Power Electronic Applications and Control Coursework 1: Average Modelling Design and Switching Model Verification of a Full-Bridge Inverter

Introduction:

In this module, the two coursework’s are aimed at providing your insight towards how professional engineers go about the design of power electronic inverters for varying applications.

Putting you in the place of a power electronics design engineer you are tasked with parametric design, controller design, and verification of the power electronic converter. You will firstly model and verify your inverter and controller design through use of average models. Then you will go on to verify your design on a physical converter through a switching simulation.

As the engineer, it’s up to you to use knowledge learnt so far in this module to aid you in designing and delivering the inverter of the expected performance. It is also equally important that the design is properly documented and clearly explained to your stakeholders (your boss, customers, directors) who may not
necessarily have advanced knowledge of power electronics.

Therefore, your Coursework 1 report must not only document very clearly your design procedure in a logical manner, but design decisions and methodologies, behaviours observed throughout the design process must be properly explained and need to convince the reader that your design is delivering what is expected as per the system specification. You should try and refer to literature where possible to back decisions and observations made during the design process.

Part 1: Average Modelling

This coursework concerns of a H-Bridge inverter, which has the following circuit:

The first step of any power converter design is to perform average modelling. Average modelling is a process by which we exclude the switching dynamics of the converter and instead represent everything in terms of the average values across the switching period. Thus, in the first part of this design you are not expected to do any simulations involving switching components.

For example, since a switching inverter simply varies the average voltage on its outputs dependant on switching conditions of the transistors, well without modelling switching, this can be simply seen as a variable voltage source, who’s voltage is then dependant on control perturbations. This allows us therefore to analyse systems in a much more simplified, numerical way, and verify our designs prior to doing more complex switching analysis, done in Part 2 of this coursework.

In Part 1, your task is to design and model a H-Bridge Inverter as discussed in the coursework summary. You will go about designing the filters, and the controller for this inverter using average modelling and verifying all your
results. The desired specification for this inverter is listed below.

Table 1 – Design Specification for Converter to be designed.

The rubric as to how the assessment will be marked can be found on Moodle, however there are some general deliverables which are to be expected in the report.

Deliverables:

1. Parametric design based on the AC current ripple specification (LR Filter)
2. Passive effects on the circuit variables, such as the inverter output voltage.
3. Average model development and verification of the above designs using PLECS.
4. Development of transfer functions to design current loop controllers using MATLAB tools. The input is the voltage demand to the PWM, and the output is the grid current.
5. Verification of the Controller Design
   a. Perform appropriate steps to ensure the inverter can operate at different power levels.
   b. The inverter is bi-directional, so ensure the converter can absorb as well as transmit power.
6. Design of DC-Link capacitance to achieve desired voltage ripple.
   a. You will not be able to develop a average circuit model to integrate with the inverter as you have done above, and the voltage control part of the inverter will need to be kept in transfer function form in your simulations. Ensure to explain why this is the case.
7. Calculate the transfer function for the plant represented by the converter. The input is the AC power reference, and the output is the square of the DC voltage. Show the voltage control diagram and hence design the voltage controller.
8. Verify your voltage control performance and show voltage transients. Show the overall voltage and current control diagrams and simulations to verify global performance of the converter.

Part 2: Switching Model Verification

Completing the average simulation now we need to implement our design on to an actual switching converter to verify that if we implemented our design into physical hardware, it should work as expected.
The switching simulation will continue in PLECS and many of the performance deliverables will remain the same as that for Part 1, since we are wanting to do a direct comparison. However, you will need to build up the switching circuit in PLECS and think about the PWM implementation, so that the controller demands are properly actuated on the converter.

Deliverables:

1. Build a switching PLECs model of the system, including PWM modulator and the full bridge converter. Firstly, use a DC source power supply. Feed the PWM modulator with the voltage demand you previously evaluated in CW1 and demonstrate that the correct average power is exchanged with the AC grid. If you have designed correctly, the switching model performance should match that from deliverable 3 in Part 1.
   HINT: Make sure the full bridge is built as the combination of two half-bridges with independent PWM modulators, as you will have gone through in your lectures.
2. Show that the average over a switching period of the voltage generated by the full bridge matches the voltage demand provided to the PWM modulators. In addition, evaluate the harmonics of the generated voltage and discuss the results.
3. Verify that your calculation for L is correct, showing the current ripple is within the expected requirements.
4. Implement the controller designed in the continuous time domain into the switching PLECs model.
   Provide an AC current reference that enables the exchange of the desired power, and zero reactive power. Show that the closed loop follows the current reference at steady-state, and comment on any discrepancies you may find. In addition, show that the response of the system is as designed by applying different transients to your system. For example:
   a. Power stepping from 0 -> P/2, and then from P/2 to P.
   b. Performing the likewise analysis absorbing power from the grid.
   HINT: Make sure you implement a suitable block which accepts the desired AC power reference, and from there generates the required AC current reference.
5. Up till now, our supply is still ideal. Keep everything in your switching simulation thus far, however, now replace the DC source with a capacitor with a controllable current source in parallel, which represents the current from the photovoltaic grid / energy storage system. Set the DC current to zero and implement the DC voltage controller in the continuous time domain. Run the simulation and verify that the voltage achieves the desired steady-state value.
6. Run the simulation again as you have done for the above deliverable, by increase the DC current from 0to half rated current, and then from half rated to full rated current. Show the response of the DC voltageand compare with the expected response from the controller design. Note any discrepancies you may see.
7. Whilst operating at full DC current, the system is injecting power to the AC grid. Demonstrate during these conditions, the DC voltage ripple respects the systems specifications you have been designing to.

Tips for the Assignment:

• The format for your assignment is a technical report. Ensure to adhere to this format in your writing and ensure its targeted towards the stakeholders. Thus, use headings and subheadings in a logical manner.
• Ensure to go over the inverters section of the module, presented by Dr. Ahmed
• Figures/ Graphs/ Tables should all be clear without requiring zooming in to see clarity. Must be clearly readable for the reader as if report was printed on paper.
• Ensure to consider any limitations with any design approach or analysis being performed and document it and make these points clear to the reader.