JPH078110B2 - Power system operation device - Google Patents

Description

The present invention relates to a device for automating the operation of a power system.

There has been a method to automatically analyze, operate, and plan a power system using a digital computer, but it was a methodology based on analysis. The analytical methodology is
Several quantitative models are prepared in advance, and when the quantities of the power system match any of the quantitative models, numerical calculations and logical operations are performed using a computer to deal with various problems of the power system. This methodology has widespread application.

However, as the power system has become larger and more complex in recent years, it is not possible to clearly determine the quantitative model, or even if the model is clear, the solution space is wide and the search takes a long time. There are some cases where it is difficult to handle with scientific methods. In addition, in the conventional method, when there is a partial change in the system due to expansion or improvement of power equipment,
The entire program had to be reworked and checked to see that the professional Kurawa had been created without error, and this work was tremendous. Furthermore, after operating with the conventional method, what was executed by the computer and typed out by the typewriter shows the result of numerical calculation and logical operation, but what kind of problem occurs in the system Since it was difficult to read how the countermeasure was taken and it was difficult for the operator to understand, there was a demand for a solution method that was easy for the operator to understand.

A conventional power system operation device will be described below with reference to the drawings.
FIG. 1 shows an apparatus for automating the operation of a power system using a conventional computer. In the figure, (1) is composed of an electric power system or a part thereof, and (2) is composed of numerical data such as power flow, voltage, and current measured from the electric power system (1) and logical data indicating on / off of switchgear. A data bank, (3) is a program in which a procedure for performing arithmetic processing is written for each data in the data bank (2), and (4) is an arithmetic processing unit for sequentially executing the procedures of the program.
(5) is a conventional power system operation device composed of (2), (3), and (4). Of these, the data bank (2) and the program (3) are usually stored in the computer memory.

Next, the operation will be described. The arithmetic processing unit (4) interprets the program (3), and according to the procedure written in the program, appropriately performs the numerical operation and the logical operation while referring to the data bank (2), and the execution of the program (3) ends. Sometimes the final result is output to the power system (1) as an operating procedure of the power system. In the conventional device, the process of deriving the operation procedure of the electric power system such as the operation and control of the electric power equipment and the control of the electric power system is characterized in that the procedure is expressed in the form of a program of numerical operation and logical operation.

Now, in FIG. 2, A _O is the initial state of the power system, and B _O is the desired final state, and the initial state is set by controlling and operating.
Given that leads to final state B _O from A _O.

The conventional control-driven algorithm shown in FIG. 1 applies the system to one of the predetermined quantitative models based on appropriate assumptions at each branch point, and performs numerical calculation and logic. The calculation is performed and the branch point is moved to the next branch point according to the result, and it is repeated.However, if it is necessary to perform numerical calculation for all of the branch expansions and to check them, it would be a huge calculation. It will take time. For example, in the case of restoring a failed system,
When there are a large number of mash and breaks that can be thrown in for restoration, the calculation becomes enormous because it is necessary to calculate and confirm all combinations of those breaks and breaks.

Further, since the program (3) includes a procedure for deriving the operation procedure of the electric power system, when a part of the procedure for deriving the operation procedure is modified, the entire program (3) must be rewritten. The work of confirming the legitimacy of the subsequent program became enormous, and it took a lot of time and effort to change or modify the program (3) accompanying the expansion of electric power maintenance.

It is an object of the present invention to propose an electric power system operation device capable of guiding an operation procedure without performing an enormous calculation even when a target system is complicated and capable of easily responding to a change in a program. It is a thing.

In order to achieve this object, the present invention is based on the idea of treating a power system qualitatively, rather than fitting the power system to a model captured quantitatively. That is, the knowledge obtained from the actual operation experience in the power system is expressed in the form of a large number of rules to enable qualitative treatment, and since each rule is a unit of knowledge, the change of the program is specified. It is possible only by changing the rule of without affecting the whole.

FIG. 3 is a block diagram showing an embodiment of the power system operation device according to the present invention. Reference numeral (8) is an electric power system operation device including a storage device (6) and an inference mechanism (7). (61) is a data bank that stores data related to the state of the power system (1), and (62) is a generation rule part that represents knowledge obtained from operational experience as rules. ) And a storage device. This storage device is also called a knowledge base. A rule is expressed by dividing it into a premise part and a conclusion part that is derived if this premise part is established, and infers whether or not the premise part of the rule is established by checking the state of the system, that is, the data bank (61). A mechanism (7) is provided.

Hereinafter, the operation of the operation device of the power system shown in FIG. 3 will be described by taking the case of determining a recovery operation procedure for an accident of the power system as an example.

In the power system shown in Fig. 4, BUS is a bus, LINE is a transmission line, BANK is a transformer, white circles are open and broken, black circles are closed and broken, and white circles are permanently attached. It is a broken or broken piece that was opened due to a failure.

If a permanent failure occurs in BUS1, the area surrounded by the broken line is out of power, but the recovery procedure in this case is guided by the system macro flow shown in FIG. Following the start (step 101), the rules of the knowledge base (6) (6
2) is read into the inference mechanism (7) (step 102). This is because it is necessary to give the inference mechanism (7) what kind of rule there is because the rule is not incorporated in the program.

The rules (RULE1 to RULE5) used here can be classified into the following three according to the stage of inference.

(1) Judgment as to whether it is an element requiring restoration (2) Content of each restoration condition and configuration requirement (3) Selection of restoration method RULE1 belongs to (1) RULE2, RULE3 To (2), RULE4, RU
LE5 belongs to (3) respectively. As a basic procedure for restoration, RULE2 and RULE3 will be tried in numerical order. RULE4, RUL
The same applies to E5.

The specific rules are as follows.

Next, the inference mechanism (7) reads data from the data base (61) of the knowledge base (6) about power flow, voltage, current, system state such as switching state of switch and disconnector (step 10).
3). The inference mechanism (7) selects one element of the system (step 105). It is determined whether the element requires restoration (step 106). For example, the determination as to whether or not BUS2 requires recovery is performed as follows.

The inference mechanism (7) selects RULE1 as a rule corresponding to the first stage of inference.

The inference mechanism (7) collates the systematic data in the data bank (61) and determines whether the premise of RULE1 is true.
In this case, BUS2 is not charged and has not failed, so it can be concluded that BUS2 needs to be restored. Next, the inference mechanism (7) collates with the data bank (61) to confirm whether there is a power source capable of restoring BUS2 (step 10).
7). In this case, LINE 2,3,8 can be considered as a power source,
One of them, for example LINE2, is extracted (step 109) and BU
It is verified by RULE4,5,2,3 whether it is effective to restore S2 with LINE2 (step 110).

RULE4 and 5 relate to restoration conditions 1 and 2, respectively. The restoration conditions are, for example, as follows.

Restoration condition 1: The element that has lost power is restored from the power receiving end at all times.

Restoration condition 2: Power is restored from the standby circuit for the element that has lost power.

If restoration of BUS2 using LINE2 as the power supply meets either of restoration conditions 1 and 2, the restoration is effective. this
The reasoning mechanism (7) is verified using RULE2,3.

The inference mechanism (7) asks whether LINE2 satisfies the prerequisite of RULE2 and BUS2 satisfies the prerequisite of RLE2.
Judgment is made according to the initial state (103) of the grid. In this case BUS2
Does not satisfy the premise part, the inference mechanism (7) next uses RU
Select LE3. In this case, both assumptions are satisfied, so L
Restoring BUS2 with INE2 satisfies restoration condition 2 and can therefore be verified to be valid (step 110). As the next step, the inference mechanism (7) determines a restoration candidate of the system that may be restored (step 111). Since LINE4,5,6,7,18, BUS11 can be restored if BUS2 is restored, the portion surrounded by the broken line in FIG. 6 is a restoration candidate.

If all of these are restored, there is a risk of overload, so the power flow is calculated (step 112). The power flow calculation is performed as a subroutine in another place and is used as a known value. As a result of the power flow calculation, if it is found that an overload occurs (step 113), LINE5, which can be fed from other power sources among the restoration candidates, is preferentially disconnected and selected as a load (step 115). Under this circumstance, if the power flow is calculated again and the overload check is performed, the overload will no longer exist. In the above process, using LINE2 as a power source, BUS2, LINE4,
It can be seen that 6,7 can be restored. The range of systems that can be restored eventually becomes the range shown by the broken line in FIG.

BUS11 can be restored by the backup line LINE17.
It is verified by executing the flow shown in the figure. In this case, RULE1,5,8 is used. Based on the conclusion obtained in this way, the operator performs a specific operation for opening and closing the door and the circuit breaker.

It is not impossible to know the recovery procedure by verifying RULE2 and 3 by omitting RULE4 and 5, but when the operator deciphers the operation result typed out by the typewriter,
If RULE4 and 5 are omitted, it will be difficult to understand that the restoration was performed because the restoration conditions 1 and 2 were met.

LISP is a language used in a computer like FORTRAN. FORTRAN is suitable for numerical calculation and logical calculation, whereas LIPS is suitable for symbol processing.

FIG. 8 is a block diagram showing another embodiment of the present invention in which the inference mechanism is divided into an inference program (9) and a central processing unit (10). The inference program (9) is, for example, a program for promoting the steps shown in FIG. 5, and the central processing unit (10) collates the information in the data bank (61) to verify the rule (62). To do.

In the above, the present invention has been described by exemplifying the case where the restoration procedure for the power system can be restored without overload, but the present invention is not limited thereto, and the control of the power system,
It can be widely applied to general electric power system operation equipment such as diagnosis.

The present invention makes it possible to handle the system qualitatively based on the knowledge obtained from the experience of operating the power system. Therefore, the knowledge is expressed as a rule having a premise part and a conclusion part to be independent from the program, and the systematic data is collated to verify the rule. No numerical calculation is used in the rule verification process. Since the rules are independent of the program of the inference mechanism, when changing the system operation method, it is sufficient to change only a part of the rules and not the entire program.

As described above, the power system operation device according to the present invention expresses qualitative knowledge about the operation of the power system in a rule, verifies the rule while collating data indicating the state of the power system, and derives an operation procedure. As a result, even if the target power system is complicated, it is possible to guide the operation procedure without performing an enormous calculation, and it is possible to easily deal with the change of the program.

[Brief description of drawings]

FIG. 1 is a block diagram showing an example of a conventional power system operation device,
FIG. 2 is a diagram showing a relationship between an initial state and a desirable final state of the electric power system, FIG. 3 is a configuration diagram showing an embodiment of the electric power system operation device according to the present invention, and FIG. 4 is electric power of the present invention. A system diagram showing an example of a power system to be restored by the system operation device, Fig. 5 is a flow chart for restoration of the power system, Fig. 6 is a system diagram showing restoration candidates, and Fig. 7 is the restored system. FIG. 8 is a configuration diagram showing another embodiment of the power system operation device according to the present invention. In the figure, (1) is a power system, (6) is a storage device,
(7) is an inference mechanism, (9) is an inference program, and (10) is an arithmetic processing unit. The same reference numerals in each drawing indicate the same or corresponding parts.

Claims (2)

[Claims] 1. A rule in which qualitative knowledge about an operating procedure of a power system is divided into a premise section and a conclusion section which is derived if the premise section is true, and a memory storing data indicating the state of the power system. An electric power system operation device comprising: a device; and an inference mechanism that determines whether or not the precondition is satisfied based on the data. 2. An inference mechanism is provided with an inference program having an instruction relating to a rule selection order, and an arithmetic processing unit for determining whether or not a precondition of a rule is satisfied according to the inference program. An electric power system operation apparatus according to claim 1, wherein: