment in a setting more conducive for the modelers. Furthermore, the software suite is ship agnostic so it can be used to perform tests on a variety of plants and Siemens PLC project files. Another key attribute of the software suite is that it was developed in an object oriented framework so new features can be implemented as needed. In order to demonstrate the architecture, we present numerical validations of our work for a naval HM&E subsystem case study utilizing a nominal Siemens PLC control system and plant models.