Power System‑II (EX-601) - Unit 3 Short Notes

3.1 Significance and Types of Load Flow Studies

Load Flow Analysis (Power Flow Study) determines voltage magnitudes, angles, and real/reactive power flows in a steady-state operating condition.

Significance:

• Planning: System expansion, new line/transformer addition.

• Operation: Day-ahead scheduling, optimal power flow, contingency analysis (N-1 criterion).

• Analysis: Voltage profile assessment, loss calculation, stability studies (initial operating point).

Types of Studies:

• Planning/Design: Evaluates proposed system configurations.

• Operational: Monitors real-time system state.

• Contingency Analysis: Simulates outages (lines, generators) to assess security.

• Optimal Power Flow (OPF): Minimizes cost/losses subject to constraints.

[!TIP] Exam Focus: Always justify why load flow is needed—it's the foundation for almost all other power system analyses (stability, economic dispatch, state estimation).

3.2 Bus Classification: PQ, PV, and Slack Buses

Buses are classified by known/unknown quantities in the load flow solution.

Why Slack Bus?

• Total generation ≠ total load + losses. Slack bus supplies the difference.

• Provides a reference for all voltage angles.

• Without it, the power balance equation becomes indeterminate.

3.3 Formation of Y_bus (Admittance Matrix)

Y_bus is the nodal admittance matrix, fundamental to all load flow methods.

General Form:

$$ Y_{ij} = \begin{cases} \sum_{k=1}^n y_{ik} & \text{for } i = j \text{ (diagonal)} \\ -y_{ij} & \text{for } i \neq j \text{ (off-diagonal)} \end{cases} $$

where $$\displaystyle y_{ij} $$ is the mutual admittance between bus $i$ and $j$.

Formation Steps:

1. Initialize: Set all $$\displaystyle Y_{ij} = 0 $$.

2. For each line (i-j):

   • Self-Admittance: $$\displaystyle Y_{ii} \leftarrow Y_{ii} + y_{line} + j\frac{y_{shunt}}{2} $$ (add half-line charging to both ends).

   • Mutual Admittance: $$\displaystyle Y_{ij} \leftarrow Y_{ij} - y_{line} $$; $$\displaystyle Y_{ji} \leftarrow Y_{ji} - y_{line} $$.

3. For transformers:

   • Tap-Changing (Phase-Shifting): Incorporate tap ratio $a$ and phase shift $\phi$ into the equivalent π-model. The off-diagonal element becomes $$\displaystyle Y_{ij} = -y_{line}/a^* $$.

   • Phase Shifter: Adds an imaginary component to the off-diagonal admittance.

4. For mutual coupling: Add mutual impedance terms directly to the matrix elements.

[!TIP] Common Pitfall: Forgetting to add half the line charging capacitance to the diagonal elements. Y_bus is symmetric for passive, reciprocal networks but can be asymmetric with phase-shifting transformers.

3.4 Gauss-Seidel (GS) Method

An iterative method solving $$\displaystyle I = YV $$ sequentially for each bus.

Iterative Equation for a PQ Bus:

$$ V_i^{(k+1)} = \frac{1}{Y_{ii}} \left( \frac{S_i^*}{V_i^{(k)*}} - \sum_{j=1, j\neq i}^n Y_{ij} V_j^{(k)} \right) \quad \text{where } S_i = P_i + jQ_i $$

For PV Buses: First compute $$\displaystyle V_i^{(k+1)} $$ as above, then correct its magnitude:

$$ \delta_i = \angle V_i^{(k+1)}; \quad V_i^{(k+1)} = |V_i^{spec}| \angle \delta_i $$

Reactive power $$\displaystyle Q_i $$ is calculated from the latest voltage.

Convergence Criterion:

$$ |V_i^{(k+1)} - V_i^{(k)}| \le \epsilon \quad \forall i $$

Typical $$\displaystyle \epsilon = 0.0001 $$ p.u.

Advantages:

• Simple concept, easy programming.

• Low memory requirement (stores only Y_bus and previous voltages).

Disadvantages:

• Slow convergence (linear), especially for large systems.

• May diverge for ill-conditioned systems.

• Requires acceleration factor (0 < $\alpha$ < 2) to improve speed.

Flow Chart (PQ Buses Only):

1. Initialize all $$\displaystyle V_i^{(0)} $$ (e.g., 1.0∠0° for PQ, specified for PV, slack set).

2. For each PQ bus $i$: compute $$\displaystyle V_i^{(k+1)} $$ using GS equation.

3. Check convergence. If not met, $$\displaystyle k \leftarrow k+1 $$, go to step 2.

4. Compute $$\displaystyle P_i, Q_i $$ from final voltages.

3.5 Newton-Raphson (NR) Method

A powerful, quadratic-convergence method using Taylor series expansion and Jacobian matrix.

Formulation:

We solve the mismatch equations:

$$ \Delta P_i = P_i^{spec} - P_i(V,\delta) \quad \text{and} \quad \Delta Q_i = Q_i^{spec} - Q_i(V,\delta) $$

Linearized around operating point $$\displaystyle (V^{(k)}, \delta^{(k)}) $$:

$$ \begin{bmatrix} \Delta P \\ \Delta Q \end{bmatrix}^{(k)} = -J^{(k)} \begin{bmatrix} \Delta \delta \\ \Delta |V| \end{bmatrix}^{(k)} $$

where Jacobian Matrix $J$ is:

$$ J = \begin{bmatrix} \frac{\partial P}{\partial \delta} & \frac{\partial P}{\partial |V|} \\ \frac{\partial Q}{\partial \delta} & \frac{\partial Q}{\partial |V|} \end{bmatrix} = \begin{bmatrix} J_{1} & J_{2} \\ J_{3} & J_{4} \end{bmatrix} $$

with elements like:

$$ J_{1,ik} = |V_i| \sum_{j=1}^n |V_j| \left( G_{ij} \sin(\delta_i - \delta_j) - B_{ij} \cos(\delta_i - \delta_j) \right) \quad (i \neq k) $$

Iterative Process:

1. Initialize voltages (similar to GS).

2. Calculate $$\displaystyle P_i, Q_i $$ from current voltages.

3. Form mismatch vectors $\Delta P, \Delta Q$.

4. Form Jacobian $$\displaystyle J^{(k)} $$.

5. Solve for corrections: $$\displaystyle \begin{bmatrix} \Delta \delta \\ \Delta |V| \end{bmatrix} = -[J]^{-1} \begin{bmatrix} \Delta P \\ \Delta Q \end{bmatrix} $$.

6. Update: $$\displaystyle \delta_i^{(k+1)} = \delta_i^{(k)} + \Delta \delta_i $$; $$\displaystyle |V_i|^{(k+1)} = |V_i|^{(k)} + \Delta |V_i| $$.

7. Repeat until mismatches are negligible.

Advantages:

• Quadratic convergence (very fast, 3-5 iterations).

• Reliable for well-behaved systems.

• Handles PV buses naturally (set $$\displaystyle |V_i| $$, solve for $$\displaystyle \Delta \delta_i $$, $$\displaystyle \Delta Q_i $$).

Disadvantages:

• Complex programming, high memory (store full Jacobian).

• Each iteration requires forming and factorizing a large matrix.

• Sensitive to poor initial guess.

3.6 Comparison of Load Flow Solution Methods

[!TIP] Exam Key: NR is the benchmark for accuracy. FDLF exploits the decoupling property ($\partial P/\partial |V|$ and $\partial Q/\partial \delta$ are "small") for speed in large systems. GS is mostly obsolete for real systems but good for understanding basics.

3.7 Numerical Problems: Key Points

Y_bus Formation:

• Convert all impedances to p.u. on common base.

• Diagonal: Sum of all admittances connected to the bus (including half-line charging).

• Off-Diagonal: Negative of the mutual admittance between buses.

• Example (JUN 2025): For line 1-2 with $$\displaystyle Z = 0.25 + j0.1 $$ p.u., $$\displaystyle y_{12} = 1/(0.25+j0.1) $$. Then $$\displaystyle Y_{11} \leftarrow Y_{11} + y_{12} $$, $$\displaystyle Y_{22} \leftarrow Y_{22} + y_{12} $$, $$\displaystyle Y_{12} = Y_{21} \leftarrow -y_{12} $$. Neglect shunt capacitance as stated.

Load Flow Solution (GS/NR):

• GS: Watch for acceleration factor $\alpha$. Update: $$\displaystyle V_i^{new} = V_i^{old} + \alpha (V_i^{calc} - V_i^{old}) $$.

• NR: Ensure correct ordering of buses (PQ first, then PV). Slack bus is always first/last, fixed.

• PV Bus Limit Check: After each iteration, if computed $$\displaystyle Q_i $$ exceeds limits, convert to PQ bus for next iterations.

• Convergence: Check both voltage changes and power mismatches ($$\displaystyle |\Delta P|, |\Delta Q| < 10^{-4} $$ p.u.).

[!TIP] Problem-Solving Flow: 1) Draw single-line diagram, label buses. 2) Form Y_bus meticulously. 3) Classify buses (specify slack). 4) Choose method (GS for small, NR for large). 5) Iterate until convergence. 6) Report final $|V|$, $\delta$, $P$, $Q$.