Tokamak magnet power supply capacity optimization method and system for improving grey prediction

Abstract

The invention relates to the technical field of Tokamak magnet power supplies, and provides a Tokamak magnet power supply capacity optimization method for improving grey prediction, which comprises the following steps: transforming the weight of an original sequence of a grey prediction model; a background value is constructed by using a complex Simpson formula, and the complex Simpson formula is simplified; utilizing known plasma displacement information to verify the predicted displacement, and accurately predicting a displacement signal of the next period through verifying the domesticated gray prediction model; the plasma displacement is predicted in advance to obtain a magnet power supply reference signal in advance, and the magnet power supply only needs to output small current when the plasma generates small displacement offset, so that plasma displacement excitation balance control can be realized, and the purpose of reducing the capacity to optimize the capacity of the magnet power supply is achieved; the problem that the investment practicability of the fusion device is influenced due to the continuous and great increase of the power capacity demand of the Tokamak magnet is solved.

Description

Technical Field

[0001] The invention relates to the technical field of tokamak magnet power supply operation, and in particular to a tokamak magnet power supply capacity optimization method with improved grey prediction. Background Art

[0002] The Tokamak device is an important device for achieving controlled nuclear fusion, in which the displacement control of plasma is directly related to whether the controlled nuclear fusion can be successfully carried out. The Tokamak magnet power supply can quickly establish the corresponding magnetic field according to the real-time displacement signal of the plasma, realize the magnetic confinement control of the plasma displacement, and achieve the purpose of feedback balance control of the plasma displacement. The design of the Tokamak magnet power supply is based on the capacity required for the maximum displacement balance control of the plasma. The plasma control experts estimate the capacity required for the magnet power supply, and the magnet power supply designers design the magnet power supply of the corresponding capacity according to the estimated capacity requirements to excite the load coil and achieve the purpose of plasma magnetic confinement closed-loop control.

[0003] Tokamak devices involve a wide range of disciplines. The operating effects of magnet power supplies such as poloidal field power supplies and fast control power supplies depend on the power supply response speed and the ability to output the maximum current. In order to achieve balanced control of plasma displacement under large displacement offsets, high requirements are placed on the capacity and response performance of magnet power supplies. In the doctoral thesis "Simulation and Experimental Research on Passive Stability and Active Control of EAST Vertical Displacement" by Liu Lei of the University of Science and Technology of China in May 2015, it was introduced that when the Tokamak magnet power supply project team estimated and designed the magnet power supply capacity, the designers established a power supply model based on the response delay and current ramp delay of the power supply, set the maximum acceptable initial displacement of plasma, and estimated the relevant magnet power supply capacity to achieve robust control. With the development of controlled nuclear fusion engineering, the demand for the increase of tokamak magnet power supply capacity will continue to increase. Considering that the controlled fusion device will eventually benefit mankind in the form of grid-connected power generation, the current demand for the continued substantial growth of magnet power supply capacity will inevitably seriously affect the progress of the practical application of the fusion device. Optimizing the plasma control system and improving its performance will help reduce the unlimited growth demand for magnet power capacity and maximize the commercial value of controlled nuclear fusion.

[0004] The control target of some magnet power supplies represented by fast control power supplies is to quickly track the reference signal given by the plasma control system (PCS), output the corresponding current to excite the load coil, and realize the magnetic confinement control of the vertical displacement of the plasma. The shorter the output current and the time to establish the voltage on the load coil, the faster the plasma displacement balance control can be achieved. Furthermore, the faster the output current or the smaller the plasma displacement offset, the smaller the target current is needed to control the plasma to the equilibrium position. Therefore, fast control of the output current of the magnet power supply, or predictive control of the plasma displacement, is an important way to reduce the required current to achieve plasma displacement balance control, and is also an important method to optimize the capacity of the tokamak magnet power supply. The magnet power supply adopts a digital control structure, improves the dynamic response speed of the output current, and reduces the digital control delay and the inherent delay of the power supply system. These are important methods to improve the power supply control performance. Among them, the inherent delay of the system can be solved by predicting the plasma displacement in advance and obtaining the reference signal in advance, so that the magnet power supply can output a small current quickly in advance. The plasma displacement is obtained by prediction method and then the reference current is obtained in advance, so that when a small displacement deviation occurs, a small current is output to pull the plasma back to the equilibrium position. This is an important way to reduce the magnet power capacity to achieve the purpose of optimizing the magnet power capacity and achieve stable control of plasma displacement balance.

[0005] Grey GM (1,1) prediction can achieve the prediction of the target quantity without the need for an accurate mathematical model of the prediction target. The prediction process is simple, fast and easy to calculate. In the prior art, the Chinese invention patent "A gas concentration prediction method based on an improved grey-long short-term memory neural network combined model" with the publication number CN113190535A discloses an improved grey GM (1,1) model prediction method, which uses the ensemble learning method to combine the improved grey prediction model with the long short-term memory neural network to establish an improved grey-long short-term memory neural network combined prediction model to achieve accurate prediction of gas concentration. The paper "Application of improved grey model with initial value correction and function transformation in pipeline corrosion depth prediction" published in the Journal of Safety and Environment in June 2023 improves the initial value of grey prediction, increases the weight of new information, and achieves accurate prediction of pipeline corrosion depth. The paper "Research on grey settlement prediction model based on background value optimization and improvement" published in the Journal of Surveying, Mapping and Spatial Geographic Information in December 2022 uses the Cortes trapezoidal formula to reconstruct the background value, realizes the optimization of the grey GM (1,1) model, and improves the prediction accuracy of underground settlement.

[0006] The above patent applications and papers have modified the grey GM (1,1) prediction model, improved the grey GM (1,1) model or reconstructed the original sequence and background values, but the reconstruction function or improved algorithm used is too complex, which is not conducive to the realization of digital control of tokamak magnet power supply, and brings a heavy burden to the digital processor. In addition, in the process of grey prediction of target quantity, the original data is not used for repeated prediction, verification and correction. The accuracy of target quantity prediction needs to be further improved, and the optimization of magnet power supply capacity and plasma equilibrium displacement control are poor. Summary of the invention

[0007] The technical problem to be solved by the present invention is to obtain the displacement deviation in advance by accurately predicting the plasma displacement, thereby reducing the power supply capacity of the tokamak magnet.

[0008] The present invention solves the above technical problems through the following technical means:

[0009] The present invention provides a method for optimizing the capacity of a tokamak magnet power supply by improving grey prediction, comprising the following steps:

[0010] S1. Collect the plasma displacement data in the last four switching cycles given by PCS to form the original sequence X of the grey GM (1,1) prediction model (0) , the original sequence expression is

[0011] X (0) =[X (0) (1),X (0) (2),X (0) (3),X (0) (4)] (1)

[0012] And transform the weight of each information in the original sequence;

[0013] S2, based on the original sequence X (0) , generate the cumulative sequence X (1) (k), the cumulative sequence expression is

[0014]

[0015] Then use the simplified Simpson formula to reconstruct the background value Z (1) (k);

[0016] S3, according to the background value Z (1) (k) and the original displacement sampling data, calculate the gray development coefficient a and the gray amount b; according to the gray development coefficient a, the gray amount b and the original sequence X (0) , obtain the prediction formula of the original displacement signal;

[0017] S4. Prediction based on the prediction formula The value of and prediction error check; use the prediction formula that meets the error check to predict the displacement output value of the next cycle

[0018] S5. Predicted displacement The plasma displacement deviation is obtained by comparison with the reference displacement, and the transfer function is calculated according to the displacement deviation and the reference signal to obtain the reference signal.

[0019] Furthermore, the weights of each information in the original sequence described in step S1 are transformed. The specific transformation expression is:

[0020]

[0021] Furthermore, the background value is reconstructed using the simplified Simpson formula in step S2, specifically:

[0022] In the complex Simpson formula, the k~k+1 sampling period is divided into m intervals, and the cumulative sequence curve expression is set to f(x i ), among which

[0023]

[0024] The expression for solving the background value by complexing Simpson's formula to approximate the integral process is:

[0025]

[0026] Accumulate the sequence X within k~k+1 sampling cycles (1) (k) to X (1) The (k+1) change value is divided into m equal parts, then the cumulative sequence change value in the m intervals is

[0027]

[0028] X in the cumulative sequence (1) (k) and X (1) The value of (k+1) is known, so let

[0029] f(x i )=X (1) (k)+iΔ,i=0,1,…,m-1 (14)

[0030]

[0031] Combining formulas (12) to (15) to simplify the complex Simpson formula, the reconstructed background value is

[0032]

[0033] Furthermore, the step S3 described in accordance with the background value Z (1) (k) and the original displacement sampling data, calculate the gray development coefficient a and gray amount b, the specific calculation method is:

[0034]

[0035] X n =[X (0) (2)X (0) (3)X (0) (4)] T (7)

[0036] Furthermore, the prediction formula of the original displacement signal in step S3 is specifically:

[0037]

[0038] Furthermore, the prediction formula satisfying the error check is used in step S4 to predict the displacement output value of the next cycle. Specifically:

[0039] Assume that the acceptable prediction error is e, and use the error check formula to determine whether the predicted value meets the prediction error. The error check formula is:

[0040]

[0041] When the error check is satisfied, the next displacement output value is predicted according to the prediction formula

[0042] Otherwise, update the weight of the new information in the original sequence, recalculate the gray development coefficient a and gray quantity b, and get the new prediction value according to the prediction formula. And recheck until the error check is satisfied.

[0043] Furthermore, the reference signal calculation transfer function described in step S5 is:

[0044]

[0045] Among them, K p , K i , K d The coefficient of the transfer function is calculated for the reference signal, s is the Laplace operator; the input of the transfer function is the obtained plasma displacement deviation, and the output is the reference signal given to the tokamak magnet power supply.

[0046] The present invention also provides a tokamak magnet power supply capacity optimization system with improved grey prediction, which adopts the above method when running, and includes the following modules:

[0047] The acquisition module is used to collect the plasma displacement data in the last four switching cycles given by PCS to form the original sequence X of the grey GM (1,1) prediction model. (0) , the original sequence expression is

[0048] X (0) =[X (0) (1),X (0) (2),X (0) (3),X (0) (4)] (1)

[0049] And transform the weight of each information in the original sequence;

[0050] Background value calculation module is used to calculate the background value based on the original sequence X (0) , generate the cumulative sequence X (1) (k), the cumulative sequence expression is

[0051]

[0052] Then use the simplified Simpson formula to reconstruct the background value Z (1) (k);

[0053] Gray prediction module is used to predict the background value Z (1) (k) and the original displacement sampling data, calculate the gray development coefficient a and the gray amount b; according to the gray development coefficient a, the gray amount b and the original sequence X (0) , obtain the prediction formula of the original displacement signal;

[0054] Error checking module, used to predict according to the prediction formula The value of and prediction error check; use the prediction formula that meets the error check to predict the displacement output value of the next cycle

[0055] Reference signal output module for predicting displacement The plasma displacement deviation is obtained by comparison with the reference displacement, and the transfer function is calculated according to the displacement deviation and the reference signal to obtain the reference signal.

[0056] Furthermore, the acquisition module includes an information weight transformation unit, and the transformation expression used during operation is specifically:

[0057]

[0058] Furthermore, the background value calculation module includes a Simpson background value reconstruction unit, and its specific operation mode is as follows:

[0059] In the complex Simpson formula, the k~k+1 sampling period is divided into m intervals, and the cumulative sequence curve expression is set to f(x i ), among which

[0060]

[0061] The expression for solving the background value by complexing Simpson's formula to approximate the integral process is:

[0062]

[0063] Accumulate the sequence X within k~k+1 sampling cycles (1) (k) to X (1) The (k+1) change value is divided into m equal parts, then the cumulative sequence change value in the m intervals is

[0064]

[0065] X in the cumulative sequence (1) (k) and X (1) The value of (k+1) is known, so let

[0066] f(x i )=X (1) (k)+iΔ,i=0,1,…,m-1 (14)

[0067]

[0068] Combining formulas (12) to (15) to simplify the complex Simpson formula, the reconstructed background value is

[0069]

[0070] The advantages of the present invention are:

[0071] (1) The present invention improves the influence of old information on prediction accuracy in traditional grey prediction, transforms the weight of each information in the original sequence to increase the influence weight of new information, and further improves the prediction accuracy of the grey GM (1,1) prediction model.

[0072] (2) The present invention uses the Simpson formula instead of the median theorem to construct the background value, which reduces the background value construction error in the traditional grey GM (1,1) prediction model, and simplifies the complex Simpson formula to ensure the accuracy of grey prediction without increasing the amount of calculation, thereby ensuring that the improved grey GM (1,1) prediction model can quickly and accurately predict the plasma displacement.

[0073] (3) The present invention uses known displacement information to perform error check on the predicted displacement until a gray GM (1,1) prediction model that meets the error requirements is trained, thereby realizing a closed-loop feedback check function for gray prediction. The displacement prediction accuracy of plasma displacement is greatly improved by using the gray GM (1,1) prediction model that meets the error requirements.

[0074] (4) The improved grey GM (1,1) prediction model is combined to make advance predictions on the plasma displacement trajectory, so that the reference signal can be obtained in advance. When a small deviation occurs in the plasma displacement, only a small current needs to be output by the tokamak magnet power supply to excite the load coil, and the plasma displacement balance control can be achieved through the magnetic field, thereby reducing the requirements on the power capacity of the magnet power supply and realizing the optimization of the tokamak magnet power supply capacity. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a flow chart of a plasma displacement balance control method for optimizing the tokamak magnet power supply capacity by improving the grey GM (1,1) model in an example of the present invention;

[0076] Figure 2 A schematic diagram showing the comparison of background value structures before and after improvement in an example of the present invention;

[0077] Figure 3 This is a diagram of the simulation results of plasma balance control in an example of the present invention. DETAILED DESCRIPTION

[0078] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in combination with the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0079] Example 1

[0080] The present invention provides a method for optimizing the capacity of a tokamak magnet power supply by improving the grey prediction. Figure 1 As shown, the following steps are included:

[0081] S1. Collect the plasma displacement data in the last four switching cycles given by PCS to form the original sequence X of the grey GM (1,1) prediction model (0) , the original sequence expression is

[0082] X (0) =[X (0) (1),X (0) (2),X(0) (3),X (0) (4)] (1)

[0083] New information is often more consistent with the trend of the predicted value. In order to reduce the influence of old information and increase the weight of new information, the weight of each information in the original sequence is transformed to increase the influence of new information. The transformation expression of each information is

[0084]

[0085] The transformed δ(1)~δ(4) form a new original sequence. The oldest information δ(1) in the new original sequence also contains some new information X (0) (4), which increases the weight of new information.

[0086] S2, based on the original sequence X (0) , generate the cumulative sequence X (1) (k), the cumulative sequence expression is

[0087]

[0088] Then, according to the idea of ​​integration, the cumulative sequence X (1) (k) At the k-1th sampling cycle time t k-1 Time t to the kth sampling cycle k The integral on the background value Z (1) (k), the background value expression is

[0089]

[0090] In order to simplify the calculation in grey prediction, the median theorem is used to approximate the background value. The approximate background value expression is:

[0091]

[0092] However, the background value constructed by the median theorem has certain errors. Therefore, this embodiment uses the simplified Simpson formula to reconstruct the background value to ensure the accuracy of grey prediction without increasing the amount of calculation.

[0093] like Figure 2 As shown, the background value Z before improvement (1) (k) is X (1) (k) and X (1) The area of ​​the right-angled trapezoid formed by the (k+1) line and the k~k+1 time axis, the improved background value Z (1) (k) is X (1) (k) and X (1)The area of ​​the curved trapezoid formed by the (k+1) curve and the k~k+1 time axis. The improved background value construction process uses the simplified Simpson formula to approximate the curve integration process, making the background value construction more accurate, ensuring the accuracy of gray prediction without increasing the amount of calculation. In the complex Simpson formula, the k~k+1 sampling cycle time is divided into m intervals, and the cumulative sequence curve expression is set to f(x i ), and the specific expression requires complex calculation or fitting, among which

[0094]

[0095] The expression for solving the background value by complexing Simpson's formula to approximate the integral process is:

[0096]

[0097] The cumulative sequence curve expression f(x i ) The specific expression is difficult to obtain, so the sequence X is accumulated during the k~k+1 sampling period (1) (k) to X (1) The (k+1) change value is divided into m equal parts, then the cumulative sequence change value in the m intervals is

[0098]

[0099] X in the cumulative sequence (1) (k) and X (1) The value of (k+1) is known, so let

[0100] f(x i )=X (1) (k)+iΔ,i=0,1,…,m-1 (14)

[0101]

[0102] The complex Simpson formula is simplified by combining formulas (12) to (15). The simplified Simpson formula is used to approximate the integral process and replace the mean value theorem to reconstruct the background value. The background value reconstructed by the simplified Simpson formula is

[0103]

[0104] The simplified Simpson formula described in formula (16) is used to approximate the integration process, instead of the mean value theorem to reconstruct the background value, which not only ensures a higher accuracy of the reconstructed background value, but also simplifies the calculation amount of the integration process.

[0105] S3, according to the background value Z (1)(k) and the original displacement sampling data, calculate the gray development coefficient a and gray quantity b, the calculation formulas of a and b are:

[0106]

[0107] X n =[X (0) (2)X (0) (3)X (0) (4)] T (7)

[0108] According to the gray development coefficient a and gray amount b and the original sequence X (0) , predict the original displacement signal, the prediction formula is

[0109]

[0110] S4. Prediction based on the prediction formula The value of is used to perform prediction error check, and the displacement output value of the next cycle is predicted using the prediction formula that satisfies the error check. Assume that the acceptable prediction error is e, and use the error check formula to determine whether the predicted value meets the prediction error. The error check formula is:

[0111]

[0112] When the error check formula is met, the next displacement output value is predicted according to the prediction formula Otherwise, update the weight of the new information in the original sequence, and then continue to update the values ​​of a and b to predict Until the error check formula is satisfied, the next displacement output value is predicted

[0113] In the above steps S3 and S4, it is first necessary to predict the plasma displacement signal of the fourth switching cycle based on the plasma displacement data in the last four switching cycles. The predicted displacement signal is compared with the actual displacement signal X (0) (4) Compare and realize the error checking function until the predicted displacement signal and the actual displacement signal meet the prediction error, achieving the purpose of gray GM (1,1) prediction closed loop test, and then predict and output the plasma displacement signal of the next switching cycle The accuracy of displacement prediction is guaranteed.

[0114] S5. Predicted displacement The plasma displacement deviation is obtained by comparing with the reference displacement, and the transfer function is calculated based on the displacement deviation and the reference signal to obtain the reference signal. The tokamak magnet power supply tracks the reference signal output current, excites the load coil, and realizes the magnetic balance control of the plasma displacement.

[0115] The reference signal is obtained by obtaining the plasma displacement deviation in advance. When there is a small deviation in the plasma displacement, only a smaller reference signal is needed to be given to the tokamak magnet power supply. The magnet power supply only needs to output a smaller current to excite the load coil, that is, only a smaller capacity magnet power supply is needed to achieve magnetic balance control of the plasma displacement.

[0116] The reference signal calculation transfer function is

[0117]

[0118] Among them, K p , K i , K d The coefficients of the transfer function are calculated for the reference signal, and s is the Laplace operator; in the traditional function, the input is the plasma displacement deviation obtained in advance, and the output is the reference signal given to the tokamak magnet power supply.

[0119] In this embodiment, the tokamak magnet power supply capacity optimization method based on improved grey prediction is simulated and analyzed, and the simulation parameters are: the reference signal amplitude is ±10V, the load inductance value is 160μH, the inductor internal resistance value is 25mΩ, the magnet power supply feedback current sensor ratio is 50000:1, the plasma discharge current is set to 1MA, and the reference signal transfer function parameter K p =100, K i =24, K d =3, the initial displacement deviation of plasma is 10cm. The plasma displacement is predicted in advance by using improved grey prediction to obtain the reference signal in advance. Figure 3 As shown, analysis of the simulation waveform shows that the reference signal is limited to ±10V, the plasma displacement can be quickly controlled to a stable zero position, the peak current required to be output by the Tokamak magnet power supply is 215.7kA, and the peak voltage required to be output is 5.45kV. In contrast, when the plasma discharge current in the CFETR nuclear fusion device is set to 1MA, the designer gives the required magnet power supply capacity of ±7.46kV / ±309.74kA. After gray prediction, the plasma displacement is predicted in advance, and the reference signal can be output in advance when the plasma has a small offset. The maximum current required by the magnet power supply and the maximum voltage required to be established at the output end are greatly reduced, which shows that the present invention can greatly reduce the power capacity of the Tokamak magnet power supply, and optimize the magnet power supply capacity while also ensuring the plasma displacement balance control performance.

[0120] Example 2

[0121] It should be further explained that, based on the same inventive concept, this embodiment provides a tokamak magnet power capacity optimization system with improved grey prediction. When the system is running, it executes the method described in embodiment 1, including the following modules:

[0122] The acquisition module is used to collect the plasma displacement data in the last four switching cycles given by PCS to form the original sequence X of the grey GM (1,1) prediction model. (0) , the original sequence expression is

[0123] X (0) =[X (0) (1),X (0) (2),X (0) (3),X (0) (4)] (1)

[0124] And transform the weight of each information in the original sequence;

[0125] Background value calculation module is used to calculate the background value based on the original sequence X (0) , generate the cumulative sequence X (1) (k), the cumulative sequence expression is

[0126]

[0127] Then use the simplified Simpson formula to reconstruct the background value Z (1) (k);

[0128] Gray prediction module is used to predict the background value Z (1) (k) and the original displacement sampling data, calculate the gray development coefficient a and the gray amount b; according to the gray development coefficient a, the gray amount b and the original sequence X (0) , obtain the prediction formula of the original displacement signal;

[0129] Error checking module, used to predict according to the prediction formula The value of and prediction error check; use the prediction formula that meets the error check to predict the displacement output value of the next cycle

[0130] Reference signal output module for predicting displacement The plasma displacement deviation is obtained by comparison with the reference displacement, and the reference signal is obtained by calculating the transfer function based on the displacement deviation and the reference signal.

[0131] The acquisition module includes an information weight transformation unit, and the transformation expression used during operation is specifically:

[0132]

[0133] The background value calculation module includes a Simpson background value reconstruction unit, and its specific operation mode is as follows:

[0134] In the complex Simpson formula, the k~k+1 sampling period is divided into m intervals, and the cumulative sequence curve expression is set to f(x i ), among which

[0135]

[0136] The expression for solving the background value by complexing Simpson's formula to approximate the integral process is:

[0137]

[0138] Accumulate the sequence X within k~k+1 sampling cycles (1) (k) to X (1) The (k+1) change value is divided into m equal parts, then the cumulative sequence change value in the m intervals is

[0139]

[0140] X in the cumulative sequence (1) (k) and X (1) The value of (k+1) is known, so let

[0141] f(x i )=X (1) (k)+iΔ,i=0,1,…,m-1 (14)

[0142]

[0143] Combining formulas (12) to (15) to simplify the complex Simpson formula, the reconstructed background value is

[0144]

[0145] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An improved grey prediction method for optimizing the power supply capacity of a tokamak magnet is characterized in that: The following steps are involved: S1. Collect the plasma displacement data in the last four switching cycles given by PCS to form the original sequence X of the grey GM (1,1) prediction model (0) , the original sequence expression is X (0) =[X (0) (1),X (0) (2),X (0) (3),X (0) (4)] (1) And transform the weight of each information in the original sequence; S2, based on the original sequence X (0) , generate the cumulative sequence X (1) (k), the cumulative sequence expression is Then use the simplified Simpson formula to reconstruct the background value Z (1) (k); S3, according to the background value Z (1) (k) and the original displacement sampling data, calculate the gray development coefficient a and the gray amount b; according to the gray development coefficient a, the gray amount b and the original sequence X (0) , obtain the prediction formula of the original displacement signal; S4. Prediction based on the prediction formula The value of and prediction error check; use the prediction formula that meets the error check to predict the displacement output value of the next cycle S5. Predicted displacement The plasma displacement deviation is obtained by comparison with the reference displacement, and the transfer function is calculated according to the displacement deviation and the reference signal to obtain the reference signal.

2. The method for optimizing the tokamak magnet power supply capacity by improved grey prediction according to claim 1, characterized in that: The weight of each information in the original sequence is transformed as described in step S1. The specific transformation expression is:

3. The method for optimizing the tokamak magnet power supply capacity by improving grey prediction according to claim 1, characterized in that: The background value is reconstructed using the simplified Simpson formula described in step S2, specifically: In the complex Simpson formula, the k~k+1 sampling period is divided into m intervals, and the cumulative sequence curve expression is set to f(x i ), among which The expression for solving the background value by complexing Simpson's formula to approximate the integral process is: Accumulate the sequence X within k~k+1 sampling cycles (1) (k) to X (1) The (k+1) change value is divided into m equal parts, then the cumulative sequence change value in the m intervals is X in the cumulative sequence (1) (k) and X (1) The value of (k+1) is known, so let f(x i )=X (1) (k)+iΔ,i=0,1,…,m-1 (14) Combining formulas (12) to (15) to simplify the complex Simpson formula, the reconstructed background value is 4. The method for optimizing the tokamak magnet power supply capacity by improving grey prediction according to claim 3, characterized in that: Step S3 described in accordance with the background value Z (1) (k) and the original displacement sampling data, calculate the gray development coefficient a and gray quantity b. The specific calculation method is:

5. The method for optimizing the tokamak magnet power supply capacity by improved grey prediction according to claim 4, characterized in that: The prediction formula of the original displacement signal described in step S3 is specifically:

6. The method for optimizing the tokamak magnet power supply capacity by improved grey prediction according to claim 5, characterized in that: Step S4 uses the prediction formula that satisfies the error check to predict the displacement output value of the next cycle. Specifically: Assume that the acceptable prediction error is e, and use the error check formula to determine whether the predicted value meets the prediction error. The error check formula is: When the error check is satisfied, the next displacement output value is predicted according to the prediction formula Otherwise, update the weight of the new information in the original sequence, recalculate the gray development coefficient a and gray quantity b, and get the new prediction value according to the prediction formula. And recheck until the error check is satisfied.

7. The method for optimizing the tokamak magnet power supply capacity by improving grey prediction according to claim 1, characterized in that: The reference signal calculation transfer function described in step S5 is: Among them, K p , K i , K d The coefficient of the transfer function is calculated for the reference signal, s is the Laplace operator; the input of the transfer function is the obtained plasma displacement deviation, and the output is the reference signal given to the tokamak magnet power supply.

8. Improved grey prediction tokamak magnet power capacity optimization system, characterized in that: When the system is running, the method described in any one of claims 1 to 7 is adopted, including the following modules: The acquisition module is used to collect the plasma displacement data in the last four switching cycles given by PCS to form the original sequence X of the grey GM (1,1) prediction model. (0) , the original sequence expression is X (0) =[X (0) (1),X (0) (2),X (0) (3),X (0) (4)] (1) And transform the weight of each information in the original sequence; Background value calculation module is used to calculate the background value based on the original sequence X (0) , generate the cumulative sequence X (1) (k), the cumulative sequence expression is Then use the simplified Simpson formula to reconstruct the background value Z (1) (k); Gray prediction module is used to predict the background value Z (1) (k) and the original displacement sampling data, calculate the gray development coefficient a and the gray amount b; according to the gray development coefficient a, the gray amount b and the original sequence X (0) , obtain the prediction formula of the original displacement signal; Error checking module, used to predict according to the prediction formula The value of and prediction error check; use the prediction formula that meets the error check to predict the displacement output value of the next cycle Reference signal output module for predicting displacement The plasma displacement deviation is obtained by comparison with the reference displacement, and the reference signal is obtained by calculating the transfer function based on the displacement deviation and the reference signal.

9. The Tokamak magnet power supply capacity optimization system with improved grey prediction according to claim 8, characterized in that: The acquisition module includes an information weight transformation unit, and the transformation expression used during operation is specifically:

10. The Tokamak magnet power capacity optimization system with improved grey prediction according to claim 8, characterized in that: The background value calculation module includes a Simpson background value reconstruction unit, and its specific operation mode is as follows: In the complex Simpson formula, the k~k+1 sampling period is divided into m intervals, and the cumulative sequence curve expression is set to f(x i ), among which The expression for solving the background value by complexing Simpson's formula to approximate the integral process is: Accumulate the sequence X within k~k+1 sampling cycles (1) (k) to X (1) The (k+1) change value is divided into m equal parts, then the cumulative sequence change value in the m intervals is X in the cumulative sequence (1) (k) and X (1) The value of (k+1) is known, so let f(x i )=X (1) (k)+iΔ,i=0,1,…,m-1 (14) Combining formulas (12) to (15) to simplify the complex Simpson formula, the reconstructed background value is

Images