The development procedure of a controller for the active

This section presents the development of a reactive power controller for DER providing local AS by constraining the reactive power flow between the TN and DN.

Publications V and VI present the development of a reactive power controller from a preliminary algorithm (presented in Publications III and IV) to a lightweight IED by utilising the Sundom Smart Grid network model and the measured data. The indications for improving the controller, the test setup, the real-time simulation, and the testing platform are discussed in the publications. The CHIL main development stages were:

1) the control algorithm and distribution network model development by the Simscape Power Systems offline simulation tool in the phasor mode, 2) adaptation of the distribution network model in the real-time

ePHASORSIM solver,

3) adaptation of the control algorithm in the real-time eMEGASIM platform as SIL,

4) testing the SIL controller in the real-time OPAL-RT’s co-simulation platform (ePHASORSIM + eMEGASIM),

5) implementation of the control algorithm to a microcontroller device, and 6) CHIL implementation tests.

Figure 27 presents the used real-time co-simulation system setup.

Figure 27. The outline of the real-time simulation platform. Adapted from (K.

Sirviö, Mekkanen, et al., 2018).

Publication V presents various distribution network development scenarios of the Sundom Smart Grid to resolve the challenges because of the increased amount of reactive power injection from the DN to TN because of the network’s cabling increases in the future. Whole year measurement data was utilised, and real-time simulation models of the network development scenarios were developed in eMEGASIM and ePHASORSIM. The paper suggests that the real-time simulation setup could be developed further. For long-term tests and simulation studies, a suitable coefficient for data reading cycle Td could be defined to accelerate the simulation time reliable enough for one-year study cases, for example. Also, the controller could be improved, making it more intelligent, for instance, predictive.

Offline and SIL test results comparison indicated the communication delay effects to the control. The SIL and CHIL test results comparison indicated the processing time of the hardware affecting the results.

Publication VI presents the performance of a reactive power control scheme on a lightweight IED. The development of the control solution and communication system based on open-source IEC 61850 and implemented on two hardware platforms, FPGA and BBB, for CHIL tests, is presented. The performance of the BBB and FPGA was evaluated through CHIL versus SIL test in terms of communication latency, processing time, and control action. The FPGA performed better than the BBB, and therefore, it could be more suitable, for example, for a microgrid (secondary level) controller.

Publication VII presents an accelerated real-time co-simulation method and testing platform for long-term simulations of power systems. Long-term simulations are needed to study, for instance, the potential weekly, monthly, or yearly usage of DERs providing different ASs. Therefore, real-time simulations or HIL tests should be accelerated for testing new algorithms in long-term case studies. Developing this kind of platform aims to aid the development and testing of the ADNM functions. The accelerated long-term simulations are developed and analysed with the Sundom Smart Grid study case. The reactive power controller and the power system model developed in Publications V and VI were utilised.

Publication VII suggests a procedure with the system presented in Figure 28. The ePHASORSIM solver emulates the power system with its simulation time step ts = 0.01. The reactive power window (RPW) controller presents the developed controller with its functioning time step Tc. Td presents the data reading cycle for the loads and generation. The paper presents how to accelerate the long-term simulations by manipulating the input data reading cycle Td. The behaviour of the reactive power controller in long-term simulations was studied by the offline simulations and the SIL and CHIL real-time tests by the following procedure.

Step 1: The offline and real-time SIL simulations were executed for discovering the relevant time factors (including Td) so that the offline and real-time results are equal.

Step 2: When the accelerated SIL real-time results are closest possible to the offline results with only an effect of the communication time delay (Td

setting found), all relevant time factors for the real-time SIL simulations can be concluded.

Step 3: Discovering and defining the Td value, where SIL and CHIL real-time test results do not differ.

Figure 28. The emulated system and the controller (Publication VII).

The effects of the processed input data in the long-term simulations or tests affecting the results were demonstrated. In conclusion, it was shown that it is possible to find a data-reading cycle, coefficient Td, to accelerate the long-term SIL and CHIL simulations/tests. The accelerated real-time co-simulation platform proved to be effective in one-year power flow control simulations. When Td is equal or greater than a particular value, the simulation results do not differ. This defines the minimum data-reading cycle. The total real-time simulation time length is defined through the data-reading cycle. Further, based on the Td result, a suitable PI controller can be derived for accelerated real-time tests preventing oscillations in the system with closed-loop control.

By considering the status of microgrid standardisation, standardised tests for microgrid control are needed (Joos et al., 2017). Though, the IEEE standard 2030.8 for testing microgrid controllers has been launched (IEEE, 2018b).

Accelerated CHIL test methods are useful in testing the AS provision of microgrids since IEEE standard 2030.8 “does not address issues related to the power exchanges between the microgrid and the distribution network at the POI”. The method developed in Publication VII can be utilised to define the test procedure

for AS control development in microgrids. However, more detailed stability and technical analyses of the proposed approach is needed.