A self-contained CHIL platform for evaluating control algorithms in power applications

Shervin Salehi Rad

doi:10.17918/00011204

The increasing complexity of modern power electronics and renewable energy systems demands advanced testing and validation methodologies to ensure performance, reliability, and safety. Testing approaches typically begin with offline simulations, which provide preliminary insights into system behavior and algorithm functionality. These include model in the loop (MIL), which simulates both the model and control algorithm in an idealized environment; software in the loop (SIL), which verifies algorithm compatibility with programming languages for the target microcontroller; and processor in the loop (PIL), which evaluates the algorithm on the actual MCU without considering real hardware interactions. Although useful, offline simulations have several limitations. They cannot capture real time behavior, they rely on unrealistic assumptions of ideal hardware and software models, and they neglect timing and communication delays. To overcome these challenges, hardware in the loop (HIL) testing was developed. HIL can be applied in two ways. The first is controller HIL (CHIL), which evaluates physical controllers and their algorithms in real time using simulated systems. The second is power HIL (PHIL), which tests hardware level performance, including thermal behavior and high power characteristics, under realistic operating conditions. Existing commercial CHIL platforms, although capable, are often costly, complex, and limited in transparency, which restricts algorithm monitoring and flexibility in selecting simulation tools. To address these limitations, this thesis introduces a new, cost effective, and transparent CHIL platform. The platform operates on a single computer while supporting integration with a physical MCU. It provides a flexible communication interface and real time task scheduling using a real time operating system. It maintains compatibility with multiple simulation environments, allowing designers to test algorithms directly without modification. It also can be used to test a wide range of applications, including high power industrial systems such as furnaces and converters ranging from kW to GW, battery chargers from kilowatt to megawatt scales, renewable energy and microgrid systems such as DC microgrids and wireless power transfer, and advanced control strategies such as nonlinear proportional integral derivative control and phase shift control techniques. The platform is validated through a variety of experiments in both power systems and power electronics to demonstrate its capability in analyzing designed algorithms before deployment. In power systems, component and system level tests show differences of less than 0.4% in the extracted voltage and current behavior of devices under test on the developed CHIL platform compared to commercial platforms, and less than 12.5% compared to real world setups. In complex fault scenarios of solid state circuit breakers in a DC microgrid system, the MCU action times on the developed CHIL platform deviate by less than 7% from results obtained using the commercial OPAL-RT HIL platform. In power electronics applications, evaluation of nonlinear proportional integral derivative controllers for battery charging and analysis of phase shift control for wireless power transfer systems demonstrate more than 90% similarity with experimental results. These studies confirm that the developed CHIL platform provides a low cost, transparent, and flexible solution while significantly reducing development time. The platform enables monitoring of internal MCU variables, allowing engineers to observe and debug control algorithms with full visibility. Its modular design and flexible interface support seamless integration with various simulators, enabling accurate hardware model simulations. Overall, the CHIL platform shows effective across a variety of applications, including high-power industrial systems, battery chargers, and DC microgrids, showing that HIL testing can be achieved efficiently without the complexity or expense of commercial platforms.

A self-contained CHIL platform for evaluating control algorithms in power applications

Files and links (1)

Abstract

Metrics

Details

A self-contained CHIL platform for evaluating control algorithms in power applications

Files and links (1)

Abstract

Metrics

Details

Drexel University Social media