An Adaptive Protection Scheme for Optimal Overcurrent Relay Coordination in Interconnected Power Systems

Transcription

1 From the SelectedWorks of Almoataz Youssef Abdelaziz March, 2000 An Adaptive Protection Scheme for Optimal Overcurrent Relay Coordination in Interconnected Power Systems Almoataz Youssef Abdelaziz Available at:

2 An Adaptive Protection Scheme for Optimal Overcurrent Relay Coordination in Interconnected Power Systems A.Y. Abdelaziz H. E. A. Talaat A. I. Nosseir Department of Electrical Power and Machines Faculty of Engineering Ain Shams University, Cairo, Egypt Ammar A. Hajjar Electrical Power Department Faculty of Mech. and Electrical Engineering Tishreen University, Latakia, Syria Abstract This paper presents an adaptive protection scheme for optimal coordination of overcurrent relays (OCR) in interconnected power networks with an improved formulation. The scheme adapts to system changes; new relays settings are implemented as load, generation-level or system-topology changes. The software developed for this application is described. The developed scheme is applied to the IEEE 30-bus test system. Results showed the importance and necessity of this scheme in maintaining the optimal performance of the relays in all conditions. Keywords: overcurrent relay coordination, optimization, adaptive protection, interconnected power networks. 1. Introduction Directional overcurrent relaying, which is simple and economic, is commonly used in power system protection, as a primary protection in distribution and subtransmission systems and as a secondary protection in transmission systems. The main problem that arises with this type of protection, is the difficulty in performing the relays coordination, especially in the multi-loop, multi-source networks [1]. Since the sixties, a great effort has been devoted for solving this problem by computer. The methods, which are used, for performing this task (relay settings) can be classified into three classes: trial and error method [2], topological analysis method [3,4], and optimization method [5-8]. In the optimization method, some researchers used nonlinear programming for determining the optimal settings of the pickup current and a linear programming for optimizing the time dial settings of the relays subject to the coordination constraints, and the limits of the relay settings [5,6]. Other researchers [7] applied the linear programming technique only to minimize operating time while the pickup currents are selected based on experience. The presently followed protection philosophy assumes pre-determinism in its application. All the faults, abnormal operating conditions and system contingencies are predetermined in order to set and coordinate protective relays in an electric power system. The relays respond to these predetermined conditions in a satisfactory manner. But if a condition arises which has not been included in the analysis earlier, the response of the relays would not be satisfactory and the security of the power system, as far as the protection is concerned, is jeopardized. Furthermore, it is not only difficult to identify and analyze all the operating conditions of concerns in advance, it is also impossible to determine the relay settings, which would be optimal for all abnormal and normal operating conditions. The power system protection is improved and system security enhanced by following adaptive protection philosophy. Adaptive protection is relatively a new concept. It is defined as the ability of the protection system to automatically alter its operating parameters in response to changing power system conditions, to provide reliable relaying decisions. In recent years, a number of adaptive relaying concepts have been proposed [9-11]. The Adaptive Protection and Control Working Group of IEEE Power System Relaying Committee recently conducted a survey [12] which reveals that while protection engineers are satisfied with presently used systems, they consider it desirable to take advantage of the improvements that can be achieved using adaptive relaying concepts. The idea of adaptive protection has been applied to the problem of overcurrent relay coordination. The relay coordination in an interconnected power system is a tedious and a time-consuming task. To relieve the protection engineer from this laborious task, it is proposed to set and coordinate relays in an on-line manner. In adaptive coordination, the relays should respond to the changing system conditions and adapt according to the new prevailing conditions. The changing system conditions could be operational or topological. In this paper, the authors suggest an adaptive protection scheme for optimal coordination of overcurrent relays in interconnected power networks 556

3 through applying a linear programming technique. Results showed the importance and necessity of this scheme in maintaining the optimal performance of the relays in all conditions. 2. Optimal coordination of overcurrent relays The general relay coordination problem can be stated as a parametric optimization problem. The objective function of operating times of the primary relays is minimized subject to keeping the operation of the backup relays coordinated. One possible approach to achieve minimum shock to the system due to faults would be to minimize a sum of the operating times of all primary relays hoping that the operating times of individual primary relays would be close to the minimum individual operating times that might be possible. The objective function is taken as the sum of the time dial settings () of all primary relays irrespective of the type and location of the fault. The constraints considered here are based only on the maximum near-end faults. The considerations of the weight factors and far-end faults in the problem formulation haven t any effect on the optimal solution [13]. So, the problem dimensionality is reduced to a quarter in comparison with [8]. Consequently, a faster algorithm, which is more suitable for on line application is obtained. In the application reported in this paper, the overcurrent relay is conformed to the following IEC characteristic [14]: k1. T = k2 (1) [ M 1] where M is a multiple of the pickup current, i.e., I M =, I is the relay current (overload and/or fault), I pu is the pickup current, k 1 and k 2 are constants which depend on the relay characteristic. The problem formulation can be demonstrated with the help of Fig. (1) and by assuming a network consisting of n relays, the objective function J to be minimized can be expressed as: n J = T i=1 ii (2) where: T ii is the operating time of the primary relay R i for a near-end fault i, R j R i near-end fault T ji T ii Far-end fault The operating time of the backup relay must be greater than the sum of the operating time of its primary relay and the coordination margin. This can be expressed as: T ji T ii + CTI (3) where: T ji is the operating time of the backup relay R j for the same near-end fault at i, and CTI is the coordination time interval. From equation (1) one can see that the relation between the operating time T of the time overcurrent unit, and the multiple pickup current M, is nonlinear. Since the multiple pickup current of the relays can be predetermined, so for a fixed M, equation (1) becomes linear as follows: T = a. (4) where: k1 a = k2 M 1 (5) By substitution in equation (2), the objective function becomes: n J = a i. i (6) i= 1 In equation (6), a i s haven t any effect on the optimal solution and can be assumed ones, they are predetermined from equation (5) and substituted in (3); values of i are determined by minimizing J (the objective function) and satisfying the coordination between the primary and backup relays. Equation (6) is optimized using the Active Set Strategy two-phase method subject to the condition that the operation of the backup relays remains properly coordinated. 3. The adaptive relaying scheme Fig. (2) shows the functional block diagram of the adaptive relaying scheme proposed for overcurrent protection of the power system. Similar configurations of relays and computers are used at other substation. The relays sample line currents via current transformers. Each relay possesses quantized samples and calculates voltage and current phasors. Under normal operating conditions, each relay provides the measurements to the substation control computer at regular intervals. The SCADA system checks the status of local isolators and circuit breakers and provides the information to the substation computers. In addition to communicating with the relays, the station computer pass on the collected information to the central computer at prespecified intervals (e.g. one hour). Fig. (1) An illustrative diagram for basic definitions 557

4 C.B status RTU POWER SYSTEM HIGH VOLTAGE EQUIPMENT DATA I Digital Relay #1 Trip Signal Trip Signal Digital Relay # n SUBSTATION COMPUTER I TO OTHER SUBSTATIONS DATA RTU C.B status Transfer settings to the relays: in adaptive coordination of overcurrent relays, it is assumed that all relays are of digital type and that there is a communication channel between them and the substation control computers (optic-fiber cables) as shown in Fig. (2). Hence, once the relay settings are calculated, these are communicated to the respective relays. Monitoring: the power system is continuously monitored, using SCADA system, for any change operational or topological. If the change detected is an operational change, the computer will restart the coordination process from the load flow program and if the change is due to a topological change, the procedure will be restarted from the topology processor. Breaker status information SUBSTATION DATA OTHER SUBSTATION COMPUTERS Topology processor Load flow CENTRAL CONTROL Fig. (2) - A block diagram of the adaptive scheme for a sample substation The central computer estimates the system state and decides whether or not the relay settings should be changed. If it decides to change the settings, it will calculate the new settings and conveys them to the relays via the substation control computers. The relays implement the new settings and send confirmation messages to the central computer via substation computers. If the central computer decides not to change the settings, the decision will be communicated to the relays for the purpose of sharing information and confirming that communication facilities are working properly. 4. Algorithm for adaptive coordination Fig. (3) illustrates the flowchart of the developed MATLAB algorithm for the adaptive coordination of the overcurrent relays. Each block of the flowchart is explained as follows: Topology processor: the topology processor tracks the network topology over the time. The circuit breaker status information is the main input to this module. The topology processor feeds the network information to the load flow program and the fault analysis program. Optimal coordination procedure: just as described in section 2 using a linear programming technique (Active Set Strategy two-phase method) as follows: In phase I, a feasible solution is obtained and in phase II, the optimal solution is found. Topological Optimal coordination Transfer settings to the relays System Monitoring (SCADA system) Any change Detected? Yes Type of Change? No Operational Fig. (3) - Flowchart of the proposed adaptive optimal coordination algorithm 5. Application of the proposed methodology to the IEEE 30-bus system Considering the IEEE 30-bus test system shown in Appendix [A], it consists of two subsystems (subtransmission system 132 KV and distribution system 33 KV). The conjunction substations have 132/33 KV step-down transformers. Each line is equipped with a circuit breaker at each end. Each breaker is equipped with a directional overcurrent relay (with inverse characteristics). Directional overcurrent relays are used to protect the 33/132 KV transformers and load circuit emanating from the substations. The system data (loads and line parameters) are given in [15]. To investigate the adequacy of the developed algorithm and the necessity of adaptive scheme for changing system conditions operational and/or topological, the following four operating states of the system were considered: 558

5 State 1: Maximum system load and generation with line L2 is in service (MXLG-1). State 2: Minimum system load and generation with line L2 is in service (MNLG-1). State 3: Maximum system load and generation with line L2 is open (MXLG-2). State 4: Minimum system load and generation with line L2 is open (MNLG-2). Load flows for different operating conditions were conducted and relay currents for faults at different near-end locations in the system were calculated. A coordination time interval of 0.2s was adopted, with a in the range of [0.05-1]. The current transformer (CT) selection was achieved based on the maximum load and fault currents in order to prevent the miscoordination problem resulting from CT s saturation. Table (1) shows the load and fault currents, C.T s and I pu selections and the optimal for the four mentioned operating states for the 33 KV distribution system of the IEEE 30 bus system. Full results of the whole system are listed in [15]. Relay Table (1) - Loads, fault currents and for IEEE 30-bus system Operating condition Load Current [A] Nearend fault current [A] CT Ratio [A] optimal [0.05-1] 1 MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG MNLG / MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG MXLG / MNLG MXLG MNLG

6 Two modes of operation are assumed for the relaying scheme: nonadaptive and adaptive modes. In the nonadaptive mode, the relay settings are optimized based on the concepts discussed in section 2 corresponding to the operating condition of state 1 (MXLG-1: maximum generation and load with line- 2 is in service). Meanwhile, in the adaptive mode the relay settings will be updated according to the significant changes detected in the operating conditions discussed in this section. Tables (2)-(5) show samples of overcurrent relay operating times for the four operating conditions in the nonadaptive and adaptive modes. Table (2) - Samples of OCR operating times for IEEE 30-bus test system (MXLG-1) Optimal coordination based on maximum near-end fault Relay no. (s) Relay no. (s) Relay no. (s) Relay no (s) Table (3) - Samples of OCR operating times for IEEE 30-bus test system (MNLG-1) Nonadaptive mode Adaptive mode Relay no. (s) Relay no. (s) Relay no. (s) Relay no. (s) Table (4) - Samples of OCR operating times for IEEE 30-bus test system (MXLG-2) Nonadaptive mode Adaptive mode Relay no. (s) Relay no. (s) Relay no. (s) Relay no. (s) Table (5) - Samples of OCR operating times for IEEE-30 bus test system (MNLG-2) Nonadaptive mode Adaptive mode Relay no. (s) Relay no. (s) Relay no. (s) Relay no. (s) From the previous four studied states, it is concluded that: In state-1, optimal solution is obtained without any miscoordination case, because the optimal coordination is conformed with this state (maximum load and generation and all lines are in service). In state-2, when the maximum loads and generations are reduced to minimum (operational change), there is 26-miscoordination case in the nonadaptive mode, but in the adaptive mode this problem is removed. In state-3, when the system operates at the maximum load and generation condition, but line-2 is removed (topological change), there is 18-miscoordination case in the nonadaptive mode but in the adaptive mode this problem is also removed. It is worth to mention that only a local sub-system has been affected by the disturbance. In state-4, the system undergoes both topological and operational changes (the maximum loads reduce to minimum with line-2 outage). This case is the worst because it resulted in 35-miscoordination case. These miscoordination cases are removed by following the adaptive protection scheme. 6. Conclusion An adaptive protection scheme for optimal coordination of the directional overcurrent relays in interconnected power networks with an improved formulation has been introduced. This algorithm is applied to the IEEE 30-bus test system with system condition changes, operational and/or topological. The results showed that the proposed adaptive scheme has exhibited optimal performance in all operating states without any miscoordination case. 7. References [1] Applied Protective Relaying, Westinghouse Electric Corporation, Relay- Instrument Division, Coral Springs, Florida 33065, [2] R. E. Albrecht, et al., Digital Computer Protective Device Coordination program -I- General 560

7 Program Description, IEEE Trans on PAS, Vol. 83, No 4, April 1964, pp [3] M. J. Damborg, R. Ramswami, S. Venkata and J. Posforoosh, Computer Aided Transmission Protective System Design, Part I: Algorithms, IEEE Trans on PAS, Vol. 103, No 4, [4] L. Jenkines, H. Khincha, S. Shivakumar and P. Dash, An Application of Functional Dependencies to the Topological Analysis of Protection Schemes, IEEE Transaction on Power Delivery, Vol. 7, No. 1, January 1992, pp [5] A. Urdenta, R. Nadria and L. Jimenez, Optimal Coordination of Directional Overcurrent Relays in Interconnected Power Systems, IEEE Trans. on Power Delivery, Vol. 3, July 1988, pp [6] N. A. Laway and H. O. Gupta, A Method for Coordination of Overcurrent Relays in Interconnected Power Systems, IE Journal, Vol. 74, August 1993, pp [7] B. Chattopadhyay, M. S. Sachdev and T. S. Sidhu, An On- Line Relay Coordination Algorithm For Adaptive Protection Using Linear Programming Technique, IEEE Transactions on Power Delivery, Vol. 11, No. 1, January 1996, pp [8] A. J. Urdenta, L. G. Perez, and H. Resterbo, Optimal Coordination of Directional Overcurrent Relays Considering Dynamic Changes in the Network Topology, IEEE Transaction on Power Delivery, Vol. 12, No. 4, October 1997, pp [9] S. H. Horwitz, A. G. Phadke and J. S. Thorp, Adaptive Transmission System Relaying, IEEE Trans. on Power Delivery, Vol. 3, No. 4, Oct. 1988, pp [10] A. Jampala, S. Venkata and M. Damborg, Adaptive Transmission Protection: Concepts and Computational Issues, IEEE Trans. on Power Delivery, Vol. 4, No. 1, Jan. 1989, pp [11] A. G. Phadke and S. H. Horwitz, Adaptive Relaying, IEEE Computer Applications in Power, July 1990, pp [12] IEEE Working Group, Adaptive Protection and Control Committee, Feasibility of Adaptive Protection and Control, Paper No. 92 SM PWRD. [13] Ammar Hajjar, A.Y. Abdelaziz, H.E.A.Talaat and A.I.Nosseir, Optimal Coordination of Overcurrent Relays by Linear Programming: An Enhanced Problem Formulation, Proceedings of Third CIGRE Regional Conference for Arab Countries, Douha, May [14] IEC Standard, Single Input Energizing Quantity Measuring Relays with Dependent Specified Time, Publication No , [15] Ammar Hajjar, Adaptive Coordination of Overcurrent Relays in Power Distribution Networks, M.Sc thesis in Electrical Engineering, Ain Shams University, June Appendix [A] C G C Line G G Fig. (4) - IEEE 30-bus test system 561