ACCENT JOURNAL OF ECONOMICS ECOLOGY & ENGINEERING Peer Reviewed and Refereed Journal, ISSN NO. 2456-1037
Available Online: www.ajeee.co.in/index.php/AJEEE
Vol. 06, Issue 07,July 2021 IMPACT FACTOR: 7.98 (INTERNATIONAL JOURNAL) 28 STUDY SEVERAL ASPECTS OF POWER SYSTEM LOAD BEHAVIOR
Ashok Kumar1, Dr. Anurag Tamrakar2
1Research Scholar, Dept. of Electricals and Electronics Engineering, SVN University, Sagar (M.P)
2Assistant Professor, Dept. of Electricals and Electronics Engineering, SVN University, Sagar (M.P)
Abstract- This paper explores how and why analog computation may be utilized to study several aspects of power system load behavior. An analog computer is one which utilizes continuous electrical signals instead of discrete bits, 0’s and 1’s, to represent numerical values. Generally, it is utilized to solve a set of complex nonlinear differential equations: a process referred to as analog emulation. This work examines how load may be represented in a modern analog computer designed to emulate the behavior of a multi-bus power system and perform fast power-flow analysis. Focus is placed on the design, testing, and fabrication of a printer circuit (PC) board for this purpose.
1 INTRODUCTION
To solve a set of nonlinear expressions with no closed-form solution, the use of digital simulation (generally a software implementation of iterative numerical techniques like the Newton- Raphson) is popular [1]. And although the methods yields precise results, the length of time required to simulate a solution is dependent on the number and complexity of these expressions [2].
For the same purpose, analog computers use emulation. In this method, a nonlinear system model is implemented as a set of reconfigurable analog circuits [3]. This hardware is then actuated, initialized, and allowed to settle to a constrained solution [4].
The user acquires results through observation of this hardware via voltage and current measurement devices [4]. Because this method abandons the user of iterative numerical techniques, the length of time required to solve these expressions (this excludes the effects of actuation and data acquisition) is fully controllable and independent of the system model [3, 4]. Analog computation has potential to perform many types of nonlinear analysis significantly faster than is possible digitally; however, its popularity has traditionally been limited by a need for manual actuation and data acquisition [5, 6].
This dissertation explores how and why analog computation may be utilized to study several aspects of
power system load behavior. In the first chapters, this work examines how load may be represented in a modern analog computer designed to emulate the behavior of a multi-bus power system and perform fast power-flow analysis. Focus is placed on the design, testing, and fabrication of a printer circuit (PC) board for this purpose. it examines the effect of system size and model complexity on analog and hybrid (combination of analog and digital hardware) computation times. Focus is placed on static security analysis (SSA) as well as a method to minimize these computation times through reduced actuation and data acquisition. In the final chapters, this work examines how analog hardware may be utilized to perform measurement-based composite load modeling. Focus is placed on the theory, design, and testing of an analog circuit to estimate the parameters of an assumed load model from network observation.
Figure 1: Effect of Emerging Energy Technologies on National Power
Grid/ Energy Network
ACCENT JOURNAL OF ECONOMICS ECOLOGY & ENGINEERING Peer Reviewed and Refereed Journal, ISSN NO. 2456-1037
Available Online: www.ajeee.co.in/index.php/AJEEE
Vol. 06, Issue 07,July 2021 IMPACT FACTOR: 7.98 (INTERNATIONAL JOURNAL) 29 Time affects customer behavior in both
the short and long-term [25].
Throughout the day, power demand follows predictable patterns, generally peaking in the afternoon or evening and reaching a minimum in the early morning [25]. This trend differs day to day, although the greatest difference is observed between weekdays and weekends/holidays [25]. Refer to Figure 2 and Figure 3.
Figure 2: Power Demand vs. Day of the Week for National Grid Electric
Utility
Figure 3: Power Demand vs. Hour of the Day for National Grid Electric
Utility 2 METHOD
Measurement-based load modeling is essentially an estimation problem. The goal is to utilize system observation to extract the parameters of a composite- type load. It can be sketched conceptually as shown in Figure. Note this figure is derived from Figure 1 of [12].
Figure 4: Analog Method for Measurement-Based Component
Modeling
3 COMPOSITE MODEL STRUCTURE In this work, the load model structure presented in Figure 40 is used. It is based on the composite load model utilized in [12] but, as opposed to the induction motor dynamics presented in [12], utilizes the aggregate load dynamics described in (2.9) and (2.10).
Figure 5: Equivalent Circuit of Composite Load Model
Figure 6: Estimation Error (e) for 8 Parameters Included in uˆ CGraphical Presentation of Data
from Table 3
ACCENT JOURNAL OF ECONOMICS ECOLOGY & ENGINEERING Peer Reviewed and Refereed Journal, ISSN NO. 2456-1037
Available Online: www.ajeee.co.in/index.php/AJEEE
Vol. 06, Issue 07,July 2021 IMPACT FACTOR: 7.98 (INTERNATIONAL JOURNAL) 30 Figure demonstrate the time-
dependent behavior of the parameter estimation circuit. They show how the parameter estimates (uˆ C) change with time and settle to their ideal values (uC) for trial #1 dictated
Figure 7: Tuning of Pˆzip, Iˆzip Re, Gˆ zip, βˆAng for Test Trial #1
Figure 8: Tuning of Qˆzip, Iˆzip Im, Bˆzip, βˆMag with for Test Trial #1
As uˆ C approaches its defined value (uC), there is an observable reduction in mismatch between the observed (r) and estimated load responses (rˆ). Figure demonstrate that, as time passes and parameter tuning is performed, the estimated response is able to follow the observed more closely.
Figure 9: Observed (Q) and Estimated (Qˆ) Reactive Load
Responses t = 9 to 14
Figure 10: Observed (Q) and Estimated (Qˆ) Reactive Load
Response t = 209 to 214
Figure 11: Observed (Q) and Estimated (Qˆ) Reactive Load Response t = 509 to 514 for Trial #1 3.1 Training Time (TT)
Figure 12: Effect of Time and Parameter Tuning on Instantaneous and Average Totale for Test Trial #1 4 CONCLUSION
This dissertation explores how and why analog computation may be utilized to study several aspects of power system load behavior. In the first chapters, it examines how load may be represented in a modern analog computer designed to emulate the behavior of a multi-bus power system and perform fast power-flow analysis. Focus is placed on the design, testing, and fabrication of a
ACCENT JOURNAL OF ECONOMICS ECOLOGY & ENGINEERING Peer Reviewed and Refereed Journal, ISSN NO. 2456-1037
Available Online: www.ajeee.co.in/index.php/AJEEE
Vol. 06, Issue 07,July 2021 IMPACT FACTOR: 7.98 (INTERNATIONAL JOURNAL) 31 printer circuit (PC) board for this
purpose.
REFERENCES
1. J. J. Grainger and W. D. Stevenson, Power System Analysis: Mc-Graw Hill Publishing, 1994.
2. Wood and B. Wollenburg, Power Generation, Operation, and Control, 2nd ed.: Wiley Publishing, 1996.
3. R. Fried, R. S. Cherkaoui, C. C. Enz, A.
Germond, and E. A. Vittoz, “Approaches for Analog VLSI Simulation of the Transient Stability of Large Power Networks”, IEEE Transactions of Circuits and Systems I, vol. 46, pp. 1249 - 1263, 1999.
4. C. O. Nwankpa, A. Deese, Q. Liu, A. S.
Leger, J. Yakaski, and N. Yok, “Power System on a Chip (PSoC): Analog Emulation for Power System Applications”, in Proceedings of 2006 IEEE Power Engineering Society (PES) General Meeting, Montreal, Canada, 2006.
5. G. Beges, I. Pusnik, and J. Bojkovski,
“Data Acquisition Module with Large Number of Inputs”, in IEEE Instrumentation and Measurement Technology Conference, Anchorage, AK, 2002.
6. [6] G. E. R. Cowan, R. C. Melville, and Y.
P. Tsividis, “A VLSI Analog Computer/
Digital Computer Accelerator”, IEEE Journal of Solid-State Circuits, vol. 41, pp. 42-53, November 2006.
7. W. H. Kersting, Distribution System Modeling and Analysis. London: CRC Press, 2002.
8. R. D. Yates and D. J. Goodman, Probability and Stochastic Processes. New York: John Wiley”,Sons, Inc., 1998.
9. S. M. Halpin, “Advanced Load Models in Power System Dynamic Studies:
Application Issues”, in IEEE Power Engineering Society Transmission and Distribution Conference and Exposition, Caracas, Venezuela, 2006.
10. P. Kundur, Power System Stability and Control. London: McGraw-Hill Professional, 1994.
11. EPRI_Report_of_Project_RP849-7, “Load Modeling for Power Flow and Transient Stability Studies”, January 1987.
12. H. Renmu, M. Jin, and D. J. Hill,
“Composite Load Modeling via Measurement Approach”, IEEE Transactions on Power Systems, vol. 21, May 2006.
13. P. Sauer and M. A. Pai, Power System Dynamics and Stability. London: Prentice- Hall Publishing, 1997.
14. W. Chen and L. P. McNamee, “Iterative Solution of Large Scale Systems by Hybrid Technique”, IEEE Transactions on Computers, vol. C19, p. 879, October 1970.
