The Strejc method is a classic system‑identification technique that estimates the transfer function of a non‑periodic, black‑box system using only its step response. It is widely employed in industrial and mechanical engineering to determine a system’s order, time constant, and delay from a single experimental measurement. The method was originally devised to provide a quick, reliable way to model complex processes without requiring detailed knowledge of internal dynamics.
1. What is the Strejc method?
The Strejc method belongs to the family of step‑response identification techniques. It relies on the following facts:
- Black‑box assumption – the internal structure of the system is unknown or too complex to model directly.
- Non‑periodic input – a step signal is applied; the system’s output is measured over time.
- Inflection‑point analysis – the inflection point of the step‑response curve is used to extract two characteristic times, denoted \(t_{\mathrm{u}}\) (rise time) and \(t_{\mathrm{g}}\) (settling time).
- Numeric table lookup – a pre‑computed table of \(t_{\mathrm{u}}\) and \(t_{\mathrm{g}}\) values for standard transfer‑function orders is consulted.
- Parameter extraction – the table provides the system order \(n\) that best matches the measured times, and the second column of the table gives the time constant \(t\) for that order.
By following these steps, the Strejc method produces a first‑order or higher‑order transfer function that approximates the measured dynamics. The resulting model can then be used for controller design, performance prediction, or system analysis.
2. Theoretical Foundations
2.1 Transfer Functions and System Order
A transfer function \(G(s)\) describes the input–output relationship of a linear time‑invariant system in the Laplace domain:
\[ G(s) = \frac{Y(s)}{U(s)}. \]
The order \(n\) of the system is the number of poles in the transfer function. In the Strejc method, the order is inferred from the shape of the step response, specifically from the ratio of the two characteristic times.
2.2 Time Constants and Delays
- Time constant (\(t\)) – the time it takes for the system’s response to reach a certain percentage of its final value (commonly 63.2 % for a first‑order system).
- Delay (\(t_{\mathrm{d}}\)) – the time between the application of the step input and the point at which the system’s output begins to change.
The Strejc method estimates the combined effect of these parameters from the step‑response curve.
2.3 Inflection Point Analysis
The inflection point is where the curvature of the response changes sign. In practice, it is the point where the slope of the output reaches its maximum. By measuring the times at which the output crosses specific thresholds relative to the final value, the method obtains \(t_{\mathrm{u}}\) and \(t_{\mathrm{g}}\):
- \(t_{\mathrm{u}}\): time from the step input to the inflection point.
- \(t_{\mathrm{g}}\): time from the step input to when the output reaches a chosen percentage (often 90 % or 95 %) of its final value.
These times capture the system’s dynamic behavior without requiring full knowledge of the internal structure.
3. Step‑Response Analysis
3.1 Applying the Step Signal
To use the Strejc method, a step input is applied to the system. The step should be non‑periodic and of sufficient amplitude to elicit a measurable response while avoiding saturation or nonlinear effects.
3.2 Recording the Response
The system’s output is recorded over a time window that covers the entire transient and steady‑state phases. High‑resolution sampling ensures accurate determination of the inflection point and settling time.
3.3 Extracting \(t_{\mathrm{u}}\) and \(t_{\mathrm{g}}\)
- Identify the inflection point – locate the point on the response curve where the slope is maximal.
- Measure \(t_{\mathrm{u}}\) – the time from the step onset to the inflection point.
- Determine \(t_{\mathrm{g}}\) – the time from the step onset to the point where the output reaches the chosen percentage of its final value (commonly 90 % or 95 %).
These two times are the only experimental data needed to apply the Strejc method.
4. The Numeric Table
The core of the Strejc method is a numeric table that relates measured times to system order and time constant. The table is organized as follows:
| Order \(n\) | \(t_{\mathrm{u}}\) | \(t_{\mathrm{g}}\) | Time Constant \(t\) |
|---|---|---|---|
| 1 | … | … | … |
| 2 | … | … | … |
| 3 | … | … | … |
| … | … | … | … |
- The first column lists possible orders \(n\) (typically from 1 to 5 or 6).
- The second column contains the corresponding time constants \(t\) for each order.
- The third column may contain additional parameters such as delay estimates or scaling factors, depending on the specific table used.
The table is derived from theoretical analysis of standard transfer‑function forms. By comparing the measured \(t_{\mathrm{u}}\) and \(t_{\mathrm{g}}\) to the table entries, the user selects the order that best matches the data.
5. Procedure Summary
- Apply a step input to the system.
- Record the output and identify the inflection point.
- Measure \(t_{\mathrm{u}}\) and \(t_{\mathrm{g}}\) from the recorded data.
- Consult the numeric table and find the row that matches the measured times.
- Read off the system order \(n\) and time constant \(t\) from the table.
- Construct the transfer function using the identified parameters (e.g., \(G(s) = \frac{K}{(t s + 1)^n}\) for a standard n‑th order system).
The resulting model captures the dominant dynamics of the black‑box system and can be used for further analysis or controller design.
6. Applications in Industry and Engineering
The Strejc method’s simplicity makes it attractive across a variety of fields:
- Process control – quick modeling of chemical reactors, distillation columns, or heat exchangers.
- Mechanical systems – estimation of motor dynamics, suspension systems, or robotic joints.
- Electrical circuits – characterization of filters, amplifiers, or power supplies.
- Automotive engineering – modeling of throttle response, brake dynamics, or vehicle suspension.
- HVAC systems – analysis of temperature control loops in heating, ventilation, and air‑conditioning units.
Because the method requires only a single step experiment, it is especially useful when time or resources are limited.
7. Advantages
| Advantage | Explanation |
|---|---|
| Minimal data requirement | Only a step response and two time measurements are needed. |
| Non‑invasive | The system is treated as a black box; no internal instrumentation is required. |
| Speed | The identification process can be completed in minutes. |
| Robustness | Works well for a wide range of linear, time‑invariant systems. |
| Ease of use | The numeric table provides a straightforward lookup mechanism. |
These strengths have contributed to the method’s enduring popularity in industrial settings.
8. Limitations
| Limitation | Impact |
|---|---|
| Assumes linearity | Non‑linear dynamics are not captured. |
| Requires a step input | Periodic or multi‑step inputs cannot be used directly. |
| Limited to non‑periodic systems | Systems with significant oscillatory behavior may produce ambiguous inflection points. |
| Accuracy depends on measurement quality | Noise, sampling rate, or saturation can distort \(t_{\mathrm{u}}\) and \(t_{\mathrm{g}}\). |
| Only provides a coarse model | Fine‑tuned dynamics or higher‑order effects may be missed. |
Despite these constraints, the Strejc method remains a valuable first‑pass identification tool.
9. Comparison with Other Identification Methods
| Method | Data Requirement | Complexity | Typical Use |
|---|---|---|---|
| Strejc | One step response | Low | Quick industrial modeling |
| Frequency‑domain (Bode, Nyquist) | Frequency sweep | Moderate | Detailed frequency analysis |
| Least‑squares time‑domain | Multiple input–output samples | High | Precise modeling |
| Kalman filtering | State measurements | High | Adaptive control |
While more sophisticated methods can yield higher fidelity models, the Strejc method’s minimal data needs make it ideal for rapid prototyping and educational purposes.
10. Practical Considerations
- Signal Integrity – Ensure the step input is clean and the output is free from aliasing.
- Sampling Rate – A higher sampling rate improves the accuracy of \(t_{\mathrm{u}}\) and \(t_{\mathrm{g}}\).
- Noise Reduction – Apply filtering if necessary, but avoid excessive smoothing that may shift the inflection point.
- Multiple Trials – Repeating the experiment can help average out random errors.
- Software Tools – Many data‑analysis packages (MATLAB, Python SciPy, LabVIEW) provide functions for inflection‑point detection and table lookup.
By following these guidelines, practitioners can maximize the reliability of the Strejc identification.
11. Illustrative Case Studies
11.1 Heat‑Exchanger Modeling
A chemical engineer applies a step change to the inlet temperature of a heat‑exchanger and records the outlet temperature. Using the Strejc method, they determine an order‑two model with a time constant of 120 s and a delay of 10 s. The resulting transfer function is then used to design a PI controller that stabilizes the outlet temperature within ±2 °C.
11.2 Motor Speed Control
An automotive technician steps the throttle input on an electric motor and measures the speed response. The Strejc method yields a first‑order model with a time constant of 0.8 s. This model informs the tuning of a PID controller that reduces acceleration lag during vehicle start‑up.
These examples demonstrate how the Strejc method can bridge the gap between raw experimental data and actionable control design.
12. Summary
The Strejc method is a pragmatic, step‑response–based system‑identification technique that estimates a black‑box system’s transfer function using only two time measurements derived from the inflection point of the response curve. By comparing these measurements against a numeric table, engineers can quickly infer the system order and time constant, enabling rapid model construction for control design, performance analysis, or educational demonstrations. Its minimal data requirement, ease of implementation, and broad applicability have cemented its place in industrial and mechanical engineering practice.
FAQ
How many experiments are required to use the Strejc method? A single step‑response experiment is sufficient; the method relies only on two characteristic times extracted from that response.