About this example
This page is a doctoral research proposal in electrical engineering. The proposed study asks whether combining a static synchronous compensator (STATCOM) with superconducting magnetic energy storage (SMES), and optimising where the device is placed, can improve power quality in a distribution network with a high share of photovoltaic (PV) generation. The proposal shows how to move from a broad engineering problem to measurable objectives, a simulation-based method and a fair comparison with existing approaches.
It is an illustrative teaching example, not a real student's work. Do not submit any part of it as your own work; your university's academic integrity rules apply.
The example: background and problem
1. Background
Distribution networks are receiving more PV generation, which brings power quality problems that the networks were not designed for (Bajaj & Singh, 2020a). PV arrays produce direct current that inverters convert to alternating current; the output varies with irradiance, and the power electronic converters inject harmonic currents. As the share of PV rises, voltage variations, voltage sags and harmonic distortion at the point of common coupling (PCC) become more likely, and the amount of PV a feeder can accept without breaching voltage or thermal limits, its hosting capacity, becomes a constraint (Dubey et al., 2015; Naghdi et al., 2019).
2. Problem statement
Three power quality problems are central. First, passing clouds cause rapid changes in PV output, which produce voltage variations and can contribute to voltage sags (Khatri et al., 2019). Second, inverters and other non-linear loads add harmonic distortion; studies of residential PV expansion and of low-voltage networks with high PV penetration show that harmonic levels can rise as more systems connect (Fekete et al., 2012; Deng et al., 2017). Third, voltage sags matter economically, because sensitive industrial processes can trip during short dips (Chan & Milanović, 2015). Flexible AC transmission system (FACTS) devices such as the STATCOM can mitigate several of these problems (Liao & Milanović, 2017), but their benefit depends strongly on where they are installed, and few studies combine compensation with fast energy storage and systematic placement in PV-rich distribution networks.
The example: aim and objectives
4. Aim and objectives
The aim is to evaluate whether an SMES-supported STATCOM, placed by optimisation, improves voltage profile and harmonic performance in a PV-rich distribution network compared with existing compensation strategies. The objectives are to:
- Obtain measured time series of PV output, voltage and load from an operating PV plant and its feeder, under a data-sharing agreement with the operator.
- Build and validate a simulation model of the feeder, the PV inverters and the loads against the measured data.
- Model an SMES-supported STATCOM and its control, and use ant colony optimisation to choose its location and rating.
- Compare voltage deviation, sag frequency and total harmonic distortion (THD) for the proposed scheme against no compensation, a conventionally placed STATCOM and other optimisation methods.
5. Methodology
The study is simulation-based. A maximum power point tracking (MPPT) controller will be modelled at each PV inverter; its role is to extract the maximum available power from the array, not to smooth variability, so smoothing and voltage support are assigned to the SMES and STATCOM. The STATCOM supplies or absorbs reactive power to regulate voltage, while the SMES supplies short bursts of active power during rapid changes in PV output. Ant colony optimisation will search candidate buses and ratings to minimise a weighted objective of voltage deviation, THD and device cost, subject to network constraints.
| Metric | Definition | Benchmark |
|---|---|---|
| Voltage deviation | Deviation of bus voltages from nominal over the simulated period | Limits in the applicable grid code |
| Voltage sags | Number and depth of sag events at sensitive buses | No compensation; conventional STATCOM |
| Total harmonic distortion | THD of voltage and current at the PCC | Limits in the applicable harmonic standard |
| Cost | Rating-based capital cost of devices | Alternative designs |
Because new metaheuristics are often presented without fair comparison, ant colony optimisation will be compared with at least two established algorithms using the same objective, constraints and number of function evaluations, and with exhaustive search on a reduced network where that is feasible; results will be reported over repeated runs (Kudela, 2022; Camacho Villalón et al., 2019).
The example: work plan and expected contribution
6. Work plan
| Period | Work package | Output |
|---|---|---|
| Months 1 to 6 | Literature review; data-sharing agreement; data cleaning | Review chapter; dataset |
| Months 7 to 15 | Feeder and inverter model; validation against measurements | Validated model |
| Months 16 to 24 | SMES-STATCOM model; optimisation and benchmarking | Optimised designs; comparison |
| Months 25 to 30 | Sensitivity analysis; journal paper | Submitted manuscript |
| Months 31 to 36 | Thesis writing | Thesis |
Data, ethics and risks
The study involves no human participants, but operator data are commercially sensitive. Access, storage, anonymisation of plant identity and publication rights will be set out in a written data-sharing agreement before work starts. The main risk is that measured data are delayed or incomplete; as a fallback, the model will be built first on a published benchmark distribution feeder with synthetic PV profiles, so that the optimisation work can proceed and the measured data can be added for validation later.
7. Expected contribution
The study would contribute a validated comparison of an SMES-supported STATCOM with established compensation strategies in a PV-rich feeder, and evidence on whether optimised placement improves power quality enough to justify its cost. Results will be presented as conditional on the network studied.
How to adapt this to your own proposal
- Open with the technical problem and support each part of it with a source that studied it.
- Write related work that says what each source did, and cite every reference in the text.
- Check every DOI and drop references that do not support your argument.
- Turn the aim into objectives that each produce a checkable output.
- Explain what each device, model or algorithm does, using correct terminology.
- Name your evaluation metrics and benchmarks, and plan a fair comparison.
- Add a realistic work plan and state the contribution modestly.
Common mistakes in engineering proposals
- Reference lists with sources never cited, unrelated sources or invalid DOIs.
- Garbled or machine-paraphrased sentences that misreport sources.
- Technical errors in basic concepts (here, the role of MPPT and the name of the PCC).
- Unsourced numbers such as thresholds or frequency bands.
- Calling an approach "novel" without saying what is new.
- Comparing a new algorithm only with no-compensation baselines.
- Objectives that cannot be measured.
| What examiners look for | Where this example does it | Check in your own draft |
|---|---|---|
| A focused technical problem | Sections 1 and 2 | Is every background claim sourced and relevant? |
| Accurate related work | Section 3 | Does each citation report what the source did? |
| Measurable objectives | Section 4 | Does each objective produce a checkable output? |
| Sound method | Section 5 and Table 1 | Are devices, models and metrics described correctly? |
| Fair benchmarking | Comparison with established algorithms | Would a simpler method do as well? |
| Feasible plan | Table 2 | Is data access secured and the timeline realistic? |
Key takeaways
- High PV penetration brings voltage, sag and harmonic problems at the PCC (Bajaj & Singh, 2020a).
- Different FACTS devices mitigate different problems; placement matters (Liao & Milanović, 2017).
- MPPT maximises extracted power; it does not smooth variability.
- Benchmark a metaheuristic fairly against established methods (Kudela, 2022).
- Check every reference and DOI, and cite each source for what it actually did.
