JPH051430B2 - - Google Patents

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to a device for controlling the position and shape of plasma in a torus-type nuclear fusion device,
In particular, the present invention relates to a plasma position and shape control device that provides good control characteristics that do not generate high-order multipole magnetic fields that disturb equilibrium.

[Prior art]

BACKGROUND ART Poloidal coils have conventionally been disposed in torus-type fusion devices, particularly in tokamak-type fusion devices, in order to control plasma current, plasma position, and cross-sectional shape. This poloidal coil has the function of generating an induced voltage in the plasma region and inducing a plasma current, as well as the function of generating the magnetic field necessary for the electromagnetic fluid balance of the plasma, that is, the position of the plasma and its cross-sectional shape. The magnitude and distribution of the magnetic field change depending on the electrical resistance of the plasma and changes in the internal state of the plasma. The magnetic fields required to control the plasma position and shape are multipolar magnetic fields such as a current transformer component magnetic field that determines the plasma current, a vertical component magnetic field that determines the plasma horizontal position, and a quadrupole component magnetic field that determines the ellipticity of the plasma cross-sectional shape. It is expressed as a superposition of components. Therefore, the poloidal coil must have the function of generating and independently controlling each of these multipole magnetic fields. Incidentally, the reality is that poloidal coils have conventionally been arranged according to function. As shown in Fig. 1, a coil (group) 31a serves as a current transformer coil for generating a current transformer component magnetic field for plasma current control, and a vertical magnetic field coil for generating a vertical component magnetic field for plasma horizontal position control. coil (group) 31b, quadrupole magnetic field coil (group) 3 for generating a quadrupole component magnetic field for ellipticity control in plasma cross-sectional shape;
1c, coils (groups) corresponding to control purposes are arranged around the plasma, and each coil is connected in series by a feeder 90 and driven by a control power source prepared for each control purpose. It used to be that

When poloidal coils are arranged in this way, a feedback control method as shown in Fig. 2 is an example of a control method for maintaining the equilibrium quantities such as plasma current, plasma position and its cross-sectional shape at target values. Are known. According to FIG. 2, the plasma current I _P of the plasma 62, the plasma horizontal position R _P and the ellipticity δ of the plasma cross-sectional shape are detected by the equilibrium quantity detectors 4a to 4c, respectively, and the subtractors 12a to 12c Differences ΔI _P , ΔR _P , and Δδ are calculated between these target values I ⁰ _P , R ⁰ _P , and δ ⁰ and input to the control calculation section 39 . The control calculation unit 39 calculates the differences ΔI _P , ΔR _P ,
Correction values of power supply voltage ΔEa, ΔEb, based on Δδ
ΔEc is determined, and these correction values ΔEa, ΔEb, ΔEc are used in adders 50a to 50c to set the power supply voltage set value E ⁰ _a to
By adding E ⁰ _c , the coils 31a to 3
Power supply voltage command values Ea to Ec for the control power supplies 32a to 31c corresponding to 1c are obtained. As the control calculation unit 39, a classical feedback control device has been used so far, as shown in FIG. 3, in which PID (proportional, integral, differential operation) controllers 39a to 39c are provided corresponding to the control power supplies 32a to 32c. Apart from this, optimal feedback control has also been attempted. To perform optimal feedback control, a multiplier 40 is required as shown in FIG.
In the multiplier 40, the correction values ΔEa to ΔEc are obtained by performing the calculation shown in the following equation (1) using K as an optimal feedback matrix.

(ΔEa, ΔEb, ΔEc) = K (ΔI _P , ΔR _P , Δδ) ...(1) The above-mentioned control is adopted for the poloidal coils arranged according to function in this way. In recent years, tokamak-type fusion devices have become mainstream in which poloidal coils are arranged in a hybrid manner. This is because the number of poloidal coils is small, and furthermore, it has the advantage of being easy to disassemble and assemble. As shown in Figure 5, in the hybrid system, coil 3
The feature is that control power supplies are provided for coils 3a to 3e, but in order to form the multipolar magnetic field required for plasma position and shape control, coils 3a to 3e are
e coils 3a to 3e so that the combined magnetic field formed by each of them has the required multipole component.
The current flowing through each coil is adjusted by a control power supply provided for each of the coils 3a to 3e.

However, this hybrid system still has unique control problems. This is because the hybrid system generates an unnecessary multipole magnetic field that has a negative effect on equilibrium quantities, which disturbs the equilibrium of the plasma. To explain the control in the hybrid system here, the number of poloidal coils and therefore the control power supply is set to l,
If the number of types of equilibrium quantities to be controlled is m, the required multipole magnetic field is m types, whereas l
A multipole magnetic field of the type is generated and subtracted (l−
m) There is a risk that the equilibrium may be disturbed by unnecessary multipole magnetic fields of the type. Normally, the value of l is determined based on the withstand voltage of the poloidal coil, and is from 5 to 20.
The value is as follows. Furthermore, m is 3 for a tokamak with a circular cross section, and 4 or 5 for a tokamak with a D-shaped cross section. Therefore, the value of l is usually larger than that of m, and (l-m) types of high-order multipole magnetic fields act to disturb the equilibrium. If the equilibrium is disturbed, the shape of the plasma will be distorted, causing problems such as the plasma coming into contact with the vacuum vessel, so in hybrid control, it is necessary to control unnecessary multipole magnetic fields as small as possible. In addition, Figures 1 and 5
Reference numerals 1 and 2 in the figure represent a vacuum vessel and a toroidal coil, respectively, and numeral 61 in Fig.
indicates a tokamak device.

[Purpose of the invention]

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a plasma position and shape control device that can suppress as much as possible the generation of unnecessary multipole magnetic fields that adversely affect equilibrium quantities related to plasma current, plasma position, cross-sectional shape, and the like.

[Summary of the invention]

For this purpose, the present invention determines the magnitude of the unnecessary multipole magnetic field from the coil current and the vacuum vessel current, and provides feedback to optimally control the multipole magnetic field and the equilibrium quantities. be.

[Embodiments of the invention]

The present invention will be explained below with reference to FIGS. 6 and 7.

First, the schematic configuration of the plasma position and shape control device according to the present invention will be explained with reference to FIG. In FIG. 6, the input vector E and state vector I of the tokamak device 61 and the output vector Y of the plasma 62 are as follows.

I ^T = (I _P , I ^T _c , I ^T _v ) ...(2) E ^T = (E ^T _c ) ...(3) Y ^T = (ΔI _P , ΔR _P , δ, ...) ... (4) However, the elements ΔI _P , ΔI _c , and ΔI _v represent the plasma current, coil current, and vacuum vessel current, respectively, and Ec represents the voltage of the control power supply. In this case, the vectors I, E, and Y are of the nth, lth, and mth orders, respectively, and a vector with T indicates that it is a transposed vector.

Now, in the tokamak device 61, the state equation shown in equation (5) holds true, and in the plasma 62, the output equation shown in equation (6) holds true.

I=AI+BE...(5) Y=CI...(6) However, A, B, and C are corresponding two-dimensional matrices or matrices.

Here, the l-order multipole magnetic field vector B is generated by multiplying I detected by the current detector 63 by a matrix H by the multiplier 6.
It is obtained by multiplying by .

B=HI …(7) I ^T =(B _j , B _v , …, B _n , B _n+1 …, B _l ) …(8) However, B _j , B _v , ..., B _n indicate the current transformer component, vertical component magnetic field, etc., and α _ij (i, j = 1 to l)
is a constant determined by the geometric dimensions of the coil.

On the other hand, depending on the equilibrium quantity detector 4, Y, that is,
Equilibrium quantities are detected, but in the subtracter 12, Y and Y
The difference ΔY from the target value Y ₀ is determined and input to the control calculation section 41. The control calculation unit 41 creates an l-order vector Z from the multipole magnetic field that disturbs the equilibrium in B and ΔY, and then determines the output Eopt of the optimal control feedback system based on the output feedback.
Of course, it is also possible to perform control without considering the difference ΔY, but the above control form is generally adopted.

Z ^T = (ΔY ^T , B _n+1 , ..., B _l ) ...(10) Eopt = -FZ ... (11) However, F is the matrix of feedback gain, and the target value of the equilibrium quantities This is obtained from the optimum feedback input for minimizing the deviation between the two and the formation of a multipole magnetic field that disturbs the equilibrium, that is, E that minimizes the evaluation function shown in equation (12).

J=∫ ^∞ ₀ (Z ^T PZ + E ^T RE) dt (12) However, P and R are diagonal weight matrices.

Here, we will briefly explain how to calculate the feedback gain matrix F in equation (11).
First, Z is expressed as follows using equations (6), (7), and (10).

Z=DI ……(13) However, D=C H ₂ ……(14) It is. Therefore, F can be easily obtained from the root K obtained by obtaining the root K from the Rikkatchi equation below.

KBR ^-1 B ^T K-KA-A ^T K - (D ^-1 ) ^T DD ^-1 = 0 ...... (16) F = R ^-1 B ^T KD ^-1 ... (17) In this way, control calculation The output Eopt of the feedback system obtained from the section 41 is added to the set value _E0 of the power supply voltage in an adder 50, thereby obtaining the input vector E.

FIG. 7 shows the configuration shown in FIG. 6 in more detail. In FIG. 6, the current detector 63 includes means for detecting the current flowing through each of the coils (group) and the vacuum vessel, but in FIG. 7, the vacuum vessel current is determined by calculation. That is, the current detector 63 in FIG.
Based on the plasma current I _P and the coil current I _c from the current detector 33, the vacuum vessel current calculating section 9 calculates the vacuum vessel current I _u . Although the vacuum vessel current calculation section 9 will not be explained in detail here, it is described in, for example, Japanese Patent Application Laid-Open No. 56-110087.
This is already publicly known from the publication No. Therefore, in the case of this embodiment, even if the vacuum vessel current cannot be directly measured, it can be determined by calculation, and control can be performed that also takes the vacuum vessel current into consideration.

〔Effect of the invention〕

As explained above, the present invention determines the magnitude of an unnecessary multipole magnetic field that disturbs the balance from the coil current, and provides feedback to optimally control the multipole magnetic field and the equilibrium quantities. It is something. Therefore, the present invention has the advantage that good control characteristics can be obtained in maintaining the plasma current, plasma position, cross-sectional shape, etc. at target values.

[Brief explanation of the drawing]

FIG. 1 is a partial cross-sectional view of a tokamak-type nuclear fusion device in which poloidal coils are arranged in a functional manner, FIG. 2 is a diagram showing the configuration of an example of a feedback control device for the fusion device, and FIG. 3
4 are diagrams showing the configuration of the control calculation unit in FIG. 2, FIG. 5 is a partial cross-sectional view of a tokamak-type fusion device in which poloidal coils are arranged in a hybrid manner, and FIG. , and FIG. 7 are diagrams showing a schematic configuration and a specific configuration of a plasma position and shape control device according to the present invention, together with a tokamak device, respectively. 3a to 3e... (hybrid system) toroidal coil, 4... equilibrium quantity detector, 6... multiplier, 41
...Control calculation unit, 8...Voltage detector, 9...Vacuum vessel current calculation unit, 12...Subtractor, 33, 63...Current detector, 50...Adder, 61...Tokamak device, 62...
plasma.

Claims (1)

[Claims] 1. A tokamak type in which a group of poloidal coils that generate a multipolar magnetic field for controlling plasma current, plasma position, and plasma cross-sectional shape are arranged in a hybrid manner so that each coil has a control power source. A plasma position and shape control device for a nuclear fusion device, including current detection means for detecting current flowing through a poloidal coil group and vacuum vessel, and equilibrium quantity detection means for detecting equilibrium quantities including plasma current, plasma position, and plasma cross-sectional shape. and a multipolar magnetic field calculating means for calculating the magnitude of the multipolar magnetic field based on the current value from the current detecting means, and calculating the difference between the equilibrium quantities from the equilibrium quantity detecting means and the target value of the various quantities. and a comparison means, and based on the difference from the comparison means and the magnitude of the multipole magnetic field that disturbs the equilibrium in the multipole magnetic field from the multipole magnetic field calculation means, reduce the magnitude of the multipole magnetic field that disturbs the balance and the difference. What is claimed is: 1. A plasma position and shape control device comprising at least control calculation means for calculating the optimum amount of feedback for each control power source and performing feedback control. 2. The current detection means includes a means for detecting the current flowing through the poloidal coil group, a means for detecting the current flowing in the plasma, a means for detecting the voltage of the poloidal coil, and a means for detecting the current flowing through the poloidal coil group, and a means for detecting the voltage of the poloidal coil. 2. A plasma position and shape control device according to claim 1, comprising means for calculating a current flowing in the plasma.