JPH0388007A - Automatic frequency control system - Google Patents

Abstract

PURPOSE:To reduce the control error of a control variable by converting a frequency deviation, a time difference and a total demand into fuzzy values and allowing each fuzzy input state to correspond to each control operation expression at the rate of 1 to 1. CONSTITUTION:For a system constant switching method in the automatic frequency control system, a fuzzy evaluation part 800 and an AR calculation integrating part 900 are used instead of conventional AR derivation control part and AR deriving part. The fuzzy evaluation part 800 consists of a frequency deviation fuzzy conversion part 811 for inputting a frequency deviation DELTAf outputted from a frequency deviation output part 200, evaluating the deviation DELTAf by means of a membership function mu(DELTAf) and outputting the evaluated result, a time difference fuzzy conversion part 812, a system capacity fuzzy conversion part 813, and an input status adaptation evaluating part 820. The AR calculation integrating part 900 is constituted of AR calculation parts 901 to 927 and an AR integrating part 930, inputs the frequency deviation DELTAf and system capacity Pa, executes calculation based upon an AR calculation expressing regulated by the decending part of a fuzzy control rule Rt and outputs the calculated result.

Description

[Detailed description of the invention] [Object of the invention] (Industrial application field) The present invention relates to an automatic frequency control method for controlling the frequency of a power system and interconnection line power to a reference value (prior art). If the load and supply capacity are always balanced, no frequency fluctuations will occur, but in reality, the load fluctuates from moment to moment, creating an imbalance between supply and demand, causing frequency and interconnection line power to fluctuate. Automatic frequency control (hereinafter referred to as AFC) maintains this fluctuating frequency and interconnection line power at their respective reference values.
).

Two control schemes are used, described below.

(1) Constant frequency control method (abbreviated as FFC method) A method in which the frequency deviation, which is the deviation from the reference frequency of the power grid, is detected and the generator is controlled to maintain only the frequency at the reference frequency regardless of the interconnection line power. It is. In this method, area requirement iAR (Area Requirement
nt) is determined by the following formula.

AR-= (Δf±a)xKx PaX0.1 --
-----(1) However, Δf is the frequency deviation [Hz],
α is the time difference correction constant [Hz], K is the systematic constant [%Hw
/ 0. IH2], pa is the system capacity [Hw].

(2) Frequency deviation interconnection line power control method (abbreviated as TBC method) The regional demand AR, which is the control amount within the own system, is determined from the frequency and the power deviation from the reference value of the interconnection line using the following formula, and this is This method controls the generator so that the

AR=-(Δf±a)xKxPa　xO,1-ΔP.

・・・・・・(2) However, Δpt is interconnection line power deviation [Hw].

Currently in Japan, power companies with large system capacity generally adopt the FFC method, while power companies with less capacity generally adopt the TBC method.

Conventional AFC will be explained using the FFC method as an example.

FIG. 16 shows the configuration of a conventional FFC type AFC.

In this figure, the part surrounded by the broken line is the AFC of the FFC method.
It is. In FIG. 16, 100 is a power system to be controlled, 110 is a generator that supplies power to the power system 100, and 200 is a frequency f and a reference frequency f of the power system 100.
, a frequency deviation deriving unit that derives the frequency deviation Δf (Δf=f−f,), which is the difference, 300 is a time difference deriving unit that detects the time difference Δt, which is the integral value of the frequency deviation Δf, and 400 is the power generation I! Generator power P which is the output of 1110 and power system 1
500 is an AR derivation unit for controlling the AR derivation unit, which will be described next, according to the frequency deviation Δf1 time difference Δt and the magnitude of the system capacity Pa. A control unit 600 derives iAR from frequency deviation Δf and system capacity pa using equation (1) under control from AR derivation control unit 500.
The R deriving section 700 is an AR distribution processing section for distributing the control amount AR, which is the output of the AR deriving section 600, to the generators.

The detailed configuration of the AR derivation control section 500 is shown in FIG.

In this figure, 510 is a dead zone processing section, 511 is a time difference correction live/kill section, and 512 is a system constant switching section. The dead band processing unit 510 inputs the frequency deviation Δf output from the frequency deviation derivation unit 200, and sets the control amount AR to zero when the absolute value of Δf is smaller than a predetermined value, and otherwise sets the control amount AR to zero.
The AR deriving unit 600 is controlled to calculate by equation (1). The time difference correction activation/destruction unit 511 inputs the time difference Δt output from the time difference derivation unit 300, and sets the time difference correction constant α in equation (1) to zero when the absolute value of the time difference Δt is smaller than a predetermined time (for example, 10 seconds).

If this is not the case, the time difference correction function is disabled by setting the frequency to non-zero (for example, 0.02 Hz). System constant switching unit 51
2 is the system capacity P output from the system capacity derivation unit 400
a is input, and the system constant K in equation (1) calculated by the AR derivation unit 600 is switched in a stepwise manner in multiple steps according to the magnitude of the system capacity Pa that changes from time to time. The switching characteristics of the system constant in the system constant switching unit 512 are as follows.
As shown in the figure. In this figure, the horizontal axis shows the system capacity Pa, and the vertical axis shows the system constant K. In this illustrated example, 0≦P B−<
When PaL, KN = 1.0; When PaL≦Pa×PaH, KM=1.5 When PaH≦Pa, KP=2
．． It is 0.

(Problems to be Solved by the Invention) In the conventional AFC described above, the system capacity Pa is set to the system constant switching threshold value PaL and P as shown in FIG.
Since the control amount AR is calculated using systematic constants that are approximated stepwise according to aH (usually within 10), there is a drawback that a considerable control error occurs. Also, conventional AFC
However, there is a drawback that it is difficult to determine the dead zone for the frequency deviation Δf, the time difference correction activation/destruction threshold, and the system constant switching threshold to arbitrary real values.

The present invention has been made in order to solve the above-mentioned drawbacks, and it is difficult for control errors to occur in the controlled variable AR, and it is possible to uniquely determine the frequency dead zone, the time difference correction activation/destruction threshold, and the system constant switching threshold. The purpose is to provide an automatic frequency control method that does not require

[Structure of the Invention] <Means for Solving the Problems> In order to achieve the above object, the present invention provides an automatic frequency control method that controls the frequency of a power system and interconnection line power to a reference value, with a frequency deviation of 1 time difference.

A frequency deviation fuzzification unit 9 converts each system capacity into a fuzzy quantity and evaluates the grade of these fuzzy quantities using a membership function; and one or all of a time difference fuzzification unit and a system capacity fuzzification unit; an input state suitability evaluation unit that evaluates the suitability of an antecedent part of a control rule that defines a fuzzy relationship with an AR calculation formula;
an AR calculation section that calculates the AR when each control rule is applied from the AR calculation formula defined in the consequent part of the control rule, the frequency deviation, the system capacity, and the interconnection line power deviation; and the input state compatibility evaluation section. and an AR integration section that calculates R from the output of the AR calculation section and the output of the AR calculation section to the control amount.

(Function) The frequency deviation fuzzification unit inputs the frequency deviation Δf and converts it into a fuzzy quantity: NB-Neqative Biq (larger on the negative side) l
O...Zero (approximate zero) PB=.
Fuzzy with Po5itive Big (large on the positive side), each fuzzy amount NB, 20. The PB grades μNB(Δf), μZo(Δf), μPB(Δf) are evaluated using the membership function on the frequency deviation Δf shown in FIG. 4 and are output. The time difference fuzzification section inputs the time difference Δt and converts it into a fuzzy quantity NB.

10. Fuzzy with PB, each fuzzy 1NF3.1
0°PB grade μHB (Δt), μZo (Δt〉,
The time difference Δ shown in FIG. 5 is evaluated and outputted using the above membership function.

The system capacity fuzzification unit takes the system capacity pa as input and normalizes it with the average system capacity Po, which is (Pa-Po)/P.
Convert it to a and further convert it into a fuzzy quantity NB, 20. P.B.
The grades of each fuzzy quantity NB, IO, PB are μHB(Pa), /-jZg(Pa).

μpB (Pa> is the normalized system capacity (Pa-
Po)/Evaluate using the membership function on Po and output.

The input state suitability evaluation unit uses fuzzy quantities NB and ZO.

Using PB's grade μMB'μZo and μPB as input, fuzzy quantity NB is calculated by fuzzy set operation.20. P
The degree of suitability Wt of the antecedent part of the fuzzy control rule that defines the relationship between B and the AR calculation formula is evaluated and output.

The AR calculation section receives the frequency deviation Δf and the system capacity pa as input, calculates and outputs the control amount ARt using the AR calculation formula defined by the consequent part of the fuzzy control rule R4 as shown in FIG. The AR integration section calculates the suitability W! of the output of the input state suitability evaluation section. and the AR calculation value AR of the output of the AR calculation section
The final control amount ARO is determined from t and output.

(Example) Examples will be described below with reference to the drawings.

FIG. 1 is an overall elliptical diagram for explaining the automatic frequency control method according to the present invention. In FIG. 1, the same reference numerals as in FIG. 16 are used with the same meanings as in FIG. 16, so their explanation will be omitted here, and only the different parts will be explained.

The R derivation control section 500 and AR derivation section 600 in FIG. 16 have been changed to a fuzzy evaluation section 800 and an AR calculation integration section 900, respectively, in FIG. Below, the fuzzy evaluation unit 80
0 and the AR calculation integration unit 900 will be explained using the diagram.

FIG. 2 is a side view of the configuration of the fuzzy evaluation section 800 of the automatic frequency control method of the present invention. In FIG. 2, 811 is a frequency deviation fuzzification unit, 812 is a time difference fuzzification unit, and 811 is a frequency deviation fuzzification unit;
13 is a system capacity fuzzification section, and 820 is an input state suitability evaluation section. The frequency deviation fuzzification unit 811 inputs the frequency deviation Δf output from the frequency deviation derivation unit 200, and converts this into a fuzzy quantity NB.

Fuzzy with ZO, Pa, each fuzzy quantity NB, IO
.

PB grade μ (Δf〉, μZo (Δf), μP
BB (Δf) is defined by the membership function μ(
Δf) is used to evaluate and output. Time difference fuzzification section 8
12 inputs the time difference Δt of the output of the time difference deriving unit 300, and uses this as the fuzzy quantity NB; 20. Fuzzy with PB, fuzzy amount NB, 20. PB grade μNB
(Δt), μ20(Δt〉, μ, 8(Δt〉) are evaluated and output using the membership function μ(Δt) shown in FIG. The system capacity Pa, which is the output of the
o) Convert to /Po and convert the converted value into fuzzy quantity NB, 10.
Fuzzy with PB, and the fuzzy quantity NB, 20.
PB grade μHB (Pa).

μ(Pa) and μpB(Pa) are evaluated and output using a membership function μ(Pa), which is not shown in FIG. 6, which will be described later. The input state suitability evaluation unit 820 evaluates the output grades μNB (Δf) and μ of the frequency deviation fuzzification unit 811.
(Δf), μ, 8〈Δf〉, and the grade μNB (Δt〉) of the output of the fuzzification unit 812 with the time difference O.

μ7o(Δt>, μPB(Δt), and the grade μ, 8(pa), μl of the output of the system capacitance fuzzification unit 813.

(Pa), μpB (Pa) as input, frequency deviation Δ
Each fuzzy quantity N8.20 for f1 time difference Δt, normalized system capacity (Pa-Po>/Po, etc.) Antecedent part of fuzzy control rule Rt shown in FIG. The degree of fitness Wt is evaluated and output by fuzzy set calculations described below.

FIG. 3 shows the configuration of the AR calculation integration unit 900 of the automatic frequency control method of the present invention. In FIG. 3, 901 to 927 are AR calculation units, and 930 is an AR integration unit.

Each AR calculation unit 901 to 927 includes a frequency deviation derivation unit 2
The frequency deviation Δf of the output of 00 and the system capacity pa of the output of the system capacity deriving unit 400 are respectively input, and the AR calculation formula specified by the consequent part of the fuzzy control rule Rt shown in FIG. Calculate and output the calculated value ARt.

The AR integration unit 930 calculates the final control amount AR, by using the fitness Wl of the output of the input state fitness evaluation unit 820 and the AR total X value of the output of each AR calculation unit 901 to 927. and outputs it to the AR distribution processing section 700. the above(
3) Formula is A calculated by each AR calculation unit 901 to 927.
This is a method in which the final control amount ARO is calculated as a weighted average based on the goodness of fit Wt of the R calculation value AR1.

FIG. 4 is a diagram showing an example of the membership function for the fuzzy quantity NB of the frequency deviation Δf, 20°PB.

In FIG. 4, the horizontal axis shows the frequency deviation Δf scaled 10 times, the vertical axis shows the fuzzy amount NB, and the grade μ(Δf) to which the frequency deviation Δf belongs to 20°PB.

FIG. 5 shows the fuzzy quantity NB of the time difference Δt, 10. P.B.
FIG. 2 is a diagram showing an example of a member Shira 1 function for .

In FIG. 5, the horizontal axis is the time difference Δt scaled to 1/10, and the vertical axis is the fuzzy amount NB. 20. The grade μ(Δt) of the time difference Δt to PB is shown.

Figure 6 shows the normalized system capacity (Pa-Po)/P.

Nofu, Ajii amount NB, 20. It is a figure which shows the example of the membership function with respect to Pa. In Figure 6, the horizontal axis is the normalized system capacity (Pa-P
O)/Po, the vertical axis shows the fuzzy quantity NB, the grade μ(Pa) to which the normalized system capacity (Pa Po)/Po belongs to 10°PB.

FIG. 7 is a table illustrating fuzzy m-rules of the FFC system according to the present invention. This fuzzy control rule is processed in the fuzzy evaluation section 800 and the AR calculation integration section 900 in FIG. In FIG. 7, the vertical columns are control rule numbers Rt (i = 1.2..., 27).

The horizontal rows indicate the antecedent and consequent parts of each control rule R4, and the antecedent part contains the fuzzy quantity NB, 20. The state of the normalized system capacity (Pa Pa )/Po using PB, the time difference Δt1 and the frequency deviation Δf is shown, and the consequent part shows the AR calculation formula (non-fuzzy) corresponding to the antecedent part.

For example, the seventh control rule R7 in the table is R7: i
f (Pa Po )/Po iS NB andΔ
tisPa and Δf is NB.

then AR7= −(Δf+α)XKHxPaxo
, means 1.

FIG. 8 is a table for explaining an example of frequency control response using the automatic frequency control method of the present invention.

In Figure 8, the vertical columns are the same as in Figure 7, the control rule number J, and the horizontal rows are the grade values μ of the antecedent part of each control rule R4.
(Pa), μ(Δt), μ(Δf), the goodness of fit Wt, and the AR calculated value ARt of the consequent part. In FIG. 8, frequency deviation Δf2 time difference Δt.

If the value from normalized system capacity (Pa Po )/Po is as follows, Δf 1=-o, ex 10-' [Hz] Δt, =
0.7x10 [S] (Pa1-Po)
/P□ =-0,7x Each control rule R for 1/4
A response value of 4 is shown.

The grade value μ(A
f1) Grade value μ (Af1 ) of 1 time difference Δt, grade value μ (P
It can be seen from FIGS. 4, 5, and 6 that at) has the following values as shown by broken lines.

μNB(Δf, ) −0,5(NB)/−(70(
Δf, ) −0,35(ZO)μPB Δf 1
) −〇 (PB) μNB (Af1) = 0
(NB)μZo(Δtt) =0.2 (ZO
)μPB Δt 1 ) =0.5 (
PB)μNB (ΔP,,) =0.75 (
NB)μZo(ΔPa1) =0.3 < 20)μ
Pa (Δ Pax) = 0 (PB) The parenthesis after the numerical value on the right side of the above equation is the fuzzy quantity NB represented by that numerical value. 20. PB is added to clearly indicate it on Figure 8. The above value is used as the fuzzy amount NB of the antecedent part for each control rule R1 in FIG. 7, 20. The grade value μ(Pa1
), μ(Δatt), cz(Af1).

The goodness of fit of the antecedent part in Figure 8 W! , is calculated using the following formula; % formula %() ()) (4) as an operator regarding the product (AND) fuzzy set. However, n1n() means selecting the minimum value within ().For example, the fitness W7 of the antecedent part of the seventh control rule R7 in FIG. 8 is W, -IIin (μNB (Pat),
, czPB (Af1 )μNB(Af1 )) = nin (0,75,0,5, (1,5)=0.5.

The AR calculation value AR4 of the consequent part in FIG.
It is obtained by substituting 1-0.06 [H2] and Af1 [S]. At that time, time difference correction constant α = 0.02 [Hz]
.

Three systematic constants KN = 1.0 [%Hw10.1H
2], K M-1,5 [%Hw10.1Hz],
Kp =2.0 [%Hw10.1Hz]. For example, the AR calculation value AR of the seventh control rule R7 in FIG. 8 is: 8R7= (Af 1 cr) XKN X
P61X0.1=-(-0,006+0.02)x
i, o X P at X 0.1=0.00
4 XPa. Furthermore, the AR calculation value A of the consequent part in Figure 8
In R4, where the degree of fitness of the antecedent part Wt=O, the notation of the calculated value is omitted. This is the final control amount AR,
As can be seen from equation (3), the AR calculation value ARC corresponding to the fitness Wt = 0 of the antecedent part is the final control J
This is because it does not affect IAR.

In the AR integration unit 930, the final control amount 0 calculated using equation (3) is calculated based on the goodness of fit Wl and AR shown in FIG.
It can be obtained as follows using the calculated value AR1.

1 1 As explained above, according to the present embodiment, it is possible to provide an automatic frequency control method that does not require unique determination of the dead zone and time difference correction activation/destruction threshold and the system constant switching threshold.

FIG. 9 is a block diagram of another embodiment according to the present invention. This embodiment shows TBC type automatic frequency control. In FIG. 9, the same reference numerals as those in FIG. 1 are used with the same meanings as in FIG. 1, so their explanation will be omitted below and only the differences will be explained.

950 is a TBC method AR calculation integration unit, and 1000 is an interconnection line power deviation derivation unit. Interconnection line power deviation derivation unit 100
0 is the interconnection line power deviation ΔPf, which is the difference between the interconnection line power pt of the power system 100 and the corresponding interconnection line power reference value Pf,o set in advance, ΔP,=P.

It is derived using the formula Pro and output to the TBC method AR calculation integration section 950. FIG. 10 shows the configuration of the TBC type AR calculation integration section 950. In FIG. 10, reference numerals 951 to 977 are TBC type AR calculation units, which connect the frequency deviation Δf of the output of the frequency deviation derivation unit 200, the time difference Δt of the time difference derivation unit 300, and the output of the interconnection line power deviation derivation unit 1000. The system line power deviation ΔP6 is inputted, and the R calculation value ARt is calculated by the AR calculation formula defined in the consequent part of the fuzzy control rule R4 shown in FIG. 11, which will be described later. Output.

FIG. 11 is a table illustrating fuzzy control rules for the TBC method according to the present invention. This control rule is determined by the fuzzy evaluation unit 8
00, is processed in the TBC method AR calculation units 951 to 977 and the AR integration unit 930. In Figure 11, only the AR calculation formula specified in the consequent part of Figure 7 is TB
It has been modified for C-method AR calculation. For example, the seventh control rule R7 in the table is: R7: 1f(Pa-P□) /P6 is NB
andΔt is PB andΔfisNB.

then AR7=-(Δf+a) KN X P
It means B-x0.1-ΔP6.

FIG. 12 is a block diagram of another embodiment according to the present invention. In this embodiment, the present invention is applied to the system constant switching unit 512 of the conventional automatic frequency control system of the FFC system shown in FIGS. 16 and 17. In FIG. 12, the same reference numerals as in FIG. 16, FIG. 17, and FIG. 1 are used with the same meanings as in those figures, so the explanation thereof will be omitted below, and only the differences will be explained.

830 is a system capacity fuzzy evaluation unit. The system capacity fuzzy evaluation unit 830 receives the system capacity Pa output from the system capacity derivation unit 400 as an input, and converts it into a fuzzy quantity: VS·Ve
ry 5Illall (very small) 5-8l'1a
ll (small) M...Hidiu Koh (
Medium) B...Big (Large) VB-Very Bio (Very Big) Fuzzy and each fuzzy amount VS, S, M, B.

VB grade μy6 (Pa), μ3 (Pa
), 1iH(Pa > 1.L(B(Pa)
, μyB(Pa) using a membership function μ(Pa) illustrated in FIG.

FIG. 13 shows the fuzzy quantities VS, S, and M of the system capacity Pa.

FIG. 3 is a diagram illustrating membership functions for B and VB. In FIG. 13, the horizontal axis is the system capacity Pa.

The vertical axis shows the grade μ (Pa) of the affiliation of the system capacity pa to the fuzzy quantities VS, S, M, B, and VB.

FIG. 14 is a table illustrating the fuzzy control rules of the present invention. This control rule is processed in the time difference correction activation/destruction unit 511, system capacity fuzzy evaluation unit 830, and AR calculation integration unit 900 in FIG. In Figure 14,
The vertical column is the control rule number Rt (t=1゜2, . . .
10) The horizontal rows show the first part and the second part of each control rule Rt, and in the first part, the non-fuzzy time differences At and VS, S,
Five labels M, B, and VB indicate the state of the fuzzy system capacity pa, and the consequent part shows the AR calculation formula corresponding to the antecedent part. For example, the third control rule R3 in the table is, if T (T>0) is the time difference correction threshold, then R3: ifΔt<-T and Pa1s M
.

then AR3= (Δf-(Z)XK3XPaX
It means 0.1. The constant Kt (
t = 1, 2, -, 5) is the state of system capacity pa (V
S, S, M, B, VB) corresponding system capacity. The final control amount ARO of this embodiment is based on the following conditions: in the antecedent part of the control rule shown in FIG. .

μ (Pa), μ (Pa), μB (Pa).

Since there are only two adjacent S N μyB(Pa) values that are non-zero at the same time, it can be calculated as follows: (t = 2.3..., 5).

The above formula (5) uses the AR calculation formula in the case of the time difference Δ1<-T of the consequent part in Figure 14, and furthermore, the fitness of the antecedent part Wt=
If we use the relationship μt(Pa), we can write it as 1 1 (the following is a blank space). However, μi, (Pa) (t=1
, 2.

・, 5) is 77 ji quantity VS, S, M, B, VB
G: Corresponding grade value. If the time difference Δt>T, the sign of α in the above equation (5) may be changed. () in the above equation (6) is the grade μt of the fuzzy quantity of the system capacity pa
It is a weighted average of two adjacent systematic constants KL with (Pa) as the weight.

FIG. 15 shows the weighted average systematic constant K.

The broken line shown by the bold line in FIG. 15 is a graph representing changes in the weighted average systematic constant. Therefore, equation (6) is the 15th
Since the continuously changing system constant K shown in the figure is used, a controlled variable AR with a small control error is provided.

As described above, according to the present embodiment, it is possible to provide an automatic frequency control method in which a control error does not occur in the controlled variable RO, and the system constant switching threshold value is not uniquely determined.

In each of the above embodiments, the fitness W4 of the antecedent part of the fuzzy control rule was explained using the l1in operation (formula (4)) of the product fuzzy set, but the invention is not limited to this. It goes without saying that the present invention can be applied even if other operations on the fuzzy set, such as multiplication of the grade values of the fuzzy set, are used instead. Furthermore, in the above embodiments, the shapes of the membership functions are triangular trapezoid, S shape, and Z shape, but the shape is not limited to these, and other membership functions such as bell shape may also be used. It goes without saying that the present invention is valid. Further, in the above embodiment, the simple weighted average calculation expressed by equation (3) or equation (5) is used to calculate the control amount AR, but the present invention is not limited to this. It goes without saying that other defuzzification methods can also be used, such as the square weighted average calculation method or the maximum fitness method that selects the AR calculation value corresponding to the maximum fitness. Although three or five fuzzy quantities are used when fuzzifying frequency deviation, etc., the present invention is not limited to this, and it goes without saying that it can also be applied to other fuzzy quantities such as seven or nine. .

[Effects of the Invention] As explained above, according to the present invention, the frequency deviation 1 time difference and the total demand are fuzzified, and the control calculation formula is made to correspond 1:1 to the state of these fuzzy input quantities. Therefore, it is possible to provide an automatic frequency control method that does not cause control errors in the control amount and does not require unique determination of the frequency dead zone and time difference correction activation/destruction threshold and system constant switching threshold. .

[Brief explanation of drawings]

FIG. 1 is a block diagram of an embodiment of the frequency control system according to the present invention, FIG. 2 is a block diagram of a fuzzy evaluation section of the automatic frequency control system according to the present invention, and FIG. 3 is a block diagram of an embodiment of the automatic frequency control system according to the present invention. A configuration diagram of the AR calculation integration part of the control system, Figure 4 shows the membership function of frequency deviation, Figure 5 shows the membership function of time difference, Figure 6 shows the membership function of normalized system capacity, and Figure 7 8 is a diagram showing a control rule of the FFC method of the present invention, FIG. 8 is a diagram for explaining a response example of frequency control by the automatic frequency control method of the FFC method of the present invention, and FIG.
FIG. 10 is a block diagram showing another embodiment of the TBC automatic frequency control method according to the present invention, FIG. FIG. 11 is a diagram showing the control rule of the TBC method according to another embodiment of the present invention, FIG. 12 is a block diagram showing another embodiment of the automatic frequency control method of the FFC method according to the present invention, and FIG. The membership function J number for the system capacity shown in Figure 12, Figure 14 is a diagram showing the control rule of the FFC method of another embodiment of the present invention, Figure 15 is a diagram showing changes in the weighted average system constant, Fig. 16 is a block diagram of a conventional automatic frequency control system, Fig. 17 is a block diagram of an AR derivation control section of a conventional automatic frequency control system, and Fig. 18 explains a method of switching system constants in a conventional automatic frequency control system. This is a diagram for 100... Power system 200... Frequency deviation derivation unit 300... Time difference derivation unit 400... System capacity derivation unit 500... AR derivation control unit 600... AR
R distribution processing unit to derivation unit 700 800 Fuzzy evaluation unit 811 Frequency deviation fuzzification unit 812 Time difference fuzzification unit 813 System capacity fuzzification unit 820 Input state Compatibility evaluation section 830...System capacity fuzzy evaluation section 900...AR calculation integration section 911-927...AR calculation section 930...AR integration section 950...TBC method AR calculation integration section 951
~977...TBC method AR calculation section 1000...Interconnection line power deviation derivation section Patent applicant: Toshiba Corporation

Claims (1)

[Claims] In automatic frequency control methods that control power system frequency and interconnection line power to standard values, frequency deviation fuzzy converts each frequency deviation, time difference, and system capacity into fuzzy quantities and evaluates the grade of these fuzzy quantities using a membership function. an input state for evaluating the degree of suitability of the antecedent part of the control rule that defines the fuzzy relationship between the fuzzy quantity, the time difference fuzzification part, and the system capacity fuzzification part, or all of them, and the fuzzy quantity and the AR calculation formula; a suitability evaluation section; an AR calculation section that calculates the AR when each control rule is applied from the AR calculation formula defined in the consequent part of the control rule, the frequency deviation, the system capacity, and the interconnection line power deviation; An automatic frequency control system comprising: an AR integration section that calculates a control amount AR from an output of an input state suitability evaluation section and an output of the AR calculation section.