What is the network theory?

In electrical engineering, network theory is the branch that studies the behaviour of interconnected components (resistors, capacitors, inductors, sources) as a graph of nodes and branches. Core results: Kirchhoff's current and voltage laws, Thévenin and Norton equivalents, mesh and nodal analysis, the maximum power transfer theorem, and Tellegen's theorem. It is the mathematical foundation for every linear circuit-simulator engine (SPICE included).

What is network theory?

Same body of work — network theory is the framework for analysing electrical circuits as graphs. The two foundational laws: KCL (sum of currents at any node = 0) and KVL (sum of voltages around any loop = 0). Combined with the constitutive equations of each element (V = I·R for resistor, I = C·dV/dt for capacitor, V = L·dI/dt for inductor), they let you write a system of equations that fully describes any linear network.

What is control theory?

Control theory is the engineering discipline that designs systems to make a measurable output follow a desired reference, in the presence of disturbances and parameter uncertainty. It combines linear-systems analysis (transfer functions, Laplace transforms), feedback design (PID, lead-lag compensators, state feedback), and stability analysis (Bode, Nyquist, root locus). Applications: process control, robotics, aerospace guidance, power-system stabilisers, and consumer cruise control.

What is the difference between control system theory and control systems theory?

They are the same field — control system theory and control systems theory are interchangeable phrasings. The plural "control systems theory" is more common in academic textbooks; the singular "control system theory" appears more often in engineering practice. Both refer to the mathematical study of feedback, stability, controllability, and observability in dynamic systems.

How do PID controllers work?

A PID controller continuously computes an error e(t) = setpoint − measured value and produces a control signal u(t) = Kp·e(t) + Ki·∫e(t)dt + Kd·de(t)/dt. The proportional term reacts to current error, the integral eliminates steady-state offset, and the derivative anticipates future error. The three gains (Kp, Ki, Kd) are tuned to balance response speed, overshoot, and stability.

What is the difference between open and closed loop control?

Open-loop control (a microwave timer) issues a command and never measures the result — it cannot correct for unexpected loads or disturbances. Closed-loop control (a thermostat) measures the actual output and feeds it back to compute the error; the controller then adjusts the input to drive the error toward zero. Closed-loop is more accurate and disturbance-rejecting but requires sensor hardware and adds the risk of feedback-induced instability.

What is electric network theory used for?

Electric network theory (also written as electrical network theory or network theory in electrical contexts) is used for: circuit analysis (compute currents and voltages from a schematic), filter design (Bode plots and pole-zero placement), power-system load flow and short-circuit studies, signal-integrity simulation in PCB and IC design, and as the mathematical engine inside SPICE, MATLAB Simulink, ETAP, and every other circuit-simulation tool. Without network theory, modern electrical engineering would not exist.

Reference · Concepts · IEEE CSS · IFAC · Cybernetics

Control Theory — Feedback, Stability & the Network Theory Foundation

Control theory is the engineering discipline of designing feedback systems that drive a measurable output to follow a desired reference. This page covers the core concepts (open vs closed loop, transfer functions, PID, Bode/Nyquist stability), the closely-related electrical network theory that supplies the underlying linear models, and worked examples including a DC-motor speed loop and the standard Ziegler-Nichols tuning procedure. Reviewed by a licensed PE.

Tools and software for control system design

Industry-standard tools for control system theory and control systems theory work: MATLAB / Simulink with the Control System Toolbox (industry default for linear control design), GNU Octave (open-source equivalent), Python with the python-control library, Scilab/Xcos, and LabVIEW Control Design. For real-time deployment: Speedgoat or dSPACE for hardware-in-the-loop, Allen-Bradley / Siemens / Schneider PLCs for industrial control. We don\'t host a dedicated control-design calculator yet — for now the related calculators below cover the underlying electrical network theory and impedance work.

Core formulas of control theory

Eq. 01 — Closed-loop transfer function (negative feedback) SI

T(s) = \frac{G(s) \cdot C(s)}{1 + G(s) \cdot C(s) \cdot H(s)}

·: G(s) = plant transfer function (Laplace domain)
·: C(s) = controller transfer function (P, PI, PID, lead-lag, state feedback)
·: H(s) = sensor / feedback transfer function (1 for unity feedback)
·: T(s) = closed-loop transfer function from reference to output

Eq. 02 — Standard PID controller SI

u(t) = K_p \cdot e(t) + K_i \int_{0}^{t} e(\tau)\,d\tau + K_d \cdot \frac{de(t)}{dt}

·: e(t) = setpoint − measured output (instantaneous error)
·: Kp, Ki, Kd = proportional, integral, derivative gains
·: Used in 80 % of industrial control loops; tuned via Ziegler-Nichols, Cohen-Coon, or λ-tuning

Eq. 03 — Stability margins from the Bode plot SI

GM = -|G(j\omega_{180})|_{dB}; \quad PM = 180^{\circ} + \angle G(j\omega_{0})

·: GM = gain margin: gain headroom at the −180° crossover frequency
·: PM = phase margin: phase headroom at the 0 dB gain crossover
·: Stable closed loop typically needs GM ≥ 6 dB and PM ≥ 45°

Standards and authoritative references

Standard / body	Scope
IEEE Control Systems Society	The IEEE society for control theory and applications; publishes IEEE Transactions on Automatic Control and Control Systems Magazine.
IFAC (International Federation of Automatic Control)	Worldwide federation of national control engineering societies; organises the IFAC World Congress.
IEC 61131	PLC programming languages — the practical realisation of control logic on industrial hardware.
ISA-5.1 / ISA-5.2	Instrumentation symbol standards — the P&ID notation used throughout process control.
NIST SP 800-82	Industrial control systems security — guidelines for cyber-physical systems and SCADA networks.
ANSI/ISA-95	Enterprise-control system integration — connecting control systems to MES and ERP layers.

Reference: typical PID tuning starting points

Loop type	Kp (typ.)	Ti (sec)	Td (sec)	Notes
Flow control	0.5–1.0	5–10	0	Fast process; PI is usually enough, no derivative.
Liquid level	2–10	5–10	0	Integrating process; high Kp; no derivative.
Temperature	2–10	60–600	10–60	Slow process with significant lag; derivative helps.
Pressure (gas)	5–20	2–10	0	Fast; PI; derivative noisy.
Motor speed (DC)	0.1–1.0	0.05–0.5	0–0.05	Underdamped electromechanical system.
pH	0.5–2	300–3 000	0	Highly non-linear; gain-scheduled PI typical.

Model the plant — write the dynamic equations. Apply Newton's laws (mechanical), Kirchhoff's laws (electrical), or mass balance (chemical) to derive the system's differential equations. Linearise around the operating point; take the Laplace transform to get the transfer function G(s) = Y(s) / U(s). For a DC motor: G(s) = K / [s · (J·s + b)] where K is torque constant, J is rotor inertia, b is damping.
Pick a control architecture — open or closed loop. Open-loop control sends a command without measuring the output (toaster timer, microwave). Closed-loop (feedback) compares the measured output to the setpoint and corrects the error (cruise control, thermostat, AVR). Closed-loop is more robust to disturbances and parameter drift but adds the risk of instability if the loop gain is wrong.
Choose the controller type — P, PI, PID, or model-based. Proportional only (P) gives steady-state error. Proportional-integral (PI) eliminates steady-state error at the cost of slower response. Proportional-integral-derivative (PID) adds anticipation. For high-performance applications use model-predictive (MPC), state-feedback (LQR), or H∞. PID handles 80 % of industrial loops.
Tune the controller — Ziegler-Nichols or analytical. Increase proportional gain Kp until sustained oscillation; record period Tu and ultimate gain Ku. Ziegler-Nichols PID: Kp = 0.6 Ku, Ti = 0.5 Tu, Td = 0.125 Tu. Or use the Cohen-Coon, IMC, or λ-tuning methods. Always verify with a step test before going live.
Verify stability — Bode and Nyquist plots. Plot the open-loop frequency response. Closed-loop is stable if gain margin ≥ 6 dB and phase margin ≥ 45° at the gain crossover frequency. Nyquist confirms encirclements of the −1 point. Failing margins → reduce gain, add a lead compensator, or move poles via state feedback.

Worked example — DC motor speed loop with PI control

Design a PI speed controller for a 24 V brushed DC motor: torque constant Kt = 0.05 N·m/A, armature resistance Ra = 1 Ω, rotor inertia J = 5 × 10⁻⁴ kg·m², viscous damping b = 1 × 10⁻⁴ N·m·s/rad. Target: zero steady-state error to a step setpoint and settling time ≤ 200 ms.

Step 1 — model the plant. Ignoring electrical dynamics (much faster than mechanical), the speed response to armature voltage is G(s) = (Kt / Ra) / (J·s + b + Kt² / Ra) ≈ 0.05 / (0.0005·s + 0.0026).

Step 2 — pick a PI controller: C(s) = Kp + Ki/s = (Kp·s + Ki) / s.

Step 3 — tune via pole placement. Place the closed-loop pole at s = −20 (50 ms time constant → ≈ 200 ms settling). Compute Kp = 0.96, Ki = 1.04 (via the closed-loop characteristic equation).

Step 4 — verify stability margins. Open-loop Bode at the chosen gain shows PM ≈ 60° and GM > 12 dB — well within the conventional safety margins.

Step 5 — simulate and tune in hardware. The first hardware test will show 10–20 % deviation from the model (parasitic friction, supply impedance). Re-tune empirically using the Ziegler-Nichols quarter-amplitude-decay rule.

Control theory vs network theory — when each one applies

Aspect	Control theory	Network theory
Domain	Time- and frequency-domain dynamics of feedback systems	Static and dynamic analysis of interconnected electrical components
Core variables	Setpoint, measured output, error, control signal	Node voltages, branch currents, complex impedance
Foundational laws	Linear-systems theory (Laplace), feedback equations, stability theorems	Kirchhoff\'s current and voltage laws, Ohm\'s law, Thévenin / Norton
Typical question	"Will my closed-loop system be stable, fast enough, and accurate?"	"What\'s the voltage at this node, current in this branch?"
Tools	MATLAB / Simulink, Bode / Nyquist plots, root locus	SPICE, LTspice, mesh / nodal analysis by hand
Where they meet	Control theory uses network theory to derive the transfer function G(s) of any electrical / electromechanical plant. Every motor controller is a control system whose plant is solved using network theory.

Variants and related disciplines

Network theory — electrical-engineering side

Network theory in electrical engineering is the body of mathematical results for analysing circuits as graphs of nodes and branches. The most-cited results: Kirchhoff\'s current and voltage laws, Ohm\'s law, Thévenin and Norton equivalents, the superposition principle, the maximum power transfer theorem, mesh and nodal analysis, and Tellegen\'s theorem. Modern circuit simulators (SPICE, LTspice, ETAP, PSCAD) implement these algorithms numerically — but the underlying mathematics is unchanged since Kirchhoff\'s 1845 papers.

Cybernetics — the mid-20th-century roots

Norbert Wiener\'s 1948 book Cybernetics formalised the study of feedback systems across engineering, biology, and society. The word "cybernetics" comes from the Greek kybernētēs (steersman). Modern control theory grew directly from this work, taking the mathematical formalism (transfer functions, stability theorems) and applying it to designed systems. Cybernetics today survives as a research area in adaptive systems, robotics, and AI safety.

Modern control: state-space, MPC, LQR, H∞

The classical control theory above (Bode, Nyquist, root locus) handles single-input single-output (SISO) systems. Modern control extends it to multi-input multi-output (MIMO) state-space models, with optimal designs (LQR — linear-quadratic regulator), constrained optimisation (MPC — model-predictive control), robust design (H∞ — H-infinity synthesis), and adaptive / reinforcement-learning approaches. Used in aerospace guidance, power-grid stabilisers, and quadrotor flight controllers.

Frequently asked questions

What is the network theory?: In electrical engineering, network theory is the branch that studies the behaviour of interconnected components (resistors, capacitors, inductors, sources) as a graph of nodes and branches. Core results: Kirchhoff's current and voltage laws, Thévenin and Norton equivalents, mesh and nodal analysis, the maximum power transfer theorem, and Tellegen's theorem. It is the mathematical foundation for every linear circuit-simulator engine (SPICE included).
What is network theory?: Same body of work — network theory is the framework for analysing electrical circuits as graphs. The two foundational laws: KCL (sum of currents at any node = 0) and KVL (sum of voltages around any loop = 0). Combined with the constitutive equations of each element (V = I·R for resistor, I = C·dV/dt for capacitor, V = L·dI/dt for inductor), they let you write a system of equations that fully describes any linear network.
What is control theory?: Control theory is the engineering discipline that designs systems to make a measurable output follow a desired reference, in the presence of disturbances and parameter uncertainty. It combines linear-systems analysis (transfer functions, Laplace transforms), feedback design (PID, lead-lag compensators, state feedback), and stability analysis (Bode, Nyquist, root locus). Applications: process control, robotics, aerospace guidance, power-system stabilisers, and consumer cruise control.
What is the difference between control system theory and control systems theory?: They are the same field — control system theory and control systems theory are interchangeable phrasings. The plural "control systems theory" is more common in academic textbooks; the singular "control system theory" appears more often in engineering practice. Both refer to the mathematical study of feedback, stability, controllability, and observability in dynamic systems.
How do PID controllers work?: A PID controller continuously computes an error e(t) = setpoint − measured value and produces a control signal u(t) = Kp·e(t) + Ki·∫e(t)dt + Kd·de(t)/dt. The proportional term reacts to current error, the integral eliminates steady-state offset, and the derivative anticipates future error. The three gains (Kp, Ki, Kd) are tuned to balance response speed, overshoot, and stability.
What is the difference between open and closed loop control?: Open-loop control (a microwave timer) issues a command and never measures the result — it cannot correct for unexpected loads or disturbances. Closed-loop control (a thermostat) measures the actual output and feeds it back to compute the error; the controller then adjusts the input to drive the error toward zero. Closed-loop is more accurate and disturbance-rejecting but requires sensor hardware and adds the risk of feedback-induced instability.
What is electric network theory used for?: Electric network theory (also written as electrical network theory or network theory in electrical contexts) is used for: circuit analysis (compute currents and voltages from a schematic), filter design (Bode plots and pole-zero placement), power-system load flow and short-circuit studies, signal-integrity simulation in PCB and IC design, and as the mathematical engine inside SPICE, MATLAB Simulink, ETAP, and every other circuit-simulation tool. Without network theory, modern electrical engineering would not exist.

The founding statement of cybernetics

We have decided to call the entire field of control and communication theory, whether in the machine or in the animal, by the name Cybernetics, which we form from the Greek kybernētēs, or steersman. In choosing this term, we wish to recognise that the first significant paper on feedback mechanisms is an article on governors, which was published by Clerk Maxwell in 1868, and that "governor" is derived from the Latin corruption of the same Greek word.

Wiener, N. — Cybernetics: Or Control and Communication in the Animal and the Machine → Introduction (MIT Press, 1948)

Related calculators and references

Sources and further reading

Wiener, N. Cybernetics: Or Control and Communication in the Animal and the Machine. MIT Press, 1948.
Ogata, K. Modern Control Engineering, 5th edition. Pearson, 2010. The standard graduate textbook on classical and modern control theory.
Franklin, G. F.; Powell, J. D.; Emami-Naeini, A. Feedback Control of Dynamic Systems, 8th edition. Pearson, 2018.
Hayt, W. H.; Kemmerly, J. E.; Durbin, S. M. Engineering Circuit Analysis, 9th edition. McGraw-Hill, 2018. The standard undergraduate textbook on electrical network theory.
IEEE Control Systems Society. IEEE Transactions on Automatic Control — the flagship research journal.
IFAC. Annual Reviews in Control — practical industrial control reviews.