Self-adaptive active noise suppression control system and method for power equipment

Abstract

The invention relates to the technical field of transformer noise reduction, in particular to a self-adaptive active noise suppression control system and method for power equipment, and the method comprises the steps: S1, carrying out the finite element analysis of a transformer housing, and obtaining the modal shape of the transformer housing, and the position of a response point with the response greater than a preset response demand; s2, generating a layout scheme for arranging the inertial exciters on the response points according to the obtained positions of the response points, and optimizing the layout scheme to arrange the inertial exciters on the response points; s3, estimating and compensating total disturbance by adopting a linear active-disturbance-rejection vibration control algorithm, and controlling the work of the inertial exciter; and S4, judging whether the vibration isolation effect for controlling the work of the inertial exciter meets a preset vibration isolation requirement or not, and if not, executing the step S2. According to the scheme, the environmental noise can be reduced by controlling the vibration of the transformer box body, the method has relatively strong robustness on parameter change, disturbance and noise, and internal interference and external interference are effectively suppressed, so that a better vibration suppression effect is obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of transformer noise reduction, and in particular to an adaptive active noise suppression control system and method for electric power equipment. Background Art

[0002] With the continuous expansion of urban construction, some small, medium or combined transformers (hereinafter referred to as transformers) are increasingly installed near or directly in residential and commercial areas in order to meet the needs of urban power grid transformation. As the main problem of power distribution equipment, the noise pollution problem of transformers is increasingly attracting people's attention. Therefore, reducing transformer noise has become an important task that needs to be solved urgently. Transformer noise not only affects the quietness of the surrounding environment, but also may have a negative impact on the physical and mental health of residents. Long-term exposure to high noise environments can lead to hearing loss, sleep disorders, and increased stress. In addition, many regions have strict environmental noise standards and regulations to limit noise emissions from industrial equipment. And high noise usually means that there are undesirable vibrations and mechanical stresses inside the transformer, which may lead to equipment failure and shortened life. By controlling noise, these negative effects can be reduced, the reliability and service life of the transformer can be improved, and the maintenance and replacement costs can be reduced.

[0003] Traditional noise reduction methods are difficult to achieve effective noise suppression due to limitations in economics and safety. With the advancement of active noise control (ANC) technology, it is expected to be greatly developed and valued in noise reduction applications in substations. ANC technology has been successfully applied to noise reduction in limited enclosed spaces such as automobiles and ships. In recent years, the noise pollution problem of substations has received more and more attention from the society, and the application of ANC in such open sound fields has gradually increased.

[0004] Active noise control (ANC) technology for transformers neutralizes noise by generating anti-phase sound waves, and has gradually become a key means of noise reduction in substations. Its core lies in the use of control algorithms and sensor layout to achieve effective noise suppression. Commonly used control algorithms, such as the least mean square (LMS) algorithm, strike a balance between the steady-state error, convergence speed and calculation delay of noise to optimize the noise control effect. In the complex free sound field environment of the substation, there are phenomena such as sound reflection, diffraction and wave coupling, which increase the difficulty of global noise reduction. Therefore, the research mainly focuses on combining the sound radiation characteristics of the application environment to achieve the local optimal noise reduction effect. Although the transformer noise characteristics are complex, including wide-band and low-frequency noise, ANC technology responds to these challenges through multi-channel systems and adaptive filtering. While maintaining stability, the ANC system needs to prevent distortion and noise rebound. Improving the robustness of the algorithm and the real-time response capability of the system are key research directions. Compared with traditional sound insulation measures, ANC technology achieves noise reduction without affecting the heat dissipation of the equipment, and has the advantages of energy saving and environmental protection. With the development of artificial intelligence and Internet of Things technologies, the ANC system will be further integrated with smart grid technology to adjust the noise reduction strategy in real time and improve the noise reduction effect. These developments show that ANC technology has broad application prospects in substations and other industrial fields.

[0005] For ordinary fully indoor substations, the method of installing normally closed soundproof doors and windows is often used to block the noise radiated by the transformer; for special fully indoor substations - underground substations, since the existing underground building walls have a good isolation effect on transformer noise, compared with transformer noise, the operation and maintenance personnel pay more attention to the suppression of transformer and station building structure vibration. For the noise generated by the transformer running in the semi-indoor substation, in addition to first using cooling fans with lower noise levels, sound barriers are often used to block the transformer noise on the propagation path. At present, the closed vibration isolation cover, soundproof room and transformer base shock absorption technology are often used to achieve passive noise cancellation. In addition, passive vibration absorber technology has also been applied. The closed soundproof cover adopts traditional sound insulation technology, but its effect is not ideal. There is still a lot of noise transmitted, affecting people's normal life and work. In addition, the soundproof cover hinders the heat dissipation of the transformer, causing its temperature to rise, increasing the danger. The setting process of the soundproof room is complicated and costly. Not only does it need to build a soundproof room, but it also needs to use special processes to make doors and windows, and install ventilation devices. However, the ventilation device will generate additional noise during operation, so noise reduction equipment needs to be installed. These complex installations and high costs make it difficult to widely use the soundproof room solution. Although the soundproofing modification of the transformer base vibration reduction and the rigid separation of the cooling fan can reduce the noise generated by mechanical vibration, it is ineffective for the low-frequency noise generated inside the transformer. This solution is expensive and the modification process is complicated. Generally speaking, it is not feasible.

[0006] In addition to physical sound barriers, there are also efforts to develop "active sound barriers" or "virtual sound barriers". Although these sound barriers suppress noise through active noise control, which is fundamentally different from the direct sound isolation principle of physical sound barriers, they are an important part of the perfect research system. The key points of active noise control systems are the selection of control algorithms and the placement of secondary sound sources, which help to significantly improve the noise reduction performance. When selecting a control algorithm, the characteristics of the noise must be considered. Taking the steady-state error, convergence speed and calculation delay of the algorithm as evaluation criteria, the optimal setting can be obtained in theory. However, the sound reflection, sound diffraction and sound wave coupling phenomena in the free sound field make it difficult to achieve effective noise reduction in the global space. Therefore, in order to achieve the local optimal effect, it is necessary to combine the sound radiation characteristics of the application environment and the specific noise reduction requirements. At present, the research on transformer adaptive control technology mainly focuses on the least mean square (LMS) algorithm and its derivative algorithms. The LMS algorithm is sensitive to interference sources and acoustic feedback phenomena, so there are many technical challenges in its application in transformer noise reduction. Although the LMS algorithm is simple to calculate and easy to implement, its convergence speed is slow and the step size parameter needs to balance the steady-state error and convergence speed. In practical applications, it is impossible to consider the interference problem of noise source signal pickup to ensure the best noise reduction effect.

[0007] To sum up, among the existing noise suppression methods, the traditional physical sound barrier method is expensive, the modification process is complicated, and usually has low feasibility; while active noise reduction methods such as "active sound barriers" or "virtual sound barriers" do not take into account the interference of noise signals and the stability of the system, have low robustness, and are ineffective in suppressing total disturbances (internal and external interference), which in turn leads to unsatisfactory vibration suppression effects. Summary of the invention

[0008] One of the purposes of the present invention is to provide an adaptive active noise suppression control method for power equipment, which can reduce environmental noise by controlling the vibration of the transformer box, has strong robustness to parameter changes, disturbances and noise, and effectively suppresses internal interference caused by modeling errors and unmodeled dynamics, as well as external interference such as measurement noise and external excitation, so as to obtain better vibration suppression effect.

[0009] The present invention provides a basic solution 1: an adaptive active noise suppression control method for electric power equipment, comprising:

[0010] S1. Perform finite element analysis on the transformer housing to obtain the modal vibration shape of the transformer housing and the position of the response point whose response is greater than the preset response requirement, extract the mass matrix and stiffness matrix, and model the controlled structure;

[0011] S2. According to the acquired response point positions, a layout scheme is generated to arrange the inertial exciter at the response point, and the layout scheme is optimized to output a layout scheme that meets the vibration effect. According to the layout scheme, the inertial exciter is arranged at the response point;

[0012] S3, using linear anti-disturbance vibration control algorithm to estimate and compensate for total disturbances and control the operation of the inertial exciter;

[0013] S4, judging whether the vibration isolation effect of controlling the inertial exciter meets the preset vibration isolation requirement, if so, adopting the current layout scheme and parameter setting to perform active noise reduction; if not, executing S2 to re-optimize the layout scheme.

[0014] Furthermore, the finite element analysis is performed on the transformer housing to obtain the dynamic model of the transformer system:

[0015]

[0016] Where: M∈n×n is the mass matrix of the controlled structure;

[0017] C∈n×n is the damping matrix of the controlled structure;

[0018] K∈n×n is the stiffness matrix of the controlled structure;

[0019] is the acceleration response vector of the structure;

[0020] is the acceleration response vector of the structure;

[0021] x(t)∈n is the displacement response vector of the structure;

[0022] n is the modal order;

[0023] D s ∈n×r is the disturbance force position matrix;

[0024] F(t)∈r is the disturbance force vector;

[0025] B s ∈n×p is the position matrix of the inertial actuator; the inertial actuator is the actuator;

[0026] p is the number of inertial exciters;

[0027] U(t)∈p is the p-dimensional control force vector.

[0028] Further, the S2 includes:

[0029] S201, according to safety requirements, using the optimization configuration criteria of structural response control, generating several layout schemes for arranging inertial exciters at response points;

[0030] S202, construct a fitness function Fit for evaluating parameters [B s ] performance indicators;

[0031]

[0032] Among them, y i is the actual observed value;

[0033] is based on the parameter [B s ]’s model prediction value;

[0034] N is the number of observations;

[0035] S203, calculating the fitness function value of the layout solution according to the fitness function Fit;

[0036] S204, obtain the minimum fitness function value, and [B s ] In the layout scheme that meets the preset range;

[0037] S205: Output the obtained layout solution.

[0038] Further, the execution of the linear active disturbance rejection vibration control algorithm includes:

[0039] S301. For an n-degree-of-freedom vibration system, its state space model is expressed as:

[0040]

[0041] Among them, F c (t)∈r is the disturbance force vector.

[0042] Convert the above equation into state space form:

[0043]

[0044] in, is the state vector;

[0045] U(t) is the control input;

[0046] d(t) is the external disturbance;

[0047] A, B, and E are system matrices;

[0048] S302, determine the control target: reduce the displacement x(t) and suppress the external disturbance d(t);

[0049] S303, setting the extended state observer ESO:

[0050] ESO is used to estimate the system state and total disturbance;

[0051]

[0052] y(t)=C e Z(t)

[0053] in, C e =[C 0];

[0054] S304, setting controller:

[0055] The LADRC control law is set as:

[0056] U(t)=U 0 (t)-d(t);

[0057] U 0 (t) is the basic control input based on the estimated state, d(t) is the total disturbance estimated by the ESO;

[0058] S305, basic control input settings:

[0059] Design the basic control input using proportional-derivative (PD) control:

[0060] U 0 (t) = K p x(t)+K d x(t);

[0061] Among them, K p is the proportional gain matrix, K d is the differential gain matrix;

[0062] S306: Adjust the ESO gain matrix L and controller gain K p and K d Optimize system performance.

[0063] Furthermore, the ESO gain matrix L and controller gain K are adjusted p and K d The methods include trial and error, frequency domain analysis or optimization algorithm.

[0064] The second object of the present invention is to provide an adaptive active noise suppression control system for power equipment, which can reduce environmental noise by controlling the vibration of the transformer box, has strong robustness to parameter changes, disturbances and noise, and effectively suppresses internal interference caused by modeling errors and unmodeled dynamics, as well as external interference such as measurement noise and external excitation, so as to obtain better vibration suppression effect.

[0065] The present invention provides a second basic solution: an adaptive active noise suppression control system for electric power equipment, comprising: an inertial exciter and a linear active disturbance rejection controller;

[0066] The inertial exciter is arranged on the transformer housing according to the layout plan; wherein the layout plan is obtained by: performing finite element analysis on the transformer housing, obtaining the modal vibration shape of the transformer housing, and the position of the response point whose response is greater than the preset response requirement, extracting the mass matrix and the stiffness matrix, and modeling the controlled structure; according to the obtained response point position, generating a layout plan for arranging the inertial exciter on the response point, and optimizing the layout plan to obtain a layout plan that meets the vibration effect;

[0067] The linear active disturbance rejection controller is used to execute the linear active disturbance rejection vibration control algorithm, estimate and compensate the total disturbance, and control the operation of the inertial exciter; and if the vibration isolation effect of the linear active disturbance rejection controller controlling the operation of the inertial exciter does not meet the preset vibration isolation requirements, the layout plan is re-optimized.

[0068] Furthermore, the finite element analysis is performed on the transformer housing to obtain the dynamic model of the transformer system:

[0069]

[0070] Where: M∈n×n is the mass matrix of the controlled structure;

[0071] C∈n×n is the damping matrix of the controlled structure;

[0072] K∈n×n is the stiffness matrix of the controlled structure;

[0073] is the acceleration response vector of the structure;

[0074] is the acceleration response vector of the structure;

[0075] x(t)∈n is the displacement response vector of the structure;

[0076] n is the modal order;

[0077] D s ∈n×r is the disturbance force position matrix;

[0078] F(t)∈r is the disturbance force vector;

[0079] B s ∈n×p is the position matrix of the inertial actuator; the inertial actuator is the actuator;

[0080] p is the number of inertial exciters;

[0081] U(t)∈p is the p-dimensional control force vector.

[0082] Further, generating a layout scheme for arranging the inertial exciter at the response point according to the acquired response point position, and optimizing the layout scheme to obtain a layout scheme that meets the vibration effect, includes:

[0083] According to safety requirements, the optimal configuration criteria of structural response control are used to generate several layout schemes for placing inertial actuators on response points.

[0084] Construct a fitness function Fit to evaluate the parameters [B s ] performance indicators;

[0085]

[0086] Among them, y i is the actual observed value;

[0087] is based on the parameter [B s ]’s model prediction value;

[0088] N is the number of observations;

[0089] According to the fitness function Fit, the fitness function value of the layout scheme is calculated;

[0090] Get the minimum fitness function value, and [B s ] In the layout scheme that meets the preset range;

[0091] Output the obtained layout scheme.

[0092] Furthermore, the linear active disturbance rejection controller is used to execute a linear active disturbance rejection vibration control algorithm, estimate and compensate for total disturbances, and control the operation of the inertial actuator, including:

[0093] For an n-degree-of-freedom vibration system, its state space model is expressed as:

[0094]

[0095] Among them, F c (t)∈r is the disturbance force vector.

[0096] Convert the above equation into state space form:

[0097]

[0098] in, is the state vector;

[0099] U(t) is the control input;

[0100] d(t) is the external disturbance;

[0101] A, B, and E are system matrices;

[0102] Determine the control objective: reduce the displacement x(t) and suppress the external disturbance d(t);

[0103] To set up the Extended State Observer ESO:

[0104] ESO is used to estimate the system state and total disturbance;

[0105]

[0106] y(t)=C e Z(t)

[0107] in, C e =[C 0];

[0108] Set up the controller:

[0109] The LADRC control law is set as:

[0110] U(t)=U 0 (t)-d(t);

[0111] U 0 (t) is the basic control input based on the estimated state, d(t) is the total disturbance estimated by the ESO;

[0112] Basic control input settings:

[0113] Design the basic control input using proportional-derivative (PD) control:

[0114] U 0 (t) = K p x(t)+K d x(t);

[0115] Among them, K p is the proportional gain matrix, K d is the differential gain matrix;

[0116] Adjust the ESO gain matrix L and controller gain K p and K d Optimize system performance.

[0117] Furthermore, the ESO gain matrix L and controller gain K are adjusted p and K d The methods include trial and error, frequency domain analysis or optimization algorithm.

[0118] Beneficial effects: This scheme will use an inertial exciter as an active control actuator for transformer noise, reduce environmental noise by controlling the vibration of the transformer box, and take into account the interference of noise signals and the stability of the system. A linear self-disturbance rejection vibration control algorithm is used to estimate and compensate for the total disturbance including internal and external disturbances. It has strong robustness to parameter changes, disturbances and noise, and effectively suppresses internal disturbances caused by modeling errors and unmodeled dynamics, as well as external disturbances such as measurement noise and external excitation, to achieve better vibration suppression effects.

[0119] Specifically, 1. The linear active disturbance rejection vibration control algorithm (LADRC) of this scheme is a linear form of ADRC with fewer control parameters, which is more conducive to implementation and engineering application than traditional vibration control methods;

[0120] 2. This scheme does not need to build an accurate model of the controlled object. It only uses the error between the control target and the actual behavior of the object to generate a process control strategy to eliminate this error. It can realize the control of multivariable systems and has strong followability and anti-interference characteristics. The control of multi-input and multi-output systems requires little calculation and has good robustness, which can achieve better multi-point control.

[0121] 3. Through active optimization of the inertial exciter and the use of active noise reduction technology, the problems of many action points, poor heat dissipation and high cost of traditional passive vibration isolation are overcome. BRIEF DESCRIPTION OF THE DRAWINGS

[0122] Figure 1 A schematic diagram of a flow chart of an embodiment of an adaptive active noise suppression control method for electric equipment according to the present invention;

[0123] Figure 2 It is a schematic diagram of the layout of the inertial exciter in an embodiment of an adaptive active noise suppression control method for electric power equipment of the present invention;

[0124] Figure 3 It is a LADRC control block diagram in an embodiment of an adaptive active noise suppression control method for electric power equipment of the present invention. DETAILED DESCRIPTION

[0125] The following is further described in detail through specific implementation methods:

[0126] The reference numerals in the drawings of the specification include: inertial exciter 1, transformer oil 2, transformer housing 3, magnetic flux 4, coil winding 5, transformer foundation 6.

[0127] The embodiment is basically as shown in the attached Figure 1 As shown: An adaptive active noise suppression control method for electric power equipment, comprising:

[0128] S1. Perform finite element analysis on the transformer housing 3 to obtain the modal vibration shape of the transformer housing 3 and the position of the response point whose response is greater than the preset response requirement, extract the mass matrix and the stiffness matrix, and model the controlled structure; wherein the controlled structure is the transformer housing 3; wherein the position of the response point whose response is greater than the preset response requirement, that is, the modal vibration shape vibration of the region is greater than the preset regional modal vibration shape vibration value;

[0129] Specifically, a finite element analysis is performed on the transformer housing 3 to obtain the dynamic model of the transformer system:

[0130]

[0131] Where: M∈n×n is the mass matrix of the controlled structure;

[0132] C∈n×n is the damping matrix of the controlled structure;

[0133] K∈n×n is the stiffness matrix of the controlled structure;

[0134] is the acceleration response vector of the structure;

[0135] is the acceleration response vector of the structure;

[0136] x(t)∈n is the displacement response vector of the structure;

[0137] n is the modal order;

[0138] D s ∈n×r is the disturbance force position matrix;

[0139] F(t)∈r is the disturbance force vector;

[0140] B s ∈n×p is the position matrix of the inertial actuator 1; the inertial actuator 1 is the actuator;

[0141] p is the number of inertial exciters 1;

[0142] U(t)∈p is the p-dimensional control force vector.

[0143] S2. Based on the acquired response point position, generate a layout plan for placing the inertial exciter 1 on the response point, optimize the layout plan, and output a layout plan that meets the vibration effect. Based on the layout plan, place the inertial exciter 1 on the response point, such as Figure 2 As shown, the inertial exciter 1 is arranged on a transformer housing 3, and the transformer housing 3 is arranged on a transformer foundation 6, and the interior includes transformer oil 2, magnetic flux 4, and coil winding 5;

[0144] Specifically, S2 includes:

[0145] S201, according to safety requirements, using the optimization configuration criteria of structural response control, generating a number of layout schemes for arranging the inertial exciter 1 at the response point;

[0146] S202, construct a fitness function Fit for evaluating parameters [B s ] performance indicators;

[0147]

[0148] Among them, y i is the actual observed value;

[0149] is based on the parameter [B s ]’s model prediction value;

[0150] N is the number of observations;

[0151] S203, calculating the fitness function value of the layout solution according to the fitness function Fit;

[0152] S204, obtain the minimum fitness function value, and [B s ] In a layout scheme that complies with a preset range; the preset range is the position range of the inertial exciter allowed to be arranged on the transformer housing;

[0153] S205: Output the obtained layout solution.

[0154] S206 , arranging the inertial exciter 1 at the response point according to the layout plan.

[0155] S3, using a linear anti-disturbance vibration control algorithm to estimate and compensate for the total disturbance and control the operation of the inertial exciter 1;

[0156] The parameters in the linear anti-disturbance vibration control algorithm are adjusted according to the layout plan. After adjusting the parameters, the vibration isolation effect is tested. If the effect is good, it can be used for active noise reduction. If the effect is not good, it returns to S2.

[0157] The total disturbance includes internal disturbance and external disturbance;

[0158] In this embodiment, a linear active disturbance rejection controller (LADRC) is used to execute the linear active disturbance rejection vibration control algorithm. LADRC usually consists of three parts: a trajectory planner, a state observer and a controller. For a multi-degree-of-freedom system, each degree of freedom usually needs to be designed separately. For vibration control, there is no need for a trajectory planner, and it is only necessary to reduce the vibration signal to 0. Therefore, a trajectory planner is not set in this scheme. The control logic of LADRC in this scheme is as follows: Figure 3 As shown;

[0159] Specifically, S301, an n-degree-of-freedom vibration system, whose state space model is expressed as:

[0160]

[0161] Among them, M∈n×n is the mass matrix of the controlled structure;

[0162] C∈n×n is the damping matrix of the controlled structure;

[0163] K∈n×n is the stiffness matrix of the controlled structure;

[0164] is the acceleration response vector of the structure;

[0165] is the acceleration response vector of the structure;

[0166] x(t)∈n is the displacement response vector of the structure;

[0167] n is the modal order;

[0168] F c (t)∈r is the disturbance force vector.

[0169] Convert the above equation into state space form:

[0170]

[0171] in, is the state vector;

[0172] U(t) is the control input;

[0173] d(t) is the external disturbance;

[0174] A, B, and E are system matrices.

[0175] S302, determine the control target: reduce the displacement x(t) and suppress the external disturbance d(t);

[0176] S303, setting the extended state observer (ESO):

[0177] ESO is used to estimate the system state and total disturbance;

[0178]

[0179] y(t)=C e Z(t)

[0180] in, C e =[C 0].

[0181] The design of ESO involves choosing an appropriate observer gain matrix L to make the observation error as small as possible.

[0182] S304, setting controller:

[0183] The LADRC control law is set as:

[0184] U(t)=U 0 (t)-d(t);

[0185] U 0 (t) is the basic control input based on the estimated state, and d(t) is the total disturbance estimated by the ESO.

[0186] S305, basic control input settings:

[0187] Design the basic control input using proportional-derivative (PD) control:

[0188] U 0 (t) = K p x(t)+K d x(t);

[0189] Among them, K p is the proportional gain matrix, K d is the differential gain matrix.

[0190] S306: Adjust the ESO gain matrix L and controller gain K p and K d Make the system performance optimal; the adjustment method can use trial and error, frequency domain analysis or optimization algorithm to determine these parameters; the optimal system performance is the maximum system vibration reduction value.

[0191] S4, judging whether the vibration isolation effect of controlling the work of the inertial exciter 1 meets the preset vibration isolation requirement. If so, adopt the current layout scheme and parameter settings to perform active noise reduction; if not, execute S2 to re-optimize the layout scheme; in this embodiment, the preset vibration isolation requirement is that the vibration isolation efficiency reaches 70% or more, and judging whether the vibration isolation effect of controlling the work of the inertial exciter 1 meets the preset vibration isolation requirement, that is, judging whether the vibration isolation efficiency of controlling the work of the inertial exciter 1 reaches 70% or more.

[0192] This embodiment also provides an adaptive active noise suppression control system for electric power equipment, comprising: an inertial exciter 1 and a linear active disturbance rejection controller;

[0193] The inertial exciter 1 is arranged on the transformer housing 3 according to the layout scheme; wherein the layout scheme is obtained by: performing finite element analysis on the transformer housing 3, obtaining the modal vibration shape of the transformer housing 3, and the position of the response point whose response is greater than the preset response requirement, extracting the mass matrix and the stiffness matrix, and modeling the controlled structure; according to the obtained response point position, generating a layout scheme in which the inertial exciter 1 is arranged on the response point, and optimizing the layout scheme to obtain a layout scheme that meets the vibration effect;

[0194] The linear active disturbance rejection controller is used to execute the linear active disturbance rejection vibration control algorithm, estimate and compensate for the total disturbance, and control the operation of the inertial exciter 1; and if the vibration isolation effect of the linear active disturbance rejection controller controlling the operation of the inertial exciter 1 does not meet the preset vibration isolation requirements, the layout plan is re-optimized.

[0195] The transformer housing 3 is subjected to finite element analysis to obtain the dynamic model of the transformer system:

[0196]

[0197] Where: M∈n×n is the mass matrix of the controlled structure;

[0198] C∈n×n is the damping matrix of the controlled structure;

[0199] K∈n×n is the stiffness matrix of the controlled structure;

[0200] is the acceleration response vector of the structure;

[0201] is the acceleration response vector of the structure;

[0202] x(t)∈n is the displacement response vector of the structure;

[0203] n is the modal order;

[0204] D s ∈n×r is the disturbance force position matrix;

[0205] F(t)∈r is the disturbance force vector;

[0206] B s ∈n×p is the position matrix of the inertial actuator 1; the inertial actuator 1 is the actuator;

[0207] p is the number of inertial exciters 1;

[0208] U(t)∈p is the p-dimensional control force vector.

[0209] According to the acquired response point position, a layout scheme for arranging the inertial exciter 1 at the response point is generated, and the layout scheme is optimized to obtain a layout scheme that meets the vibration effect. According to the layout scheme, the inertial exciter 1 is arranged at the response point;

[0210] Specifically, according to safety requirements, the optimization configuration criterion of structural response control is used to generate several layout schemes for arranging the inertial exciter 1 on the response point;

[0211] Construct a fitness function Fit to evaluate the parameters [B s ] performance indicators;

[0212]

[0213] Among them, y i is the actual observed value;

[0214] is based on the parameter [B s ]’s model prediction value;

[0215] N is the number of observations;

[0216] According to the fitness function Fit, the fitness function value of the layout scheme is calculated;

[0217] Get the minimum fitness function value, and [B s ] In the layout scheme that meets the preset range;

[0218] Output the obtained layout scheme.

[0219] According to the layout plan, the inertial exciter 1 is arranged at the response point.

[0220] The linear active disturbance rejection controller is used to execute the linear active disturbance rejection vibration control algorithm, estimate and compensate the total disturbance, and control the operation of the inertial actuator 1;

[0221] The parameters in the linear anti-disturbance vibration control algorithm are adjusted according to the layout plan. After adjusting the parameters, the vibration isolation effect is tested. If the effect is good, it can be used for active noise reduction. If the effect is not good, the layout is re-performed.

[0222] The total disturbance includes internal disturbance and external disturbance;

[0223] In this embodiment, the linear active disturbance rejection controller (LADRC) executes the linear active disturbance rejection vibration control algorithm. The LADRC usually consists of three parts: a trajectory planner, a state observer and a controller. For a multi-degree-of-freedom system, each degree of freedom usually needs to be designed separately. For vibration control, there is no need for a trajectory planner, and it is only necessary to reduce the vibration signal to 0. Therefore, the trajectory planner is not set in this solution;

[0224] Specifically, the state space model of an n-degree-of-freedom vibration system is expressed as:

[0225]

[0226] Among them, M∈n×n is the mass matrix of the controlled structure;

[0227] C∈n×n is the damping matrix of the controlled structure;

[0228] K∈n×n is the stiffness matrix of the controlled structure;

[0229] is the acceleration response vector of the structure;

[0230] is the acceleration response vector of the structure;

[0231] x(t)∈n is the displacement response vector of the structure;

[0232] n is the modal order;

[0233] F c (t)∈r is the disturbance force vector.

[0234] Convert the above equation into state space form:

[0235]

[0236] in, is the state vector;

[0237] U(t) is the control input;

[0238] d(t) is the external disturbance;

[0239] A, B, and E are system matrices.

[0240] Determine the control objective: reduce the displacement x(t) and suppress the external disturbance d(t);

[0241] To set up the Extended State Observer (ESO):

[0242] ESO is used to estimate the system state and total disturbance;

[0243]

[0244] y(t)=C e Z(t)

[0245] in, C e =[C 0].

[0246] The design of ESO involves choosing an appropriate observer gain matrix L to make the observation error as small as possible.

[0247] Set up the controller:

[0248] The LADRC control law is set as:

[0249] U(t)=U 0 (t)-d(t);

[0250] U 0 (t) is the basic control input based on the estimated state, and d(t) is the total disturbance estimated by the ESO.

[0251] Basic control input settings:

[0252] Design the basic control input using proportional-derivative (PD) control:

[0253] U 0 (t) = K p x(t)+K d x(t);

[0254] Among them, K p is the proportional gain matrix, K d is the differential gain matrix.

[0255] Adjust the ESO gain matrix L and controller gain K p and K d Optimize system performance; the adjustment method can use trial and error, frequency domain analysis or optimization algorithm to determine these parameters.

[0256] The above is only an embodiment of the present invention. The common sense such as the known specific structure and characteristics in the scheme is not described in detail here. The ordinary technicians in the relevant field know all the common technical knowledge in the technical field of the invention before the application date or priority date, can obtain all the existing technologies in the field, and have the ability to apply the conventional experimental means before that date. The ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent. The protection scope required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. An adaptive active noise suppression control method for power equipment, characterized in that: include: S1. Perform finite element analysis on the transformer housing to obtain the modal vibration shape of the transformer housing and the position of the response point whose response is greater than the preset response requirement, extract the mass matrix and the stiffness matrix, and model the controlled structure; wherein the controlled structure is the transformer housing; S2. According to the acquired response point positions, a layout scheme is generated to arrange the inertial exciter at the response point, and the layout scheme is optimized to output a layout scheme that meets the vibration effect. According to the layout scheme, the inertial exciter is arranged at the response point; S3, using linear anti-disturbance vibration control algorithm to estimate and compensate for total disturbances and control the operation of the inertial exciter; S4, judging whether the vibration isolation effect of controlling the inertial exciter meets the preset vibration isolation requirement, if so, adopting the current layout scheme and parameter setting to perform active noise reduction; if not, executing S2 to re-optimize the layout scheme.

2. The adaptive active noise suppression control method for electric power equipment according to claim 1, characterized in that: The finite element analysis of the transformer housing is performed to obtain the dynamic model of the transformer system: Where: M∈n×n is the mass matrix of the controlled structure; C∈n×n is the damping matrix of the controlled structure; K∈n×n is the stiffness matrix of the controlled structure; is the acceleration response vector of the structure; is the acceleration response vector of the structure; x(t)∈n is the displacement response vector of the structure; n is the modal order; D s ∈n×r is the disturbance force position matrix; F(t)∈r is the disturbance force vector; B s ∈n×p is the position matrix of the inertial actuator; the inertial actuator is the actuator; p is the number of inertial exciters; U(t)∈p is the p-dimensional control force vector.

3. The adaptive active noise suppression control method for electric power equipment according to claim 2, characterized in that: The S2 comprises: S201, according to safety requirements, using the optimization configuration criteria of structural response control, generating several layout schemes for arranging inertial exciters at response points; S202, construct a fitness function Fit for evaluating parameters [B s ] performance indicators; Among them, y i is the actual observed value; is based on the parameter [B s ]’s model prediction value; N is the number of observations; S203, calculating the fitness function value of the layout solution according to the fitness function Fit; S204, obtain the minimum fitness function value, and [B s ] In the layout scheme that meets the preset range; S205: Output the obtained layout solution.

4. The adaptive active noise suppression control method for electric power equipment according to claim 3, characterized in that: The execution of the linear active disturbance rejection vibration control algorithm includes: S301. For an n-degree-of-freedom vibration system, its state space model is expressed as: Among them, F c (t)∈r is the disturbance force vector. Convert the above equation into state space form: in, is the state vector; U(t) is the control input; d(t) is the external disturbance; A, B, and E are system matrices; S302, determine the control target: reduce the displacement x(t) and suppress the external disturbance d(t); S303, setting the extended state observer ESO: ESO is used to estimate the system state and total disturbance; y(t)=C e Z(t) in, C e =[C 0]; S304, setting controller: The LADRC control law is set as: U(t)=U0(t)-d(t); U0(t) is the basic control input based on the estimated state, d(t) is the total disturbance estimated by the ESO; S305, basic control input settings: Design the basic control input using proportional-derivative (PD) control: U0(t)=K p x(t)+K d x(t); Among them, K p is the proportional gain matrix, K d is the differential gain matrix; S306: Adjust the ESO gain matrix L and controller gain K p and K d Optimize system performance.

5. The adaptive active noise suppression control method for electric power equipment according to claim 4, characterized in that: Adjust the ESO gain matrix L and controller gain K p and K d The methods include trial and error, frequency domain analysis or optimization algorithm.

6. An adaptive active noise suppression control system for electric power equipment, characterized in that: include: Inertial actuator and linear active disturbance rejection controller; The inertial exciter is arranged on the transformer housing according to the layout scheme; The layout scheme is obtained by: performing finite element analysis on the transformer shell, obtaining the modal vibration shape of the transformer shell, and the position of the response point whose response is greater than the preset response requirement, extracting the mass matrix and the stiffness matrix, and modeling the controlled structure; according to the obtained response point position, generating a layout scheme in which the inertial exciter is arranged on the response point, and optimizing the layout scheme to obtain a layout scheme that meets the vibration effect; The linear active disturbance rejection controller is used to execute the linear active disturbance rejection vibration control algorithm, estimate and compensate the total disturbance, and control the operation of the inertial exciter; and if the vibration isolation effect of the linear active disturbance rejection controller controlling the operation of the inertial exciter does not meet the preset vibration isolation requirements, the layout plan is re-optimized.

7. The adaptive active noise suppression control system for electric power equipment according to claim 6, characterized in that: The finite element analysis of the transformer housing is performed to obtain the dynamic model of the transformer system: Where: M∈n×n is the mass matrix of the controlled structure; C∈n×n is the damping matrix of the controlled structure; K∈n×n is the stiffness matrix of the controlled structure; is the acceleration response vector of the structure; is the acceleration response vector of the structure; x(t)∈n is the displacement response vector of the structure; n is the modal order; D s ∈n×r is the disturbance force position matrix; F(t)∈r is the disturbance force vector; B s ∈n×p is the position matrix of the inertial actuator; the inertial actuator is the actuator; p is the number of inertial exciters; U(t)∈p is the p-dimensional control force vector.

8. The adaptive active noise suppression control system for electric power equipment according to claim 7, characterized in that: The method of generating a layout scheme for arranging the inertial exciter at the response point according to the acquired response point position, and optimizing the layout scheme to obtain a layout scheme that meets the vibration effect includes: According to safety requirements, the optimal configuration criteria of structural response control are used to generate several layout schemes for placing inertial actuators on response points. Construct a fitness function Fit to evaluate the parameters [B s ] performance indicators; Among them, y i is the actual observed value; is based on the parameter [B s ]’s model prediction value; N is the number of observations; According to the fitness function Fit, the fitness function value of the layout scheme is calculated; Get the minimum fitness function value, and [B s ] In the layout scheme that meets the preset range; Output the obtained layout solution.

9. The adaptive active noise suppression control system for electric power equipment according to claim 8, characterized in that: The linear active disturbance rejection controller is used to execute the linear active disturbance rejection vibration control algorithm, estimate and compensate the total disturbance, and control the operation of the inertial actuator, including: For an n-degree-of-freedom vibration system, its state space model is expressed as: Among them, F c (t)∈r is the disturbance force vector. Convert the above equation into state space form: in, is the state vector; U(t) is the control input; d(t) is the external disturbance; A, B, and E are system matrices; Determine the control objective: reduce the displacement x(t) and suppress the external disturbance d(t); To set up the Extended State Observer ESO: ESO is used to estimate the system state and total disturbance; y(t)=C e Z(t) in, C e =[C 0]; Set up the controller: The LADRC control law is set as: U(t)=U0(t)-d(t); U0(t) is the basic control input based on the estimated state, d(t) is the total disturbance estimated by the ESO; Basic control input settings: Design the basic control input using proportional-derivative (PD) control: U0(t)=K p x(t)+K d x(t); Among them, K p is the proportional gain matrix, K d is the differential gain matrix; Adjust the ESO gain matrix L and controller gain K p and K d Optimize system performance.

10. The adaptive active noise suppression control system for electric power equipment according to claim 9, characterized in that: Adjust the ESO gain matrix L and controller gain K p and K d The methods include trial and error, frequency domain analysis or optimization algorithm.

Images