Simulink Model Development

After completing the MATLAB analysis, I focused on developing a Simulink model to validate the analytical approach through visual simulation. The objective was to show that the MATLAB-developed multi-turbine grid integration techniques could be effectively applied in a block-diagram simulation environment, offering a supplementary validation technique and laying the groundwork for possible real-time implementation.

Simulink was selected because, in contrast to pure code, it allows for the visual depiction of system design, which facilitates the understanding of signal flow and data connections.

The next natural step after this software-focused research would be to deliver Simulink models to embedded hardware for real-time control applications.

Implementation Approach:

Model Architecture

The Simulink model was designed with a hierarchical structure representing the complete wind farm system:

• Turbine layer (4 constant Blocks): Represents individual turbine power outputs
• Turbine 1: 1.8W (Small turbine, moderate wind)
• Turbine 2: 2.3W (Medium turbine, good wind)
• Turbine 3: 1.5W (Small turbine, lower wind)
• Turbine 4: 2.6W (Large turbine, high wind)

Grid integration layer (Sum Block): Aggregates power from all turbines

• Implements the mathematical summation: Ptotal = P1 + P2 + P3 + P4
• Represents the physical DC bus where all turbines connect

Stability assessment layer (Gain Block): Calculates grid stability metrics.

• A gain factor of 12 scales the total power to create a stability indicator
• Simplified version of the comprehensive MATLAB stability algorithm

Visualisation Layer (Display and scope): Real-time monitoring

• Individual turbine displays show the contribution from each source
• The total power display shows the aggregated grid output
• The stability metric displays system health
• The scope block visualises power trends over time

Construction Process:

    Step 1: Turbine Power Sources

Four constant blocks were used to represent turbine power outputs. These values were used to demonstrate different turbine sizes and wind conditions, similar to scenario 1 from the MATLAB analysis. The variation in values (1.5-2.6W) shows realistic differences in turbine performance due to size and wind exposure.

    Step 2: Signal Multiplexing

A mux (multiplexer) block was added to connect the four turbine signals into a single signal. This is necessary for the visualisation in the scope block. allowing all turbine signals to be displayed simultaneously for comparison.

    Step 3: Grid Integration

The Add block (configured with four inputs: ++++) sums up all turbine signals to calculate the total grid output. This represents the DC bus aggregation point where all the turbines connect. In a real system, this would be the common voltage rail after the DC-DC conversion.

    Step 4: Stability Metric Calculation

A Gain block with a factor of 12 scales the total power to create a stability metric. While it is simpler than the MATLAB algorithm. It demonstrates the concept of deriving system health indicators from power measurements.

    Step 5: Individual Monitoring

Four display blocks were connected to each turbine output to show each turbine's contribution. This allows operators to identify underperforming turbines more easily.

    Step 6: System-level visualisation

Two more display blocks were added to show total grid power and stability metrics. The scope block gives us a time-domain visualisation of all four turbine outputs. This allows us to identify power fluctuations and synchronising issues.

Results:

The Simulink model was successfully executed and produced the following results:

Validation against MATLAB analysis:

The Simulink model serves as an independent validation of the MATLAB analysis.

Even though each specific turbine configuration differs, Simulink uses constant values, whereas MATLAB uses time-varying simulations. However, the fundamental calculation method is the same:

• Power Aggregation: Both environments sum individual turbine outputs
• Stability Assessment: Both derive system health readings from power measurements
• Individual Monitoring: Both track the contribution from each turbine
• Visual Analysis: Both provide a graphical representation of system behaviour

The main difference is representation: MATLAB uses sequential code, whereas Simulink uses concurrent block processing. This indicates that the analysis method is sound regardless of the implementation.

Why Simulink matters for this project

1. Visual System Architecture

The block diagram shows the system architecture, structure and signal flow. Anyone looking at the model can understand the four-turbine configuration. grid integration point, and monitoring infrastructure without reading any of the code. This is useful for:

•  Project documentation and reporting
• Explaining the system to non-programmers, such as clients and shareholders
• Identifying potential issues in the system design before coding

2. Complementary validation method

Having two independent implementations (MATLAB scripts and Simulink models) that produce consistent results strengthens confidence in the analytical approach. If both methods arrive at the same conclusions about grid stability and power aggregation, the methodology is more credible.

3. Foundation for future hardware implementation

Simulink models can be compiled for embedded hardware (microcontrollers and FPGAs) using the code generation tool. The current model could eventually become the actual control system running on physical hardware, making it more than just a simulation – it's a prototype of the real implementation.

4. Real-time simulation capability

Unlike batch-processing MATLAB scripts, Simulink can run in real-time mode, simulating system behaviour as it would occur in reality. This allows us to test control algorithms with realistic time constraints.

Engineering decisions and trade-offs:

Decision 1: Constant vs Time-varying turbine outputs

I used constant blocks instead of implementing the full time-varying wind simulation from MATLAB. This was a deliberate choice to focus on demonstrating system architecture rather than replicating the entire MATLAB analysis. The trade-off is that this model shows steady-state behaviour only and won't respond to wind fluctuations.

Decision 2: Simplified Stability Metric

The gain block stability metric is a lot simpler than the MATLAB algorithm that takes power variation, voltage standard deviation and output consistency. I made this choice because:

• The demonstration model focuses on grid integration instead of a comprehensive stability analysis
• A single gain block is easier to understand than a complex MATLAB function block
• The full stability algorithm is already validated in MATLAB-Simulink, which demonstrates the concept

Decision 3: Individual Turbine Displays

Adding separate displays for each turbine wasn't necessary, as the scope shows all the outputs, but I included them to make the model more operator-friendly. In a real monitoring system, being able to see individual turbine contributions at a glance is valuable for troubleshooting.

Decision 4: Using Mux for scope visualisation

The Mux block bundles all four turbine signals for the scope. This adds one extra block but enables simultaneous visualisation of all turbines on separate sub-plots, making it easier to identify which turbines are underperforming or showing unusual patterns.

How this Simulink model differs from existing wind monitoring solutions:

1. Focus on Multi-Turbine integration, not single-turbine monitoring

Most small-scale wind monitoring systems show data for individual turbines in isolation. This model explicitly demonstrates the grid integration challenge – how multiple turbines with different outputs can combine into a single power system. The sum block represents the DC bus aggregation that existing hobbyist solutions ignore.

2. Visual simulation before hardware implementation

Modern commercial SCADA systems require purchasing expensive hardware up front. This Simulink model allows testing different turbine configurations by simply changing the constant block values without any physical equipment. For example:

• Change turbine values to test different wind scenarios
• Adjust the gain factor to test different stability thresholds
• Add more turbines by duplicating blocks and changing values accordingly.

This simulation-first approach means we can validate the concept before committing to hardware purchases

3. Demonstrable System architecture

The block diagram itself is a valuable deliverable. Unlike code, which requires prior programming knowledge to understand, the Simulink model can be shown to

• Potential clients
• Hardware engineers

4. Bridge between analysis and implementation

Existing solutions are either pure analysis tools (MATLAB scripts, spreadsheets) or pure monitoring hardware (data loggers, SCADA). This Simulink model connects both worlds:

• It's analytical enough to test different scenarios
• It's implementable enough to become real control code
• It validates the MATLAB analysis through an independent method

What this Simulink model enables:

For this project:

• Visual demonstration of the multi-turbine grid integration concept
• Independent validation of MATLAB code
• Professional-looking system diagram for the final report
• Clear illustration of signal flow and system architecture

For future development

• Implement full wind turbine power equations in MATLAB function blocks
• Add DC-DC converter models with efficiency curves
• Implement the complete stability scoring algorithm 
• Add battery storage dynamics and charge controller logic
• Test MPPT (Maximum power point tracking) algorithms

For real-world application

• Compile to embedded code for microcontroller deployment
• Connect to real sensors via hardware-in-the-loop interfaces
• Implement automated control logic (load balancing, fault detection)
• Deploy to the IoT dashboard for remote monitoring.

Technical insights

1. Block diagrams are powerful communication tools. Explaining the four-turbine system with the Simulink model is a lot simpler than walking through MATLAB code. The visual representation immediately conveys the system structure.

2. Simplification doesn't mean loss of value. The simplified stability metric (Gain block) is less sophisticated than the MATLAB algorithm, but it still demonstrates the concept effectively. Sometimes, a simpler model that people can understand is more valuable than a complex one that works slightly better

3. Multiple validation methods strengthen credibility. Having both a MATLAB script and a Simulink model that arrive at a consistent conclusion makes the overall project more credible. If only one implementation existed, there would always be doubt about whether bugs or errors affect the results

Process Insights

1. Building incrementally is essential. Starting with one turbine, adding the Sum block, then gradually enhancing with Mux, Scope and individual displays worked much better than trying to build everything at once.

2. Visual debugging is faster. When the model didn't work initially, seeing the block diagram made it obvious where connections were missing. This would have been harder to debug in code.

Connection to Dr Purav's Original Research questions:

Dr. Purav's initial guidance was to investigate multi-turbine grid integration challenges. The

Simulink model addresses this in three ways:

    1. Power discrepancy visualisation

The individual turbines display (1.8W, 2.3W, 1.5W, 2.6W), making power variation immediately

visible. The 42% variation from smallest to largest turbine demonstrates the challenge of

integrating mismatched sources.

    2. Grid integration demonstration

The Sum block explicitly shows the aggregation point—where four independent power

sources must combine into a single grid output. This is the DC bus that requires careful

voltage regulation in the real system.

    3. System health monitoring

The stability metric display provides a single number indicating system health. While

simplified, it demonstrates how complex multi-turbine behaviour can be reduced to actionable indicators for operators.

Conclusion:

The Simulink model successfully demonstrates multi-turbine grid integration through visual simulation. While simplified compared to the MATLAB analysis, it serves three critical purposes:

1. Independent Validation – Confirms that the MATLAB analytical approach is sound
2. Visual Communication - Provides a clear system architecture diagram for reporting
3. Implementation Foundation - Creates a pathway for future hardware deployment

 

The model produced expected results (8.2W total power, 98.4 stability metric) and successfully visualised how four turbines with varying outputs combine into a single grid system. This complements the MATLAB analysis by showing the same concepts through a different lens—block diagrams instead of code.

Most importantly, the Simulink model bridges the gap between pure analysis (MATLAB

scripts) and physical implementation (hardware systems). It's analytical enough to test

different scenarios but concrete enough to eventually become deployed control code. This

dual nature makes it a valuable addition to a simulation-focused final year project.

This Simulink development phase demonstrates that visual simulation can effectively

complement analytical code-based approaches, providing validation through independent

implementation while creating clearer documentation for technical reporting