Gerilim kararlılığı gevenilirliğinin expert sistemlerle incelenmesi

Abstract

scheduling, control, design, prediction and instruction and training. The development of an expert system dealing with large and complex mathematical problem, requires a more sophisticated knowledgebase than the one uses human expertise alone. In an expert system for voltage and reactive power control on distribution system with lateral branches using shunt capacitors and voltage regulators, human expertise is not enough to set the rules and database required by the inference engine to build up the desicions for capacitors locations, size and switching time to obtain optimum savings. There are many devices in power system and therefore, in order to describe a power system and its protective system accurately, a great number of consistent rules are required. This makes acquisition and verification of the knowledge difficult. A sample expert system configuration can be used as a money saving on-line controller in co-operation with the input/output interface unit. It also can be used in the distribution system planning stage. Sample expert system utilzes the human experts (HE) heuristics rules as well as the technical literature expertise (TLE) such as algorithms and formula to build the knowledge base of the system. The distribution system component data and network topography are stored in the database. The distribution system voltage profile kept within the desired limits by proper choice of both shunt capacitors and voltage regulators. The most economical mode of operation for the distribution system is ensured all the time, without violating any of the system voltages constraints. In our study, the calculated quantities are shown as tables and as graphics for visual inspectation by operator to decide on how the system going on, is it near to a voltage instability or collapse condition. The expert system represents the characteristics of protective relays by using numerical figures corresponding to the protection zone and time setting. This method provides high maintainability of the system, and also allows the system to display inference process in a user friendly manner. XXXIFirst, power- voltage characteristics of sending end is studied for changing power factor and series compensation ratio, to have an idea about critical values. Then, critical values are directly calculated using given formulas. The most important problem at the functioning of a power system is to limit the voltage of the different points at the system between specific values. At modern power systems, all the relevant information and control operations are recorded to the computers continously. The load -flow studies of the lines and results of the line parameters changes are performed by computers and answers a lot of relevant questions. It is difficult to detect the power transmitted in the region near to the condition of voltage collapse. A computer will detect this condition by a divergence and it is difficult to assess if this is due to the network or to other conditions. One possible method is to programme the power- voltage characteristics of the line and of the load; the intersection of these two characteristics will give the operating conditions of the system. Using a digital computer, successive load flows can be performed with high accuracy in computation. Expert systems are another technological advance which may be used to improve the performance of the these computations. An expert system uses knowledge to solve problems within a limited domain of expertise. Somehow, that knowledge has to be stored within the computer for manipulation by programs that carry out inference and other tasks. The knowledge will be culled from the system in a hotch-potch of different forms and from a variety of sources, like digital relays, controllers etc. Electric utilities often face the problem of finding the exact location of a failure in a distribution line. These failures often occur at the worst possible time and cause the maximum amount of inconvinience to the utility's customers. The utility must quickly find and isolate the failure to restore electric service. Available fault locating methods use fault indicators, thumpers, radars, or combination of these. These methods can produce less than desired results for rapid fault location and may inflict additional damage to the line. As a principle for the stable power supply, any fault sections in a power system must be disconnected quickly. To meet the principle, utilities are making efforts to educate and train their operators for recovering power supply in proper manners. Restoration of power system requires not only expensive knowledge and skills about system components, system operation and protective relays sections, but also accurate and extremely quick judgment based on the knowledge. Because of the importance of the system restoration, development of an expert system efficiently supporting operators is eagerly awaited. Therefore, fault section judgment has been studied as one of the major themes in the field of artificial intelligence application to electric power systems. The expert system applications to the power systems have increased considerably in the last decade. More than 41 % of the expert systems developed recently for power systems are for fault diagnosis. Expert systems are also used in power systems for planning and xxx %^}^'< ??><?'/

scheduling, control, design, prediction and instruction and training. The development of an expert system dealing with large and complex mathematical problem, requires a more sophisticated knowledgebase than the one uses human expertise alone. In an expert system for voltage and reactive power control on distribution system with lateral branches using shunt capacitors and voltage regulators, human expertise is not enough to set the rules and database required by the inference engine to build up the desicions for capacitors locations, size and switching time to obtain optimum savings. There are many devices in power system and therefore, in order to describe a power system and its protective system accurately, a great number of consistent rules are required. This makes acquisition and verification of the knowledge difficult. A sample expert system configuration can be used as a money saving on-line controller in co-operation with the input/output interface unit. It also can be used in the distribution system planning stage. Sample expert system utilzes the human experts (HE) heuristics rules as well as the technical literature expertise (TLE) such as algorithms and formula to build the knowledge base of the system. The distribution system component data and network topography are stored in the database. The distribution system voltage profile kept within the desired limits by proper choice of both shunt capacitors and voltage regulators. The most economical mode of operation for the distribution system is ensured all the time, without violating any of the system voltages constraints. In our study, the calculated quantities are shown as tables and as graphics for visual inspectation by operator to decide on how the system going on, is it near to a voltage instability or collapse condition. The expert system represents the characteristics of protective relays by using numerical figures corresponding to the protection zone and time setting. This method provides high maintainability of the system, and also allows the system to display inference process in a user friendly manner. XXXIFirst, power- voltage characteristics of sending end is studied for changing power factor and series compensation ratio, to have an idea about critical values. Then, critical values are directly calculated using given formulas. The most important problem at the functioning of a power system is to limit the voltage of the different points at the system between specific values. At modern power systems, all the relevant information and control operations are recorded to the computers continously. The load -flow studies of the lines and results of the line parameters changes are performed by computers and answers a lot of relevant questions. It is difficult to detect the power transmitted in the region near to the condition of voltage collapse. A computer will detect this condition by a divergence and it is difficult to assess if this is due to the network or to other conditions. One possible method is to programme the power- voltage characteristics of the line and of the load; the intersection of these two characteristics will give the operating conditions of the system. Using a digital computer, successive load flows can be performed with high accuracy in computation. Expert systems are another technological advance which may be used to improve the performance of the these computations. An expert system uses knowledge to solve problems within a limited domain of expertise. Somehow, that knowledge has to be stored within the computer for manipulation by programs that carry out inference and other tasks. The knowledge will be culled from the system in a hotch-potch of different forms and from a variety of sources, like digital relays, controllers etc. Electric utilities often face the problem of finding the exact location of a failure in a distribution line. These failures often occur at the worst possible time and cause the maximum amount of inconvinience to the utility's customers. The utility must quickly find and isolate the failure to restore electric service. Available fault locating methods use fault indicators, thumpers, radars, or combination of these. These methods can produce less than desired results for rapid fault location and may inflict additional damage to the line. As a principle for the stable power supply, any fault sections in a power system must be disconnected quickly. To meet the principle, utilities are making efforts to educate and train their operators for recovering power supply in proper manners. Restoration of power system requires not only expensive knowledge and skills about system components, system operation and protective relays sections, but also accurate and extremely quick judgment based on the knowledge. Because of the importance of the system restoration, development of an expert system efficiently supporting operators is eagerly awaited. Therefore, fault section judgment has been studied as one of the major themes in the field of artificial intelligence application to electric power systems. The expert system applications to the power systems have increased considerably in the last decade. More than 41 % of the expert systems developed recently for power systems are for fault diagnosis. Expert systems are also used in power systems for planning and xxx %^}^'< ??><?'/scheduling, control, design, prediction and instruction and training. The development of an expert system dealing with large and complex mathematical problem, requires a more sophisticated knowledgebase than the one uses human expertise alone. In an expert system for voltage and reactive power control on distribution system with lateral branches using shunt capacitors and voltage regulators, human expertise is not enough to set the rules and database required by the inference engine to build up the desicions for capacitors locations, size and switching time to obtain optimum savings. There are many devices in power system and therefore, in order to describe a power system and its protective system accurately, a great number of consistent rules are required. This makes acquisition and verification of the knowledge difficult. A sample expert system configuration can be used as a money saving on-line controller in co-operation with the input/output interface unit. It also can be used in the distribution system planning stage. Sample expert system utilzes the human experts (HE) heuristics rules as well as the technical literature expertise (TLE) such as algorithms and formula to build the knowledge base of the system. The distribution system component data and network topography are stored in the database. The distribution system voltage profile kept within the desired limits by proper choice of both shunt capacitors and voltage regulators. The most economical mode of operation for the distribution system is ensured all the time, without violating any of the system voltages constraints. In our study, the calculated quantities are shown as tables and as graphics for visual inspectation by operator to decide on how the system going on, is it near to a voltage instability or collapse condition. The expert system represents the characteristics of protective relays by using numerical figures corresponding to the protection zone and time setting. This method provides high maintainability of the system, and also allows the system to display inference process in a user friendly manner. XXXI