JP2001186651A - Protective relay system and storage medium storing processing program therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To decide relay operation in accordance with the change of a power system. SOLUTION: A protective relay system knowledge base 11 stores knowledge of experts on a protective relay system beforehand as a knowledge base. A power system data base 12 takes in information changing every moment on a power system, and stores it in the data base. A setting support inferring engine 13 infers whether or not a setting for the relay operation judgment is appropriate, on the basis of the power system information stored in the data base 12 and the knowledge base stored in the knowledge base 11. The setting 7 is changed so as to perform relay operation judgment by an appropriate value obtained by the inference.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a protection relay system for removing an accident in an electric power system and preventing the accident from spreading, and a storage medium storing a processing program for the system.

[0002]

2. Description of the Related Art When an accident occurs in a power system, a protection relay system for removing the accident and preventing the propagation of the accident is applied to the power system.

In recent years, with the spread of digital relays, research examples and application examples of a protection relay system using an intelligent system have been known. The intelligent system is known as a more advanced problem solving system that applies computer technology represented by AI (artificial intelligence), fuzzy, neural network, GA (genetic algorithm), and chaos. As a system to which AI is applied, mainly related to business support, design support, settling,
There are application examples for test support. In addition, the adaptive relay is known as a research example and an application example as a relay that changes a relay logic and an operation sensitivity according to a state of a power system.

[0004]

In recent years, with the introduction of power devices such as series capacitors and FACTS devices and the complicated power system, the setting work of protection relays has become complicated, and extensive and specialized knowledge has been required. ing. Furthermore, the power system also changes in a more complicated manner, and in many cases, it is desirable to review the setting and logic of the protection relay according to the situation of the power system in order to perform more effective accident elimination. The setting of protection relays and the change of logic that are optimal for complicated power systems require extensive and specialized knowledge, and these cannot be easily and quickly changed. There is a need for a protection relay system that can adaptively and quickly set protection relays and change logic in response to changes in the power system due to extensive knowledge of experts.

[0005] In order to solve the above problems, the present invention provides a protection relay system and a processing program for the protection relay system which enable the logic and set value of the protection relay system to be quickly and accurately changed in response to changes in the power system. It is intended to provide a storage medium.

[0006]

According to the first aspect of the present invention, there is provided a protection relay system for eliminating an accident when an accident occurs in a power system and preventing the propagation of the accident. A protection relay system knowledge base that is stored in advance as a knowledge base, a power system database that captures the power system information that changes every moment of the power system and stores it in a database, a power system information stored in the power system database, and a protection relay system. Based on the knowledge base stored in the knowledge base, it is inferred whether or not the set value for the relay operation determination is appropriate, and the setting value is changed so that the relay operation is determined based on the appropriate value obtained by the inference. And an inference engine that performs the following. According to this measure, extensive and related to power system information and protection relay system,
Expert knowledge is used to infer the set value of the protection relay system. Then, the set value of the sophisticated and complicated protection relay system is quickly and accurately changed according to the state of the power system. Therefore, when an accident occurs in the power system,
Accidents can be effectively eliminated by the revised setting value of the protection relay system, and the spread of accidents can be prevented.

According to a second aspect of the present invention, in a protection relay system for eliminating an accident when an accident occurs in a power system and preventing the propagation of the accident, knowledge of an expert on the protection relay system is stored in advance as a knowledge base. A protection relay system knowledge base, a power system database that captures the power system information that changes every moment of the power system and stores it in a database, the power system information stored in the power system database, and the protection relay system knowledge base An inference engine that infers whether the relay logic for relay operation determination is valid based on the knowledge base and changes the relay logic to make a relay operation determination based on the valid relay logic obtained by the inference. Is provided. According to this means, the logic of the protection relay system is inferred based on the power system information and the extensive and expert knowledge of the protection relay system. Then, the logic of the sophisticated and complicated protection relay system is quickly and accurately changed according to the state of the power system. Therefore, when an accident occurs in the power system, the accident can be effectively eliminated by the reviewed logic of the protection relay system, and the propagation of the accident can be prevented.

According to a third aspect of the present invention, there is provided the protection relay system according to the first aspect, further comprising a power system knowledge base for preliminarily storing knowledge of the power system including characteristics of the power system, and the power stored in the power system database. Based on the system information and the knowledge base stored in the protection relay system knowledge base and the knowledge base stored in the power system knowledge base, the inference engine infers whether the set value for relay operation determination is appropriate, The setting value is changed so that the relay operation is determined based on a reasonable value obtained by the inference. According to this means, the settling value of the protection relay system is inferred by utilizing the power system information and the extensive and expert knowledge of the protection relay system and the wide power system knowledge including the characteristics of the power system, The set value is quickly and accurately changed according to the state of the power system. Therefore, when an accident occurs in the power system,
Accidents can be effectively eliminated by the revised setting value, and the spread of accidents can be prevented.

According to a fourth aspect of the present invention, there is provided the protection relay system according to the second aspect, further comprising a power system knowledge base for preliminarily storing knowledge of the power system including characteristics of the power system, and the power stored in the power system database. Based on the system information and the knowledge base stored in the protection relay system knowledge base and the knowledge base stored in the power system knowledge base, the inference engine infers whether the relay logic for determining the relay operation is valid, The relay logic is changed so that the relay operation is determined based on the appropriate relay logic obtained by the inference. According to this measure, the logic of the protection relay system is inferred by utilizing the power system information and the extensive and expert knowledge of the protection relay system and the knowledge of the power system including the characteristics of the power system. Is quickly and accurately changed according to the state of the power system. Therefore, when an accident occurs in the power system, the accident can be effectively eliminated by the reviewed logic of the protection relay system, and the propagation of the accident can be prevented.

According to a fifth aspect of the present invention, in the protection relay system according to the third aspect, there is provided a power device knowledge base for preliminarily storing knowledge of a power device including a series capacitor,
Based on the power system information stored in the power system database, the knowledge base stored in the protection relay system knowledge base, the knowledge base stored in the power system knowledge base, and the knowledge base stored in the power equipment knowledge base, The inference engine infers whether the set value for the relay operation determination is appropriate, and changes the set value so that the relay operation is determined based on the appropriate set value obtained by the inference. . According to this measure, extensive and related to power system information and protection relay system,
Expert knowledge and knowledge of a wide range of power systems including power system characteristics and knowledge of power equipment including series capacitors are used to infer the set values of the protection relay system,
The set value is quickly and accurately changed according to the state of the power system. Therefore, when an accident occurs in the power system, it is possible to effectively eliminate the accident by the reviewed set value of the protection relay system and prevent the propagation of the accident.

According to a sixth aspect of the present invention, in the protection relay system according to the fourth aspect, there is provided a power device knowledge base for preliminarily storing knowledge of a power device including a series capacitor,
Based on the power system information stored in the power system database, the knowledge base stored in the protection relay system knowledge base, the knowledge base stored in the power system knowledge base, and the knowledge base stored in the power equipment knowledge base, The inference engine infers whether the relay logic for determining the relay operation is valid or not, and changes the relay logic so that the relay operation is determined based on the valid relay logic obtained by the inference. According to this measure, extensive and expert knowledge of the power system information and protection relay system, and extensive knowledge of the power system including the characteristics of the power system are utilized by utilizing the knowledge of power equipment including series capacitors. The logic of the relay system is inferred, and the logic is quickly and accurately changed according to the situation of the power system. Therefore, when an accident occurs in the power system, the accident can be effectively eliminated by the reviewed logic of the protection relay system, and the propagation of the accident can be prevented.

According to a seventh aspect of the present invention, in a protection relay system for eliminating an accident and preventing the spread of the accident when an accident occurs in the power system, the knowledge of an expert on the protection relay system is stored in advance as a knowledge base. Using the knowledge base of the protection relay system, the power system database that captures and stores the power system information that changes every moment of the power system, and the operation determination by the neural network using the power system information and necessary parameters stored in the power system database Performs a relay operation based on the determination result output by the neural network, the knowledge base stored in the protection relay system knowledge base, and the power system information stored in the power system database. Infer whether the set value for judgment is appropriate, and It is obtained as provided and inference engine using the integer value changes to the determined relay operation by appropriate setting value obtained by logical. According to this means, the power system information is input, the operation is determined by the neural network, and the obtained determination result is output to the inference engine. Then, the validity of the set value of the protection relay system is inferred by utilizing the input determination result, the power system information, and the expert knowledge about the protection relay system. As a result of the inference, the set value of the sophisticated and complicated protection relay system is quickly and accurately changed according to the power system. Therefore, when an accident occurs in the power system, the accident can be effectively eliminated by the set value that has been reviewed by the change, and the propagation of the accident can be prevented.

According to the present invention, in a protection relay system for eliminating an accident and preventing the spread of the accident when an accident occurs in the electric power system, the knowledge of an expert on the protection relay system is stored in advance as a knowledge base. Operation using a neural network using the protection relay system knowledge base, a power system database that captures and stores the power system information that changes every moment of the power system, and the power system information and necessary parameters stored in the power system database. A neural network that performs a determination and outputs a determination result; and a relay based on the determination result output by the neural network, the knowledge base stored in the protection relay system knowledge base, and the power system information stored in the power system database. Inference of the validity of relay logic for motion judgment Performed is obtained by so provided and inference engine to a change of the relay logic to decision relay operation by appropriate relay logic obtained by inference. According to this means, the power system information is input, the operation is determined by the neural network, and the obtained determination result is output to the inference engine. And
The validity of the relay logic of the protection relay system is inferred by utilizing the input determination result, the power system information, and the expert knowledge about the protection relay system. As a result of this inference, the relay logic of the sophisticated and complicated protection relay system is quickly and accurately changed according to the power system. Therefore, when an accident occurs in the power system, the accident can be effectively eliminated by the relay logic reviewed by the change, and the propagation of the accident can be prevented.

According to a ninth aspect of the present invention, there is provided the protection relay system according to the seventh aspect, further comprising a power system knowledge base for preliminarily storing knowledge of the power system including characteristics of the power system, and a determination result output by a neural network. Based on the power system information stored in the power system database, the knowledge base stored in the protection relay system knowledge base, and the knowledge base stored in the power system knowledge base, for inference engine to determine the relay operation. Inference is made as to whether the set value is appropriate, and the set value is changed so that the relay operation is determined based on the appropriate set value obtained by the inference. According to this means, the operation is determined by the neural network, the obtained determination result is output to the inference engine, and the knowledge of the power system including the power system information and the expert knowledge of the protection relay system and the power system characteristics is included. Is used comprehensively to infer the validity of the set value of the protection relay system. Based on the inference result, the set value is quickly and accurately changed according to the power system.

According to a tenth aspect of the present invention, there is provided the protection relay system according to the eighth aspect, further comprising a power system knowledge base for preliminarily storing knowledge of the power system including characteristics of the power system, and a determination result output by a neural network. Based on the power system information stored in the power system database, the knowledge base stored in the protection relay system knowledge base, and the knowledge base stored in the power system knowledge base, for inference engine to determine the relay operation. Inference is made as to whether or not the relay logic is appropriate, and the relay logic is changed so that the relay operation is determined based on the appropriate relay logic obtained by the inference. According to this means, the operation is determined by the neural network, the obtained determination result is output to the inference engine, and the knowledge of the power system including the power system information and the expert knowledge of the protection relay system and the power system characteristics is included. Is used comprehensively to infer the validity of the relay logic. The inference result allows the relay logic to be quickly and accurately changed according to the power system.

According to an eleventh aspect of the present invention, there is provided the protection relay system according to the ninth aspect, further comprising a power device knowledge base for preliminarily storing knowledge of the power device including the series capacitor, and a determination result output by a neural network. When,
Based on the power system information stored in the power system database, the knowledge base stored in the protection relay system knowledge base, the knowledge base stored in the power system knowledge base, and the knowledge base stored in the power equipment knowledge base The inference engine infers whether the set value for determining the relay operation is appropriate, and changes the set value so that the relay operation is determined based on the appropriate set value obtained by the inference. . According to this means, the operation is determined by the neural network, the obtained determination result is output to the inference engine, and the knowledge of the power system including the power system information, the expert knowledge of the protection relay system, and the characteristics of the power system is included. The knowledge of the power equipment including the capacitor and the power is comprehensively utilized to infer the validity of the set value of the protection relay system. Based on the inference result, the set value is quickly and accurately changed according to the power system.

According to a twelfth aspect of the present invention, there is provided the protection relay system according to the tenth aspect, further comprising a power device knowledge base for preliminarily storing knowledge of a power device including a series capacitor, and a determination result output by a neural network. When,
Based on the power system information stored in the power system database, the knowledge base stored in the protection relay system knowledge base, the knowledge base stored in the power system knowledge base, and the knowledge base stored in the power equipment knowledge base The inference engine infers whether the relay logic for determining the relay operation is valid or not, and changes the relay logic so as to determine the relay operation based on the valid relay logic obtained by the inference. . According to this means, the operation is determined by the neural network, and the obtained determination result is output to the inference engine.
Expertise on power system information and protection relay systems, power system knowledge base including power system characteristics and power equipment knowledge including capacitors are comprehensively used to infer the validity of relay logic and infer. As a result, the logic is quickly and accurately changed according to the power system.

According to a thirteenth aspect of the present invention, there is provided a storage medium storing a processing program of a protection relay system for eliminating an accident and preventing the spread of the accident when an accident occurs in the power system. A protection relay system knowledge base that stores knowledge in advance as a knowledge base, a power system database that captures and stores momentarily changing power system information of the power system, a power system information stored in the power system database, and a protection relay system. Based on the knowledge base stored in the knowledge base, it is inferred whether the set value for relay operation determination or the relay logic is appropriate, and the relay operation is determined based on the appropriate set value or relay logic obtained by the inference. To change the set value or relay logic It is an storage medium storing a processing program for executing processing with down.

According to a fourteenth aspect of the present invention, there is provided a storage medium storing a processing program of a protection relay system for eliminating an accident and preventing the propagation of the accident when an accident occurs in the power system. Using a protection relay system knowledge base that stores knowledge as a knowledge base in advance, a power system database that captures and stores power system information that changes every moment of the power system, and a power system information stored in the power system database, A neural network that performs operation determination by a net and outputs a determination result, and a determination result output by the neural network and power system information stored in a knowledge base and a power system database stored in a protection relay system knowledge base. Set value for relay operation judgment or relay A processing program for inferring whether the logic is appropriate and executing a process with an inference engine that changes the set value or the relay logic so as to determine the relay operation based on the appropriate set value or the relay logic obtained by the inference. It is assumed to be a storage medium for storing.

[0020]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a conceptual configuration diagram of hardware of a protection relay system applied to an embodiment of the present invention.

In the figure, the protection relay system 100
Is composed of an input conversion unit 1, a CPU 2, a program storage unit 3, a data storage unit 4, a relay output unit 5, a bus 10, and the like.
The data storage unit 4 stores data including a relay logic 6, a set value 7, a database 8, and a knowledge base 9,
Further, the program storage unit 3 stores various processing programs including the inference engine 3a and the like.

Here, the knowledge base and the inference engine are represented by A
A term often used in the field of artificial intelligence (I), a knowledge base stores factual knowledge about a target problem area and heuristic (hierarchical) knowledge of an expert, and stores the knowledge of the expert. It is a database that is hierarchically stored in an easy-to-understand manner. Also, the inference engine is generally
It is a control mechanism that solves and infers problems based on the knowledge base.

First, a voltage value and a current value are taken from the power system 101 to the input converter 1 of the protection relay system 100 via the CT 102 and the PT 103. When the voltage value and the current value are taken into the input conversion unit 1, the input conversion unit 1 converts the voltage value and the current value into digital signals that can be processed by the CPU 2. The converted digital signal is stored in the database 8 of the data storage unit 4 via the bus 10. CPU2
Then, the inference engine 3 stored in the program storage unit 3
a, the processing of the first to twelfth embodiments described below is performed, and the circuit breaker (CB) 1
04 is operated.

FIG. 2 is an explanatory diagram showing the processing of the protection relay system according to the first embodiment of the present invention.

The first embodiment of the present invention is characterized by including a protection relay system knowledge base 11, a power system database 12, and a settling support inference engine 13.

Here, the protection relay system knowledge base 1
Reference numeral 1 stores expert knowledge about the protection relay system in advance as a knowledge base. The power system database 12 captures power system information of the power system and stores it in the database. The settling support inference engine 13 infers the power system information of the power system database 12 with the knowledge base of the protection relay system knowledge base 11 and changes the set value.

First, information on the power system 101 that changes every moment is stored in the power system database 12. In the protection relay system knowledge base 11, expert knowledge about the protection relay system is stored in advance. The settling support inference engine 13 uses the protection relay system knowledge base 1
It is inferred based on 1 whether the set value 7 is appropriate.

For example, when the load current is large, when the influence of the neighboring line is large, when the system configuration changes and the zero-phase current increases, it is necessary to change the set value 7 or the relay operation sensitivity. The validity of the set value 7 for a system change such as a large load current is stored in advance in the protection relay system knowledge base 11, and when information on a system change such as a large load current is input to the power system database 12, the setting support is performed. Inference engine 1
3 is set value 7 based on protection relay system knowledge base 11
Is inferred as appropriate, and the set value 7 is changed to an appropriate value.

FIG. 3 is a simple explanatory diagram of the knowledge base.

The knowledge base is usually hierarchical,
It is composed of several frames. Each frame has several slots, and the slots store values or rules. Further, the frame has a hierarchical structure, and a frame of an upper layer (the parent frame 26 in this figure) is a frame of several lower layers (in this figure, two child frames 26A,
26B), and the slot of the frame in the upper layer can inherit the slot and the values and rules contained in the slot in the frame of the lower layer. In the example of this figure, the lower layer frame is the slot 1 of the upper layer frame,
Slot 2, slot 8, and slot 9 are inherited.

FIG. 4 shows the protection relay system knowledge base 1
1 is an explanatory diagram of a configuration example of FIG.

To explain an example composed of three layers of an upper frame, a parent frame and a child frame, the upper frame 11A usually contains information for managing the connection and the configuration of the entire knowledge base. The parent frames are a relay frame 11C and an electric station frame 11B. The relay frame 11C is configured for each type of relay. DZS (short circuit type distance relay), DZG (ground fault type distance relay) and the like.

The electric station frame 11B stores information input to an electric station where a relay is installed. For example, the voltage, current, and the like for each line required for protection of the line A and protection of the line B. There are several protection systems in electric stations, such as track protection and bus protection. For example, when there are two protection systems for line protection of the A line and the B line,
In this figure, the protection system for the line A is a line A protection frame 11E, and the protection system for the line B is a line A protection frame 11D.

FIG. 5 shows the circuit A protection frame 1 shown in FIG.
It is a figure explaining 1E.

The line A protection frame 11E has a slot in which information such as voltage and current is stored, which is inherited from the electric station frame 11B shown in FIG. 4, and a set value which is inherited from the relay frame 11C shown in FIG. , A slot in which rules are stored.

Information necessary for the line A protection frame 11E, such as voltage and current, is periodically input to the substation frame 11B shown in FIG. When the value of the slot storing the information such as the voltage and current of the substation frame 11B is updated,
The value of the slot storing the information such as the voltage and current of the line A protection frame 11E is also updated by inheriting the value of the parent frame. When the values of these slots are updated, the rules stored in the affected slot 11F of the line adjacent to the substation frame 11B and the affected slot 11G of the load current are activated (these rules are used to determine the change in the value of the designated slot). Triggered by degree).

Next, when the rule stored in the affected slot 11F of the neighboring line or the rule stored in the affected slot 11G of the load current is activated, the relay stored in the slot of the line A protection frame 11E is activated. If the set value is not valid, change the set value to a valid value. Thus, the change in the load current is sensed and the line A protection frame 11 is detected.
The rule stored in the slot of E is activated, and the set value is changed to an appropriate value. These series of operations are driven by the settling support inference engine 13.

Influence slot 11G of load current shown in FIG.
The following example is a rule stored in.

(Rule 1) If the load current of the A line is large (more than a certain value), the DZ which is affected by the change of the load current
What is the setting item of G? Here, from the DZG relay frame, a setting item that affects the change in the load current is selected. (Rule 2) The change in the change in the load current is compared with the set value of the current set item to determine whether it is appropriate. Rule 3) If the set value is not appropriate, calculate the recommended value.

FIG. 6 is a diagram for explaining a change in the impedance seen by the relay due to the influence of the load current.

In FIG. 6, the horizontal axis represents the impedance (for R) 23 seen by the relay, and the vertical axis represents the impedance (for X) 22 seen by the relay. As shown in FIG. 6, assuming that the operation area in which the relay operates is inside the circle shown in the relay operation area 19, the impedance seen by the relay from point A depends on the magnitude and direction of the load current even at the same fault point. B
The point or the point B changes to the point A, and the former may cause the malfunction of the relay and the latter may cause the malfunction. When changing from the point A to the point B, it is necessary to expand the relay operation area 19 so as to cover the point B in order to prevent a malfunction and a malfunction.
Conversely, when changing from point B to point A, it is necessary to reduce the size so that point A does not enter relay operation area 19 so as to prevent malfunction. The change in the impedance seen by the relay due to the effect of the load current can be obtained from a table or an approximate expression based on a preliminary simulation.

As described above, the protection relay system according to the first embodiment of the present invention uses a knowledge base of the protection relay system and a database in which information of a power system that changes every moment is input, and uses the power system by an inference engine by using these. The setting value of the protection relay system is adaptively changed in accordance with the change of. Thereby, even if the state of the power system changes, it is possible to quickly correct the set value using the knowledge of the expert of the protection relay system.

FIG. 7 is an explanatory diagram showing the processing of the protection relay system according to the second embodiment of the present invention.

The second embodiment of the present invention is characterized by having a protection relay system knowledge base 11, a power system database 12, and a logic support inference engine 14.

Here, the protection relay system knowledge base 1
Reference numeral 1 stores expert knowledge about the protection relay system in advance as a knowledge base. The power system database 12 captures power system information and stores it in the database. Logic support inference engine 14
Is to make inferences based on the power system information of the power system database 12 and the knowledge base of the protection relay system knowledge base 11 and change the logic.

First, the system information of the power system 101 that changes every moment is stored in the power system database 12. In the protection relay system knowledge base 11, expert knowledge regarding the protection relay system is stored in advance. The logic-assisted inference engine 14 infers whether the relay logic 6 is valid based on the protection relay system knowledge base 11.

For example, there are cases where the change of the power system 101 input to the power system database 12 is large and cannot be dealt with only by changing the set value 7. For example, if the zero-phase current is large or the branch current is large, it is necessary to add compensation logic to the relay logic. In such a case, when information on system change such as a large zero-phase current is input, the logic support inference engine 14 makes an inference based on the protection relay system knowledge base 11 to determine whether the relay logic 6 is appropriate, and If so, a logic such as zero-phase compensation is added and the relay logic 6 is changed.

FIG. 8 is a diagram showing an example of a line A protection frame 11E similar to that shown in FIGS. 4 and 5 for explaining the addition of logic such as zero-phase compensation according to the second embodiment.

The illustrated DZG zero-phase compensation slot 11H
Does not require zero-phase compensation when the zero-phase current is small or the branch current is small. In this case, the contents of the slot are set to none. However, when the zero-phase current is large or the branch current is large, zero-phase compensation is required. If there is a possibility of malfunction or malfunction due to the zero-phase current, the rule in the affected slot of the adjacent line is activated, and the contents of the DZG zero-phase compensation slot 11H are assumed to exist, and logic for zero-phase compensation and the like is executed. Add. These operations are driven by the logic-assisted inference engine 14.

As described above, the protection relay system according to the second embodiment of the present invention uses a knowledge base of the protection relay system and a database in which information of a power system that changes every moment is input, and uses the database to input power system information by an inference engine. Adaptively change the logic of the protection relay system in response to changes in As a result, even if the state of the power system changes, it is possible to quickly modify the logic by using the expert knowledge of the protection relay system.

FIG. 9 is an explanatory diagram showing the processing of the protection relay system according to the third embodiment of the present invention.

In the third embodiment of the present invention, a power system knowledge base 15 is added to the first embodiment shown in FIG. 1, and the set value of the protection relay system is changed according to a change in the power system. It has a function to perform.

Here, the power system knowledge base 15 stores in advance the power system knowledge including the characteristics of the power system.

For example, when an emergency load shedding is performed when the frequency drops, in order to obtain an optimum load shedding amount, the instantaneous reserve capacity of the generator, the load characteristics, and the system capacity are greatly related. In addition, when considering the abnormal voltage after the load is cut off, it is necessary to consider the reactive power characteristics of the load and the capacity that can be input (opened) of the phase adjustment equipment. Knowledge in consideration of these characteristics of the power system is stored in the power system knowledge base 15.

Further, it is necessary to coordinate the timing of load shedding with other protection relay systems, and knowledge of the cooperation is stored in the protection relay system knowledge base 11. It is necessary to determine an optimum set value of the cutoff amount based on the knowledge of the power system and the knowledge of the protection relay system. The settling support inference engine 13 is a power system knowledge base 1
Set value 7 based on 5 and protection relay system knowledge base 11
Is inferred as appropriate, and the set value 7 is changed to an appropriate value.

As described above, since the protection relay system according to the third embodiment has the knowledge base of the power system, in addition to the effect of the protection relay system according to the first embodiment, the state of the power system changes. In addition, it is possible to quickly correct the set value of the protection relay system according to changes in the power system, using the knowledge of the protection relay system and power system experts.

FIG. 10 is an explanatory diagram showing the processing of the protection relay system according to the fourth embodiment of the present invention.

The fourth embodiment differs from the second embodiment shown in FIG. 3 in that a power system knowledge base 15 is additionally provided, and a function having an inference engine for changing the logic of the protection relay system according to a change in the power system. It is provided with.

For example, when an emergency load shedding is performed when the frequency drops, a voltage abnormality after the load shedding is large, and when this is taken into consideration, if the load shedding is not desired or if the phase adjustment equipment needs to be turned on (opened), a logic operation is performed. Need to be changed. In this case, the power system knowledge base 15 stores knowledge that load shedding is not desirable or that it is necessary to input (open) a phase adjustment facility.

The logic-assisted inference engine 14 includes a power system knowledge base 15 and a protection relay system knowledge base 11.
The relay logic 6 is appropriate based on
If necessary, logic such as input (opening) of the phase adjustment equipment is added, and the relay logic 6 is changed.

As described above, since the protection relay system according to the fourth embodiment of the present invention has the power system knowledge base, in addition to the effect of the protection relay system according to the second embodiment, the state of the power system can be changed. Even if it changes, it becomes possible to quickly modify the logic of the protection relay system according to the change of the power system, using the knowledge of the protection relay system and power system experts.

FIG. 11 is an explanatory diagram showing the processing of the protection relay system according to the fifth embodiment of the present invention.

In the fifth embodiment, a power device knowledge base 16 such as a series capacitor is added to the third embodiment shown in FIG. 9, and the set value of the protection relay system is changed according to a change in the power system. Is provided with a function having an inference engine for changing

For example, when a series capacitor is installed in the power system, the electrical distance (impedance) of the protection relay to the fault point changes depending on the impedance behind and the installation point of the series capacitor. Also, when equipment protection is added to the series capacitor, the electrical distance of the protection relay up to the accident point varies depending on the operation status of equipment protection.

Assuming such a case, the knowledge of the change in the electrical distance seen by the protection relay up to the point of accident when the series capacitor is installed in the power system is stored in the power system knowledge base 1.
5 and the knowledge of device protection of the series capacitor is stored in the power device knowledge base 16. The settling support inference engine 13 infers whether the set value 7 is valid based on the power device knowledge base 16, the power system knowledge base 15, and the protection relay system knowledge base 11, and changes the set value 7 to a proper value. .

As described above, since the protection relay system according to the fifth embodiment of the present invention has a knowledge base of power devices, it has the effect of the protection relay system according to the third embodiment as well as the power system. Even if the state changes, it becomes possible to quickly correct the set value of the protection relay system according to the change in the power system, using the knowledge of the specialist in the protection relay system, the power system and the power equipment.

FIG. 12 is an explanatory diagram showing the processing of the protection relay system according to the sixth embodiment of the present invention.

In the sixth embodiment, a power device knowledge base 16 such as a series capacitor is added to the fourth embodiment shown in FIG. 10, and the logic of the protection relay system is changed according to a change in the power system. It has a function having an inference engine to be changed.

For example, if a series capacitor is present in a place where the electric distance seen by the protection relay is short due to system switching or the like in the electric power system, it is necessary to change the logic to take account of the series capacitor. In this case, the power system knowledge base 15 stores the knowledge when a series capacitor exists at a place where the electric distance is short. If a series capacitor exists in the protection relay system knowledge base 11, knowledge that a logic change is required is stored. In addition, the knowledge of equipment protection for series capacitors is transferred to the power equipment knowledge base 16.
To be stored.

The logic support inference engine 14 includes a power system knowledge base 15 and a protection relay system knowledge base 11.
Logic 6 based on the power device knowledge base 16
Is appropriate, and if necessary, the logic considering the series capacitor is added, and the relay logic 6 is changed.

As described above, since the protection relay system according to the sixth embodiment of the present invention has the power device knowledge base, in addition to the effects of the protection relay system according to the fourth embodiment, the state of the power system Even if the power supply system changes, it is possible to quickly modify the logic of the protection relay system according to the change in the power system, using the knowledge of the specialist in the protection relay system, the power system, and the power equipment.

FIG. 13 is an explanatory diagram showing the processing of the protection relay system according to the seventh embodiment of the present invention.

The protection relay system according to the seventh embodiment of the present invention uses a protection relay system knowledge base 11, a power system database 12, and a neural network 20, and uses a protection relay system set value 7 according to a change in the power system. Is provided with a function having a settling support inference engine 13 for changing.

Here, the protection relay system knowledge base 1
Reference numeral 1 stores expert knowledge about the protection relay system in advance as a knowledge base. The power system database 12 captures power system information of the power system and stores it in the database. Neural net 20
Is for performing operation determination using power system information and necessary parameters, and outputting a determination result. The setting support inference engine 13 infers a setting value based on the determination result of the neural network 20, the knowledge base of the protection relay system knowledge base 11, and the power system information, and changes the setting value.

First, information of the power system 101 that changes every moment is input to the power system database 12. In the protection relay system knowledge base 11, expert knowledge about the protection relay system is stored in advance. The settling support inference engine 13 uses the neural network 20 to infer whether the set value 7 is valid based on the protection relay system knowledge base 11.

For example, when the load current is large, when the influence of the adjacent line is large, when the system configuration changes and the zero-phase current increases, it is necessary to change the set value 7 or the relay operation sensitivity. The validity of the set value 7 for system changes such as a large load current is stored in the protection relay system knowledge base 11 in advance.

When information on a system change such as a large load current is input to the power system database 12, the neural network 20 outputs a neural network determination result 21 using the information in the power system database 12. The settling support inference engine 13 infers whether the set value 7 is appropriate based on the protection relay system knowledge base 11 by using the determination result 21 of the neural network, and changes the set value 7 to an appropriate value.

Here, the neural network used in the present invention will be described with reference to FIG.

The neural network is a technique that simulates the nerve cells of the human brain in an engineering manner. Recently, researches have been conducted in the field of electric power systems. Are known. As a feature, it is excellent in pattern recognition ability and interpolation characteristics.

As shown in FIG. 14, the neural network 2
Numeral 0 is composed of neurons 30 corresponding to human brain cells and interneuron connections 31 corresponding to nerves connecting the brain cells. It is said that human brain cells develop nerves that connect brain cells by learning. In a neural network, this is simulated by weighting the connections 31 between neurons. That is, the weight of the inter-neuron connection 31 is determined by learning.

As shown in FIG. 14, neurons are divided into several layers. That is, the input layer 32 and the intermediate layer 3
Third, the output layer 34. The example of FIG. 14 is called a three-layer type neural network, and includes an input layer 32, an intermediate layer 33, and an output layer 34.
Each layer. The phenomenon is taught to the input layer 32 and the output layer 34
Is taught the judgment result. By teaching the phenomena of an appropriate number of times and the judgment result, the weight of the inter-neuron connection 31 is determined (learning), and the neural network can judge even the phenomena that have not been learned (null). -Called pattern recognition and interpolation of the Ralnet). Generally, it takes a long time to learn a neural network, but once the learning is completed, the determination becomes faster.

FIG. 15 is an explanatory diagram of the neural network 20 used in the seventh embodiment.

In the figure, the neural network 20 includes, as an input layer 32, a load current 35, an adjacent line current 36, a system configuration 37, an impedance (for R) 38 seen by a relay,
The impedance (for X) seen by the relay was 39. The operation determination 40 is set as the output layer 34.

When an accident occurs in the electric power system, the load current 35 and the adjacent line current 3
6. The impedance seen by the relay (for R) 38, the impedance seen by the relay (for X) 3 depending on the system configuration 37
9 is different. Therefore, when the set values of the relays are the same, there is a possibility that the relay malfunctions or malfunctions.

The above is described with reference to FIG. The vertical axis shown in FIG. 16 is the impedance seen by the relay (X component)
39, the horizontal axis is the impedance seen by the relay (R) 38
It is. The relay operation area 41 operates in the garden shown in FIG. 16 when the impedance (for R) 38 seen by the relay and the impedance (for X) 39 seen by the relay enter the garden. However, due to the effects of the load current 35, the adjacent line current 36, and the system configuration 37, the impedance seen by the relay (for R) 38 and the impedance seen by the relay (for X) 3
As shown in FIG. 16, when the point 9 moves from the point A to the point B, a malfunction occurs, and when the point 9 moves from the point A to the point B, a malfunction occurs.

Next, an example of learning of the neural network 20 used in the seventh embodiment will be described with reference to FIG.

The input layer 32 has a load current 35, an adjacent line current 3
6. System configuration 37, impedance seen by the relay (R
) And the impedance (X component) 39 seen by the relay to learn the neural network 20 using these as parameters. The output layer 34 is an operation determination 40.

First, the range of the fault point where the relay operates is determined. Next, a simulation is performed using the load current 35, the adjacent line current 36, the system configuration 37, and the fault point as parameters.
If the fault point is within the range of the fault point at which the relay operates, the operation judgment 40 is set to 1; The impedance seen by the relay at this time (for R) 3
8. Record the impedance (for X) 39 seen by the relay. In this way, a simulation result is created,
Based on this, the neural network 20 is made to learn.

After the learning is completed, the impedance (R component) 38 and the impedance (X component) 39 seen by the relay are set in the neural network 20 using the load current 35, the adjacent line current 36, and the value of the system configuration 37 in a certain system state. Make the parameter make a motion decision.

FIG. 17 shows an example of the operation judgment by the neural network 20.

Assuming that the area A is the area where the neural network 20 is determined to be inactive and the area B is the area where the neural network 20 is determined to be inactive, the area A (inactive) and the relay operating area 41 partially overlap, and the original In the setting of the relay operation area 41, a malfunction occurs at an overlapping portion.

FIG. 18 is a diagram in which the original relay operation area 41 is changed to the new relay operation area 42, and the portion overlapping the area A (non-operation) is eliminated. Since there is no overlapping part, malfunction of the relay can be prevented. However, whether or not to settle in the new relay operation area 42 requires coordination with other setting and other relays, and requires the knowledge of a protection relay expert.

As described above, the protection relay system according to the seventh embodiment of the present invention uses a knowledge base of the protection relay system, a database in which information of a power system that changes every moment is input, and an inference engine using a neural network. The setting value of the protection relay system is changed according to the change of the system. Thereby, even if the state of the power system changes, it is possible to quickly correct the set value using the knowledge of the expert of the protection relay system.

FIG. 19 is an explanatory diagram showing the processing of the protection relay system according to the eighth embodiment of the present invention.

The protection relay system according to the eighth embodiment of the present invention uses a protection relay system knowledge base 11, a power system database 12, and a neural network 20, and changes the relay logic 6 according to a change in the power system. It has a function having a support inference engine 14.

Here, the protection relay system knowledge base 1
Reference numeral 1 stores expert knowledge about the protection relay system in advance as a knowledge base. The power system database 12 captures power system information of the power system and stores it in the database. Neural net 20
Is for performing operation determination using power system information and necessary parameters, and outputting a determination result. The logic support inference engine 14 performs logic inference based on the determination result of the neural network 20, the knowledge base of the protection relay system knowledge base 11, and the power system information, and changes the logic.

First, the change of the power system 101 input to the power system database 12 is so large that it may not be possible to cope with it only by changing the set value 7 according to the seventh embodiment.
For example, if the zero-phase current is large or the branch current is large, it is necessary to add compensation logic to the relay logic. In such a case, when information on system change such as a large zero-sequence current is input, the logic support inference engine 14 checks the neural network 20 based on the protection relay system knowledge base 11 to determine whether the relay logic 6 is appropriate. And make inferences,
If necessary, a logic such as zero-phase compensation is added, and the relay logic 6 is changed.

FIG. 20 is an explanatory diagram of the neural network 20 used in the eighth embodiment.

In the figure, as the input layer 32, a load current 35, an adjacent line current 36, a system configuration 37, an impedance (for R) 38 seen by a relay, and an impedance (for X) 39 seen by a relay are used. The compensation coefficient (
R component) 43 and compensation coefficient (X component) 44.

As described with reference to FIG.
5. The impedance seen by the relay changes due to the influence of the adjacent line current 36 and the system configuration 37, resulting in malfunction or malfunction.

Next, an example of learning of the neural network 20 used in the eighth embodiment will be described with reference to FIG.

First, the range of the fault point where the relay operates is determined. Next, a simulation is performed using the load current 35, the adjacent line current 36, the system configuration 37, and the fault point as parameters.
From this result, the impedance (R
Min) and the impedance (R component) 38 as seen by the relay, and the compensation coefficient (R component) 43, which is the quotient of the impedance (R component) up to the actual accident point and the impedance (X component) 39 seen by the relay. Is calculated as the compensation coefficient (for X). That is, the calculation is performed as in the following equations (1) and (2).

Compensation coefficient (for R) 43 = impedance to actual accident point (for R) / impedance seen by relay (for R) 38 (1) Compensation coefficient (for X) 44 = actual accident point Impedance up to (R) / Impedance seen by the relay (X) 39 ... (2)

Next, the load current 35, the adjacent line current 36, the system configuration 37, and the impedance (for R) seen by the relay 3
8. The impedance (for X) 39, the compensation coefficient (for R) 43, and the compensation coefficient (for X) 44 seen by the relay are recorded, and learning data for the neural network 20 is created. The neural network 20 is trained based on the learning data thus created. After the learning, even if the system state changes, the neural network 20 uses the values of the load current 35, the adjacent line current 36, and the system configuration 37 to obtain a compensation coefficient (for R) 4.
3. Using the compensation coefficient (for X) 44, the correct impedance up to the actual accident point can be obtained. However, the compensation coefficient (for R) 43 obtained from the neural network 20,
Whether the compensation coefficient (for X) 44 is used as it is requires coordination with other relays or the like, and requires the knowledge of a protection relay expert.

For this purpose, as shown in FIG. 19, the logic assisted inference engine 14 infers whether the relay logic 6 is appropriate based on the protection relay system knowledge base 11 by using the neural network 20, and if necessary, if necessary. Logic such as zero-phase compensation such as addition of a compensation coefficient is added, and the relay logic 6 is changed.

As described above, the protection relay system according to the eighth embodiment of the present invention uses a knowledge base of the protection relay system, a database in which information of a power system that changes every moment is input, and an inference engine using a neural network. Adaptively change the logic of the protection relay system in response to changes in the system. As a result, even if the state of the power system changes, it is possible to quickly modify the logic by using the expert knowledge of the protection relay system.

FIG. 21 is an explanatory diagram showing the processing of the protection relay system according to the ninth embodiment of the present invention.

The protection relay system according to the ninth embodiment of the present invention is different from the protection relay system of the seventh embodiment in that a power system knowledge base 15 is additionally provided, and a neural network 20 is used. Setting support inference engine 13 that changes the setting value of protection relay system according to
It has the function of having.

Here, the power system knowledge base 15 stores in advance the power system knowledge including the characteristics of the power system.

First, for example, when an emergency load shedding is performed when the frequency drops, in order to obtain an optimum load shedding amount,
The generator's spinning reserve, load characteristics, and system capacity are closely related. In addition, when considering the abnormal voltage after the load is cut off, it is necessary to consider the reactive power characteristics of the load and the capacity that can be input (opened) of the phase adjustment equipment. Knowledge in consideration of these characteristics of the power system is stored in the power system knowledge base 15.

Further, it is necessary to coordinate the timing of load shedding and the like with other protection relay systems, and knowledge of these cooperations is stored in the protection relay system knowledge base 11. It is necessary to determine the optimal set value of the cutoff amount based on the knowledge of the power system and the knowledge of the protection relay system.
The settling support inference engine 13 uses the neural network 20 to infer whether the set value 7 is appropriate based on the power system knowledge base 15 and the protection relay system knowledge base 11,
Change the setting value 7 to an appropriate value.

FIG. 22 is an explanatory diagram of the neural network 20 used in the ninth embodiment.

In the figure, as the input layer 32, the supply-demand unbalance ratio 45, the load characteristic (V) 46, the load characteristic (F)
47, generator spinning reserve 48, frequency minimum 49, and frequency steady value 50. The load shedding amount 51 was used as the output layer.

The terms shown in FIG. 22 are defined as follows. Supply-demand imbalance rate 45 = shortage of power generation to total demand /
Total power generation of the system Load characteristics (V) 46; Voltage characteristics of load Load characteristics (F) 47; Frequency characteristics of load Generator instantaneous reserve 48; Sum of instantaneous reserve of generators of the entire system Minimum frequency Value 49: When power generation is insufficient due to generator disconnection or interconnection disconnection, frequency stabilization control is performed by load interruption, but the frequency continues to decrease due to inertia and eventually recovers. The minimum value of the frequency at this time Frequency steady value 50: The value when the frequency is recovered and becomes almost constant after performing the frequency stabilization control by load shedding, the load shedding amount 51; The amount of load shedding to stabilize

Next, an example of learning of the neural network 20 used in the ninth embodiment will be described.

First, a target system is determined, and a method of simulating a sudden power generation shortage due to generator disconnection or interconnection interruption is determined. Then, a simulation is performed using the supply / demand imbalance ratio 45, the load characteristic (V) 46, the load characteristic (F) 47, the instantaneous power reserve 48 of the generator, and the load shedding amount 51 as parameters. From this result, a minimum frequency value 49 and a steady frequency value 50 are obtained.

Further, a supply / demand imbalance rate 45, a load characteristic (V) 46, a load characteristic (F) 47, an instantaneous power reserve 48 of the generator, a load shedding amount 51, a minimum frequency value 49, and a steady frequency value 50 are recorded. Then, learning data for the neural network 20 is created. The neural network 20 is trained based on the learning data thus created.

After the learning is completed, the supply-demand imbalance rate 45, the load characteristic (V) 46, the load characteristic (F) 47,
Generator spinning reserve 48, minimum frequency 4 to be finished
9. If the frequency steady value 50 is input, the required load shedding amount 5
1 can be obtained.

However, whether to use the load shedding amount 51 obtained by the neural network 20 as it is is necessary to consider coordination with other relays or to maintain the voltage after load shedding, and the like. The latter requires knowledge of the power system.

Therefore, as described in the description of FIG. 21, the settling support inference engine 13 determines whether the voltage after the load is cut off can be maintained with the protection relay system knowledge base 11 in which knowledge of cooperation with other relays is stored. Is inferred based on the power system knowledge base 15 using the determination result 21 of the neural network, and the set value 7 (load shedding amount) is changed to an appropriate value.

As described above, the protection relay system according to the ninth embodiment of the present invention has a power system knowledge base, so that in addition to the effects of the protection relay system according to the seventh embodiment, Even if the power supply changes, it becomes possible to quickly correct the set value of the protection relay system according to the change of the power system by using the knowledge of the protection relay system and the power system expert and the neural network.

FIG. 23 is an explanatory diagram showing the processing of the protection relay system according to the tenth embodiment of the present invention.

The protection relay system according to the tenth embodiment of the present invention differs from the protection relay system of the eighth embodiment in that it has a power system knowledge base 15, uses a neural network 20, and responds to changes in the power system. And a logic assisted inference engine 14 for changing the relay logic of the protection relay system.

For example, when the emergency load shedding is performed when the frequency drops, the voltage abnormality after the load shedding is so large that if this is taken into consideration, the load shedding is not desirable, or if the phase adjustment equipment needs to be turned on (opened), the relay logic is changed. Need to change. In this case, the power system knowledge base 15 stores knowledge that load shedding is not desired or that requires inputting (opening) of the phase adjustment equipment. Protection Relay System Knowledge Base 1
1 stores the knowledge that the relay logic needs to be changed in such a case.

The logic-assisted inference engine 14 includes the power system knowledge base 15 and the protection relay system knowledge base 11
The relay logic 6 is deduced on the basis of the above, and if necessary, logic such as input (opening) of the phase adjustment equipment is added, and the relay logic 6 is changed.

FIG. 24 is an explanatory diagram of the neural network 20 used in the tenth embodiment.

In the figure, as the input layer 32, the supply-demand unbalance ratio 45, the load characteristic (V) 46, the load characteristic (F)
47, generator spinning reserve 48, frequency minimum 49, and frequency steady value 50. Load shedding amount 5 as output layer 34
1, the lowest voltage 52 and the highest voltage 53. Here, it is determined as follows. Supply-demand imbalance rate 45 = shortage of power generation to total demand /
Total power generation of the system Minimum voltage 52; Minimum voltage in the target system after frequency control by load shedding Maximum voltage 53; Maximum voltage in the target system after frequency control by load shedding

Next, an example of learning of the neural network 20 used in the tenth embodiment will be described.

First, in the target system, a method of simulating a sudden power generation shortage due to generator disconnection or interconnection line interruption is determined.
Then, supply-demand unbalance rate 45, load characteristic (V) 4
6. A simulation is performed using the load characteristics (F) 47, the generator's momentary reserve 48, and the load shedding amount 51 as parameters.
From this result, the minimum frequency 49, the steady-state frequency 50, the minimum voltage 52, and the maximum voltage 53 are obtained.

Further, the supply / demand imbalance rate 45, load characteristic (V) 46, load characteristic (F) 47, generator momentary reserve 48, load shedding amount 51, frequency minimum 49, frequency steady value 50, The voltage 52 and the maximum voltage 53 are recorded, and learning data for the neural network 20 is created.

The neural network 20 is trained based on the learning data thus created. After the learning is completed, the supply-demand unbalance ratio 45 of the target system, the load characteristic (V) 46, the load characteristic (F) 47, the instantaneous reserve 48 of the generator, the minimum frequency 49 to be finished, and the steady-state frequency 50
Is input, the required load shedding amount 51 can be obtained. At this time, since the minimum voltage 52 and the maximum voltage 53 are also obtained at the same time, if the voltage falls below the allowable minimum voltage determined in the operation of the power system or exceeds the allowable maximum voltage, the phase adjustment equipment is turned on (opened). . The power system knowledge base 15 is used to determine how many phase adjustment facilities are to be input (opened).
Further, using the protection relay system knowledge base 11, the operating conditions are determined in consideration of the setting (opening) of the phase adjustment equipment.

As described above, in FIG. 23, from the determination result 21 of the neural network, the logic support inference engine 14 determines whether the relay logic 6 is appropriate based on the power system knowledge base 15 and the protection relay system knowledge base 11. Then, if necessary, logic such as input (opening) of the phase adjustment equipment is added, and the relay logic 6 is changed.

As described above, the protection relay system according to the tenth embodiment of the present invention has the knowledge base of the power system, so that in addition to the effects of the protection relay system according to the eighth embodiment, Even if the power supply system changes, it is possible to quickly modify the logic of the protection relay system according to the change in the power system by using the knowledge of the protection relay system and power system experts and the neural network.

FIG. 25 is an explanatory diagram showing the processing of the protection relay system according to the eleventh embodiment of the present invention.

The protection relay system according to the eleventh embodiment of the present invention is different from the protection relay system of the ninth embodiment in that the power device knowledge base 1 such as a series capacitor is used.
6 is additionally provided, and has a function having a settling support inference engine 13 for changing the set value 7 of the protection relay system according to a change in the power system using the neural network 20.

First, for example, when a series capacitor is installed in the power system, the system configuration 37 such as the impedance behind and the installation point of the series capacitor and the capacitance of the series capacitor 5 are set.
4 changes the electrical distance (impedance) of the protection relay to the fault point. Also, when equipment protection is added to the series capacitor, the electrical distance of the protection relay up to the accident point varies depending on the operation status of equipment protection.

Assuming such a case, the knowledge of the change in the electrical distance seen by the protection relay up to the point of accident when the series capacitor is installed in the power system is stored in the power system knowledge base 1.
5 and the knowledge of device protection of the series capacitor is stored in the power device knowledge base 16. The settling support inference engine 13 makes an inference using the neural network 20 to determine whether the set value 7 is appropriate based on the power device knowledge base 16, the power system knowledge base 15, and the protection relay system knowledge base 11, and sets the appropriate value. Change the set value 7.

FIG. 26 is an explanatory diagram of the neural network 20 used in the eleventh embodiment.

In the figure, the input layer 32 has a capacity 54 of a series capacitor, a system configuration 37, an impedance (for R) 38 seen by a relay, and an impedance (for X) 39 seen by a relay. The operation determination 40 was set as the output layer 34.

When an accident occurs in the power system, the impedance (R) seen by the relay depends on the system configuration 37 such as the impedance behind and the installation point of the series capacitor and the capacity 54 of the series capacitor, even if the accident point and the appearance of the accident are the same. Minutes)
38, the impedance (for X) 39 seen by the relay is different. Therefore, when the set values of the relays are the same, there is a possibility that the relay malfunctions or malfunctions as described with reference to FIG.

Next, an example of learning of the neural network 20 used in the eleventh embodiment will be described.

The input layer 32 has a capacity of 5
4. System configuration 37, impedance seen by relay (R
), And the neural network 20 is made to learn the parameter by the impedance (X component) 39 seen by the relay. The output layer 34 is an operation determination 40.

First, the range of the fault point where the relay operates is determined. Next, the capacity 54 of the series capacitor and the system configuration 3
7. Perform a simulation using the accident point as a parameter. If the fault point is within the range of the fault point at which the relay operates, the operation judgment 40 is set to 1;
At this time, the impedance (for R) 38 seen by the relay and the impedance (for X) 39 seen by the relay are recorded. Thus, the simulation result is created, and the neural network 20 is learned based on the simulation result.

After the learning is completed, the neural network 20 stores a series capacitor capacity 54 and a system configuration 3 in a certain system state.
Using the value of 7, the operation of the impedance (for R) 38 and the impedance (for X) 39 seen by the relay are determined by the parameters.

Next, the operation range of the relay is determined as described with reference to FIGS. However, as described in FIG. 18, whether or not to settle in the new relay operation area 42 depends on coordination with other setting, coordination with other relays and coordination with equipment protection of the series capacitor. Knowledge of relay specialists and equipment protection of series capacitors is required.

Therefore, as described in the description of FIG. 25, the settling support inference engine 13 is a protection relay system knowledge base 1 in which knowledge of other settling and cooperation with other relays is stored.
1, Power equipment knowledge base 16, Power system knowledge base 1
5 based on the neural network 20 for inference,
Change the setting value 7 to an appropriate value.

As described above, the protection relay system according to the eleventh embodiment of the present invention has the knowledge base of the power equipment, so that in addition to the effect of the protection relay system according to the ninth embodiment, the protection relay system according to the ninth embodiment has the following advantages. Even if the state changes,
Using the knowledge of a protection relay system, a power system and power equipment experts and a neural network, it becomes possible to quickly correct the set value of the protection relay system in response to changes in the power system.

FIG. 27 is an explanatory diagram showing processing of the protection relay system according to the twelfth embodiment of the present invention.

The protection relay system according to the twelfth embodiment of the present invention is different from the protection relay system of the tenth embodiment in that it has a power device knowledge base 16 such as a series capacitor, It has the function of having a logic support inference engine 14 that changes the relay logic 6 of the protection relay system according to a change in the system.

For example, in the case where the power system has a system switching or the like and a series capacitor is present at a place where the electrical distance seen by the protection relay is short, it is necessary to change to a relay logic that considers the series capacitor. In this case, the power system knowledge base 15 stores the knowledge when a series capacitor exists at a place where the electric distance is short. If there is a series capacitor in the protection relay system knowledge base 11, the knowledge that the relay logic needs to be changed is stored.

The knowledge of equipment protection of the series capacitor is stored in the power equipment knowledge base 16. The logic-assisted inference engine 14 makes an inference using the neural network 20 based on the power system knowledge base 15 and the protection relay system knowledge base 11 to determine whether or not the relay logic 6 is appropriate. In addition,
Change the relay logic 6.

FIG. 28 is an explanatory diagram of the neural network 20 used in the twelfth embodiment.

In the figure, as input layer 32, system switching 55, series capacitor capacity 54, system configuration 37,
The impedance (for R) 38 seen by the relay and the impedance (for X) 39 seen by the relay were set. The operation determination 40 was set as the output layer 34.

As learning, a simulation is performed using the system switching 55, the series capacitor capacity 54, the system configuration 37, and the fault point as parameters, and the impedance seen by the relay (for R) 38 and the impedance seen by the relay (X
Min) 39 is calculated. The operation judgment 40 is 1 if the fault point is within the range of the fault point where the relay operates, and 0 if it is outside the range.
And Using the calculation results, as described in the examples of the seventh embodiment and the eighth embodiment, the range of the malfunction and malfunction of the relay is found, and the relay logic 6 is set so as to eliminate the malfunction and malfunction. change.

Further, in order to make changes, the power equipment knowledge base 1 having knowledge of equipment protection of series capacitors is required.
6. A protection relay system knowledge base 11 having knowledge of how to change logic, and a power system knowledge base 15 having knowledge of a case where a series capacitor exists at a place where the electrical distance is short are required.

As described with reference to FIG. 27, the logic support inference engine 14 makes an inference using the knowledge and the decision result 21 of the neural network, and if necessary, adds a logic in consideration of a series capacitor to form a relay logic. 6
To change.

As described above, the protection relay system according to the twelfth embodiment of the present invention has the knowledge base of the power equipment, so that in addition to the effects of the protection relay system according to the tenth embodiment, the protection relay system according to the twelfth embodiment has the following advantages. Even if the state changes, it is possible to quickly modify the logic of the protection relay system according to the change in the power system, using the knowledge of the protection relay system, the power system, and the experts of the power equipment and the neural network.

FIGS. 29 to 32 are explanatory diagrams of a storage medium for recording a processing program of the protection relay system according to the thirteenth embodiment of the present invention.

The invention described in the above-described first to twelfth embodiments can be implemented as a processing program that can be executed by a computer, for example, by using a storage device such as a magnetic disk, an optical disk, or a semiconductor memory. It is also possible to write and apply it to various devices, or to transmit it via a communication medium and apply it to various devices. The computer that realizes the present apparatus reads the program stored in the storage medium, and executes the above-described processing by controlling the operation of the program.

Storage Medium 1 of Protection Relay System of FIG. 29
FIG. 7 shows an example corresponding to the fifth embodiment, but is also applicable to the first embodiment and the third embodiment.

FIG. 30 is an explanatory diagram in which the storage medium 17 of the protection relay system shown in FIG.

The storage medium 17 of the protection relay system according to the thirteenth embodiment of the present invention shown in FIG. The storage medium 17 of the protection relay system inputs necessary information from the power system 101 and changes the set value 7 according to the function of the protection relay system according to the fifth embodiment of the present invention shown in FIG.

FIG. 31 shows the functions of the protection relay system according to the sixth embodiment of the present invention shown in FIG. 12 recorded in the storage medium 17 of the protection relay system.

FIG. 32 shows the storage medium 17 of the protection relay system shown in FIG. The storage medium 17 of the protection relay system inputs necessary information from the power system 101 and changes the relay logic 6 according to, for example, the function of the protection relay system according to the sixth embodiment of the present invention shown in FIG.

FIGS. 33 to 36 are explanatory diagrams of storage media for storing processing programs of the protection relay system according to the fourteenth embodiment of the present invention.

FIG. 33 shows the functions of the protection relay system according to the eleventh embodiment shown in FIG. 25 recorded in the storage medium 17 of the protection relay system. FIG.
In this example, the storage medium 17 of the protection relay system shown in FIG. The storage medium 17 of the protection relay system inputs necessary information from the power system 101 and changes the set value 7 according to the function of the protection relay system according to the eleventh embodiment of the present invention shown in FIG.

FIG. 35 records the functions of the protection relay system according to the twelfth embodiment of the present invention shown in FIG. 27 in the storage medium 17 of the protection relay system.

FIG. 36 shows that the storage medium 17 of the protection relay system of FIG. The storage medium 17 of the protection relay system inputs necessary information from the power system 101 and changes the relay logic 6 according to the function of the protection relay system according to the twelfth embodiment of the present invention shown in FIG.

[0170]

As described above, according to the first aspect of the present invention, the set value of the relay operation determination is inferred based on the power system information and the extensive and expert knowledge of the protection relay system, and a reasonable value is obtained. Since the set value changes quickly and accurately according to the situation of the power system, when an accident occurs in the power system, the reviewed set value can effectively eliminate the accident and prevent the spread of the accident.

According to the second aspect of the present invention, the relay logic for determining the relay operation is inferred based on the power system information and the extensive and expert knowledge of the protection relay system. Therefore, when an accident occurs in the power system, the revised relay logic can effectively eliminate the accident and prevent the propagation of the accident.

Further, according to the third aspect of the present invention, the set value can be set by utilizing a wide range of power system information and protection relay system, and a wide range of power system knowledge including power system characteristics. By making inferences, reasonable setpoints can be quickly and accurately changed according to the state of the power system.

According to the fourth aspect of the present invention, the relay logic is utilized by utilizing a wide range of power system information and protection relay systems, and a wide range of power system knowledge including power system characteristics. By making inferences, the appropriate relay logic can be quickly and accurately changed according to the state of the power system.

Further, according to the invention of claim 5, a wide range of power system information and protection relay system, and a wide range of power system knowledge including the knowledge of the power system and the characteristics of the power system and the power for power including the series capacitor are included. Knowledge of equipment is used to infer set values, and appropriate set values can be quickly and accurately changed according to the state of the power system.

Further, according to the invention of claim 6, a wide range of power system information and protection relay system, and a wide range of power system knowledge including the knowledge of the power system and the characteristics of the power system, and the power Knowledge of equipment is used to infer relay logic, and appropriate relay logic can be quickly and accurately changed according to the state of the power system.

According to the seventh aspect of the present invention, the power system information is input, the operation is determined by a neural network, and the obtained determination result, the power system information, and the expert knowledge about the protection relay system are utilized. Then, the validity of the set value is inferred, and the set value can be quickly and accurately changed according to the power system based on the inference result. Therefore, when an accident occurs in the power system, the accident can be effectively eliminated by the set value that has been reviewed by the change, and the propagation of the accident can be prevented.

According to the invention of claim 8, the power system information is input, the operation is determined by a neural network, and the obtained determination result, the power system information, and the expert knowledge about the protection relay system are utilized. Then, the validity of the relay logic is inferred, and the relay logic can be quickly and accurately changed according to the power system based on the inference result.
Therefore, when an accident occurs in the power system, the accident can be effectively eliminated by the relay logic reviewed by the change, and the propagation of the accident can be prevented.

According to the ninth aspect of the present invention, the power system information is input, the operation is determined by a neural network, the obtained determination result, the power system information, the expert knowledge of the protection relay system, and the power system information. The knowledge of the power system including the above characteristics is comprehensively used to infer the validity of the set value, and the set value can be quickly and accurately changed according to the power system based on the inference result.

According to the tenth aspect of the present invention, the power system information is input, the operation is determined by a neural network, the obtained determination result, the power system information, the expert knowledge of the protection relay system, and the power system The knowledge of the power system including the characteristics of (1) and (2) is comprehensively used to infer the validity of the relay logic, and the inference result allows the relay logic to be changed quickly and accurately according to the power system.

According to the eleventh aspect of the present invention, the power system information is input, the operation is determined by a neural network, the obtained determination result, the power system information, the expert knowledge of the protection relay system and the power system information are obtained. The knowledge base of the power system including the characteristics of the power system and the knowledge of the power equipment including the capacitor are comprehensively utilized to infer the validity of the set value, and the set value is quickly determined according to the power system based on the inference result.
Can be changed exactly.

According to the twelfth aspect of the present invention, the power system information is input, the operation is determined by a neural network, the obtained determination result, the power system information, the expert knowledge about the protection relay system and the power system The knowledge base of the power system including the characteristics of the power system and the knowledge of the power equipment including the capacitor are comprehensively used to infer the validity of the relay logic, and based on the inference result, the relay logic quickly performs according to the power system, It is changed exactly.

[Brief description of the drawings]

FIG. 1 is a conceptual configuration diagram of a protection relay system according to first to twelfth embodiments of the present invention.

FIG. 2 is an explanatory diagram of a process of the protection relay system according to the first embodiment of the present invention.

FIG. 3 is a conceptual diagram of a general knowledge database.

FIG. 4 is a configuration example of a protection relay system knowledge base of FIG. 2;

FIG. 5 is an explanatory diagram of a line A frame of FIG. 4;

FIG. 6 is an explanatory diagram showing a change in impedance seen by a relay due to the influence of a load current.

FIG. 7 is an explanatory diagram of a process of a protection relay system according to a second embodiment of the present invention.

FIG. 8 is an explanatory diagram of a line A protection frame for adding a zero-phase compensation logic according to the second embodiment of this invention.

FIG. 9 is an explanatory diagram of a process of a protection relay system according to a third embodiment of the present invention.

FIG. 10 is an explanatory diagram of a process of a protection relay system according to a fourth embodiment of the present invention.

FIG. 11 is an explanatory diagram of a process of a protection relay system according to a fifth embodiment of the present invention.

FIG. 12 is an explanatory diagram of a process of a protection relay system according to a sixth embodiment of the present invention.

FIG. 13 is an explanatory diagram of processing of a protection relay system according to a seventh embodiment of the present invention.

FIG. 14 is a configuration example of a neural network used in the present invention.

FIG. 15 is an explanatory diagram of a neural network used in a seventh embodiment of the present invention.

FIG. 16 is a first explanatory diagram showing an example of a relay operation determination using a neural network according to a seventh embodiment of the present invention.

FIG. 17 is a second explanatory diagram showing an example of the relay operation determination using the neural network according to the seventh embodiment of the present invention.

FIG. 18 is a third explanatory diagram showing an example of relay operation determination using a neural network according to the seventh embodiment of the present invention.

FIG. 19 is an explanatory diagram of a process of the protection relay system according to the eighth embodiment of the present invention.

FIG. 20 is a configuration example of a neural network used in an eighth embodiment of the present invention.

FIG. 21 is an explanatory diagram of processing of a protection relay system according to a ninth embodiment of the present invention.

FIG. 22 is a configuration example of a neural network used in a ninth embodiment of the present invention.

FIG. 23 is an explanatory diagram of processing of the protection relay system according to the tenth embodiment of the present invention.

FIG. 24 is a configuration example of a neural network used in the tenth embodiment of the present invention.

FIG. 25 is an explanatory diagram of processing of the protection relay system according to the eleventh embodiment of the present invention.

FIG. 26 is a configuration example of a neural network used in an eleventh embodiment of the present invention.

FIG. 27 is an explanatory diagram of processing of the protection relay system according to the twelfth embodiment of the present invention.

FIG. 28 is a configuration example of a neural network used in a twelfth embodiment of the present invention.

FIG. 29 is a first explanatory diagram of a storage medium of a protection relay system according to a thirteenth embodiment of the present invention.

FIG. 30 is a second explanatory diagram of the storage medium of the protection relay system according to the thirteenth embodiment of the present invention.

FIG. 31 is a third explanatory diagram of the storage medium of the protection relay system according to the thirteenth embodiment of the present invention.

FIG. 32 is a fourth explanatory diagram of the storage medium of the protection relay system according to the thirteenth embodiment of the present invention.

FIG. 33 is a first explanatory diagram of a storage medium of a protection relay system according to a fourteenth embodiment of the present invention.

FIG. 34 is a second explanatory diagram of the storage medium of the protection relay system according to the fourteenth embodiment of the present invention.

FIG. 35 is a third explanatory diagram of the storage medium of the protection relay system according to the fourteenth embodiment of the present invention.

FIG. 36 is a fourth explanatory diagram of the storage medium of the protection relay system according to the fourteenth embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Input conversion part 2 CPU 3 Program storage part 4 Data storage part 5 Relay output part 6 Relay logic 7 Setting value 8 Database 9 Knowledge base 10 Bus 11 Protection relay system knowledge base 12 Power system database 13 Setting support inference engine 14 Logic support inference Engine 15 Power system knowledge base 16 Power equipment knowledge base 17 Storage medium 20 Neural network

Claims (14)

[Claims] 1. A protection relay system for eliminating an accident and preventing the propagation of an accident when an accident occurs in a power system, wherein the knowledge of a specialist concerning the protection relay system is stored in advance as a knowledge base. Base, a power system database that captures the power system information that changes every moment of the power system and stores it in a database, power system information stored in the power system database, and a knowledge base stored in the protection relay system knowledge base. An inference engine that infers whether the set value for relay operation determination is appropriate based on the inference value, and changes the set value so as to make a relay operation determination based on a reasonable value obtained by the inference. Characterized protection relay system. 2. A protection relay system for eliminating an accident and preventing the propagation of an accident when an accident occurs in a power system, wherein the knowledge of an expert on the protection relay system is stored in advance as a knowledge base. Base, a power system database that captures the power system information that changes every moment of the power system and stores it in a database, power system information stored in the power system database, and a knowledge base stored in the protection relay system knowledge base. And an inference engine that changes the relay logic so as to determine whether the relay logic for determining the relay operation is appropriate based on A protection relay system characterized by the following. 3. A power system knowledge base for preliminarily storing knowledge of a power system including characteristics of the power system, wherein power system information stored in the power system database and knowledge stored in the protection relay system knowledge base. Based on the base and the knowledge base stored in the power system knowledge base, the inference engine makes an inference as to whether or not the set value for the relay operation determination is appropriate. The protection relay system according to claim 1, wherein the set value is changed so as to determine a relay operation. 4. A power system knowledge base for preliminarily storing knowledge of a power system including characteristics of the power system, wherein power system information stored in the power system database and knowledge stored in the protection relay system knowledge base. Based on the base and the knowledge base stored in the power system knowledge base, the inference engine makes an inference as to whether the relay logic for determining the relay operation is valid or not, according to the valid relay logic obtained by the inference. 3. The protection relay system according to claim 2, wherein a relay logic is changed so as to determine the relay operation. 5. A power device knowledge base for storing knowledge of power devices including series capacitors in advance, wherein power system information stored in the power system database and knowledge stored in the protection relay system knowledge base. Based on the base and the knowledge base stored in the power system knowledge base and the knowledge base stored in the power equipment knowledge base, the inference engine determines whether the set value for relay operation determination is appropriate or not. The protection relay system according to claim 3, wherein the setting value is changed so that the relay operation is determined based on a proper setting value obtained by the inference. 6. A power device knowledge base for preliminarily storing knowledge of power devices including a series capacitor, wherein power system information stored in the power system database and knowledge stored in the protection relay system knowledge base. Based on the base and the knowledge base stored in the power system knowledge base and the knowledge base stored in the power equipment knowledge base, the inference engine determines whether the relay logic for determining a relay operation is appropriate or not. The protection relay system according to claim 4, wherein the relay logic is changed so that the relay operation is determined based on an appropriate relay logic obtained by inference. 7. A protection relay system for eliminating an accident and preventing the propagation of an accident when an accident occurs in a power system, wherein the knowledge of an expert on the protection relay system is stored in advance as a knowledge base. Using the base, the power system database that captures and stores the power system information that changes with time, and the power system information stored in the power system database and necessary parameters, performs operation determination using a neural network and makes a determination. A neural network that outputs results, a relay operation determination based on the determination result output by the neural network, the knowledge base stored in the protection relay system knowledge base, and the power system information stored in the power system database. To determine whether the settling value for Protection relay system characterized by comprising an inference engine by thus obtained reasonable setpoint to the integer value changes to the decision relay operation. 8. A protection relay system for eliminating an accident and preventing the propagation of an accident when an accident occurs in a power system, wherein the knowledge of an expert on the protection relay system is stored as a knowledge base in advance. Base, a power system database that captures and stores the power system information that changes every moment of the power system, and using the power system information and necessary parameters stored in the power system database, performs operation determination using a neural network. A neural network that outputs a determination result, a relay operation determination based on the determination result output by the neural network, a knowledge base stored in the protection relay system knowledge base, and power system information stored in the power system database. To make sure that the relay logic for , Protection relay system characterized by comprising an inference engine which changes the relay logic to decision relay operation by appropriate relay logic obtained by inference. 9. A power system knowledge base for preliminarily storing knowledge of a power system including characteristics of the power system, a determination result output by the neural network, and power system information stored in the power system database. Based on the knowledge base stored in the protection relay system knowledge base and the knowledge base stored in the power system knowledge base, infer by the inference engine whether the set value for relay operation determination is appropriate or not. 8. The protection relay system according to claim 7, wherein the set value is changed so that the relay operation is determined based on a proper set value obtained by the inference. 10. A power system knowledge base for storing in advance a knowledge of a power system including characteristics of the power system, a determination result output by the neural network, and power system information stored in the power system database. Based on the knowledge base stored in the protection relay system knowledge base and the knowledge base stored in the power system knowledge base, infer by the inference engine whether relay logic for determining relay operation is appropriate or not. 9. The protection relay system according to claim 8, wherein the relay logic is changed so that the relay operation is determined based on the appropriate relay logic obtained by the inference. 11. A power device knowledge base for preliminarily storing knowledge of a power device including a series capacitor, a determination result output by the neural network, and power system information stored in the power system database. Determining a relay operation by the inference engine based on the knowledge base stored in the protection relay system knowledge base, the knowledge base stored in the power system knowledge base, and the knowledge base stored in the power equipment knowledge base. 10. The protection relay system according to claim 9, wherein inference is made as to whether or not the set value is appropriate, and the set value is changed so as to determine the relay operation based on the appropriate set value obtained by the inference. . 12. A power device knowledge base for preliminarily storing knowledge of a power device including a series capacitor, a determination result output by the neural network, and power system information stored in the power system database. Determining a relay operation by the inference engine based on the knowledge base stored in the protection relay system knowledge base, the knowledge base stored in the power system knowledge base, and the knowledge base stored in the power equipment knowledge base. 11. The protection relay system according to claim 10, wherein the relay logic is inferred whether or not the relay logic is appropriate, and the relay logic is changed so as to determine the relay operation based on the appropriate relay logic obtained by the inference. . 13. A storage medium storing a processing program of a protection relay system for eliminating an accident and preventing the propagation of an accident when an accident occurs in a power system, wherein knowledge of an expert on the protection relay system is previously obtained. A protection relay system knowledge base to be stored as a base, a power system database that captures and stores the power system information that changes every moment of the power system, a power system information stored in the power system database, and the protection relay system knowledge base. Based on the stored knowledge base, it is inferred whether the set value for relay operation determination or the relay logic is appropriate, and the relay operation is determined based on the appropriate set value or relay logic obtained by the inference. With an inference engine that changes the value or relay logic Storage medium storing a processing program for executing management. 14. A storage medium storing a processing program of a protection relay system for eliminating an accident and preventing the spread of the accident when an accident occurs in a power system, wherein knowledge of an expert on the protection relay system is previously obtained. Using a protection relay system knowledge base stored as a base, a power system database that captures and stores power system information that changes every moment of the power system, and an operation determination using a neural network using the power system information stored in the power system database And outputs a judgment result based on the judgment result output by the neural network, the knowledge base stored in the protection relay system knowledge base, and the power system information stored in the power system database. Set value or relay low for relay operation judgment Processing program for inferring whether the logic is appropriate or not and performing processing with an inference engine that changes the set value or the relay logic so as to determine the relay operation based on the appropriate set value or the relay logic obtained by the inference. Storage medium that stores.