KR20130142827A - Numerical method for generating in-cabinet response spectra(icrs) - Google Patents

Abstract

The present invention relates to a method for generating an in-cabinet response spectrum. The method comprises: a first step for enabling an operation unit to operate a target layer response spectrum of a corresponding cabinet, and generating an artificial earthquake corresponding to the layer response spectrum by interlocking a program; a second step for inputting the shape and size of the cabinet and the weight of internal devices through an input unit, and enabling the operation unit to perform idealization by a lumped mass numerical model after dividing the cabinet into a plurality of nodes according to the height; a third step for performing an impact hammer test through an impact tester with respect to each node of the lumped mass numerical model computed in the second step, inputting the result value through the input unit, and operating the result value through the operation unit to estimate a dynamic state equation model; a fourth step for inputting an artificial earthquake value generated in the first step through the input unit, and applying the input artificial earthquake value to the dynamic state equation model of the third step through the operation unit so as to obtain in-cabinet response; and a fifth step for enabling the operation unit to convert the in-cabinet response, obtained in the fourth step, into an in-cabinet response spectrum and performing verification. [Reference numerals] (S100) Step for operating a layer response spectrum and generating an artificial earthquake;(S110) Step for idealizing a lumped mass numerical model;(S120) Step for performing an impact test and estimating a dynamic state equation model;(S130) Step for applying the artificial earthquake to the dynamic state equation model;(S140) Step for performing conversion into in-cabinet response spectrum and performing verification

Description

Numerical method for generating in-cabinet response spectra (ICRS)

The present invention relates to a method for numerically analyzing the seismic response prediction for a cabinet structure. More specifically, the dynamic characteristics of a power plant cabinet are obtained through an impact test, and then the dynamic state equation model is estimated using the numerical value. Through analysis, it is possible to predict the earthquake response of the cabinet efficiently and accurately.

Generally, power plants are equipped with many steel cabinets that house electrical or control instrumentation. Most of these are important facilities that are responsible for the safe operation and shutdown of power plants. They must maintain structural stability in the event of a strong earthquake, such as a design earthquake, and the unique functions assigned to each must be performed smoothly and without errors. . For this purpose, when the power plant is constructed, the structures must be seismic safe through seismic design, and important equipment and facilities installed inside the structure must be verified by seismic verification.

In particular, commercial operation of nuclear power plants among power plants began in the 1960s. Based on more than 40 years of experience in this industry, we now have more advanced design technologies than in the early days, and our licensing requirements have been strengthened. Since the 1970s, the need to reassess the seismic safety of equipment installed in older nuclear power plants built before the requirements of the seismic design procedure have been comparable. Work is underway to reevaluate stability and improve performance.

Among them, the seismic performance can be evaluated by constructing an analytical model based on the schematic and analyzing the seismic response. On the other hand, devices such as cabinets can be proved to be shockproof by shaking table testing or seismic response analysis.

However, since the equipment installed in the operating nuclear power plant is not allowed to be transported to the outside for the purpose of shaking table test, it is not easy to prepare the accurate dynamic analysis model for the seismic response analysis.

To diagnose the seismic safety of the internal electrical equipment of a nuclear power plant, the dynamic characteristics of them must be analyzed first. In particular, in the case of replacing all or some parts of the existing exercise equipment, it is necessary to prove whether the seismic performance requirements are met both before and after the replacement of parts, and even in this case, dynamic characteristic analysis of the device is required first.

However, since electrical devices such as cabinets or panels among nuclear power plants have not only structural characteristics of the facilities themselves, but also the connection of the components included therein, it is very important to accurately determine the rigidity and contribution mass of such facilities. it's difficult.

Therefore, these types of facilities are very difficult to accurately build a dynamic analysis model, there is a problem that it is not easy to grasp their dynamic characteristics by the existing shaking table test and analysis method.

Accordingly, the present invention has been made to solve the above problems, the problem of the present invention is the impact that can be applied in the field to the cabinet for power plant equipment, which is very difficult to actually create a dynamic analysis model for seismic performance evaluation In this paper, we propose a method to numerically generate the cabinet internal response spectrum of the cabinet by accurately estimating the dynamic state equation model that can reflect the hammer test results.

The present invention has been proposed in order to achieve the above object, the present invention includes a first step of the calculation unit calculates the target floor response spectrum of the cabinet, the artificial earthquake corresponding to the layer response spectrum by a program;

A second step of inputting a shape, dimensions, and weights of internal devices through the input unit, and separating the cabinet into a plurality of nodes according to heights, thereby idealizing a concentrated mass numerical model;

The impact hammer test is performed on each node of the concentrated mass numerical model calculated in the second step through an impact tester, the result is input through the input unit, and the calculation unit calculates the dynamic state equation model by calculating. Step 3;

Inputting the artificial earthquake value generated in the first step through the input unit, and inputting the artificial earthquake value to the dynamic state equation model of the third step by the calculating unit and performing numerical integration to obtain the internal response of the cabinet; Step 4; And

Provided is a method for creating an internal response spectrum of a cabinet including a fifth step in which the internal response of the cabinet obtained in the fourth step is numerically integrated by the calculating unit and converted into the cabinet internal response spectrum.

The present invention can be applied in the field without the movement and reinstallation of the cabinet by the method using the impact hammer test compared to the existing shaking table test, which is difficult to apply to the seismic verification of the cabinet in operation in the power plant, By using the impact hammer test result data and the dynamic state model estimation, it is possible to accurately estimate the earthquake response of the cabinet.

1 is a flow chart illustrating a method for creating an internal response spectrum of a cabinet according to the present invention.
FIG. 2 is a view showing the appearance and internal structure of a cabinet for creating a cabinet internal response spectrum shown in FIG. 1.
FIG. 3 is a schematic diagram of a computing device used in the method for creating an internal response spectrum of the cabinet shown in FIG.
FIG. 4 is a graph showing a first step of generating an artificial earthquake corresponding to the floor response spectrum of the method for creating the internal response spectrum of the cabinet shown in FIG. 1.
FIG. 5 is a diagram schematically illustrating a second step of idealizing a cabinet into a concentrated mass numerical model among methods for creating an internal response spectrum of the cabinet shown in FIG. 1.
FIG. 6 is a view showing a third step of obtaining an input load and a response at each node through an impact hammer test at each node of the concentrated mass numerical model among the methods for creating the internal response spectrum of the cabinet shown in FIG. to be.
FIG. 7 is a graph showing a third step of estimating the dynamic state equation model of the numerical model created in step 2 using the impact hammer test data among the methods for creating the internal response spectrum of the cabinet shown in FIG. 1.
8 is a diagram illustrating a fourth step of obtaining an internal response of a cabinet by applying an artificial earthquake generated in step 1 to the dynamic state equation model of the cabinet estimated in step 3 of the method for creating an internal response spectrum of the cabinet shown in FIG. A graph showing an example.
FIG. 9 is a graph illustrating an example of a fifth step of converting and verifying the internal response of the cabinet into the cabinet internal response spectrum in the method for creating the internal response spectrum of the cabinet illustrated in FIG. 1.

1 is a flowchart illustrating a prediction process of an cabinet internal response spectrum (ICRS) according to the present invention.

As shown, the numerical method of seismic response for the cabinet structure according to the present invention comprises the first step (S100) of generating an artificial earthquake corresponding to the floor response spectrum; A second step (S110) of idealizing the cabinet into a concentrated mass numerical model; A third step (S120) of estimating the dynamic state equation model through the impact hammer test at each node of the concentrated mass numerical model; A fourth step (S130) of obtaining an internal response of the cabinet by inputting and integrating the artificial earthquake value generated in the first step (S100) to the dynamic state equation model of the cabinet estimated in the third step (S120); A fifth step S140 of converting the internal response of the cabinet into the cabinet internal response spectrum is integrated.

In the numerical analysis method of the earthquake response to the cabinet structure proceeds to this step, in the first step (S100) to generate an artificial earthquake corresponding to the layer response spectrum.

Equipment subject to seismic design of nuclear power plant should be designed in consideration of seismic verification from the design stage. For this purpose, the seismic input value to be used for earthquake proofing should be specified. In general, the earthquake input to be used for earthquake proofing is presented as response spectrum.

The response spectrum (RS) is a graph showing the maximum response of the terminal inductometer according to the natural frequency and the damping ratio for the specific ground acceleration.

In addition, the response spectrum is the frequency-amplitude plane that lists the maximum value of the response time history for each earthquake motion in each one-dimensional system with different natural frequencies.

The reason for using the response spectrum in the seismic design is that once the input earthquake motion is determined, the response of the system can be known immediately.

That is, under a given input earthquake, the response of the system with each natural frequency is constant, so if we know the attenuation coefficient, the response of the system with any dominant natural frequency is given to the corresponding damping coefficient corresponding to the natural frequency in the response spectrum. Just read the response and you can get it roughly without any extra calculations.

In particular, the floor response spectrum refers to a spectrum representing the dynamic behavior of each floor due to a design earthquake in a structure composed of a plurality of floors such as a cabinet of a nuclear power plant.

In this first step (S100), as shown in FIGS. 3 and 4, first, a response spectrum spectrum 2, which is a target for the seismic design, may be created by the response spectrum calculator 12 of the calculator 10. . That is, when the input earthquake motion is input through the input unit 14, the calculation unit 10 calculates the corresponding design response spectrum 2 by the response spectrum calculation unit 12.

At this time, the horizontal axis of the design response spectrum (2) is a period, the vertical axis means the response acceleration.

When the design response spectrum 2 is created by the calculation unit 10 as described above, the artificial earthquake generating unit 16 generates the artificial earthquake 3 corresponding to the design response spectrum 2. The artificial earthquake 3 means time history acceleration.

The reason for generating the artificial earthquake is to derive a dynamic state equation model in step S120 as described below. In order to obtain a result similar to an actual earthquake, an artificial earthquake is generated and the result is input to the dynamic state equation model. To do this.

In this case, in order to generate an artificial earthquake corresponding to the design response spectrum 2, the artificial earthquake generating unit 16 uses a random vibration theory and a program.

This artificial earthquake is generated in accordance with the time history acceleration input motion in the cabinet base, which corresponds to the target response spectrum 2 shown in FIG.

In the second step (S110), the cabinet structure is idealized as a concentrated mass numerical model.

That is, as shown in Figure 5, the weight of the various devices arranged for each floor in the cabinet is grasped. In this case, the weight can be grasped through an installation drawing, a device manual, and direct measurement. Then, the determined shape and weight are input by the input unit 14, and the calculation unit 10 sets nodes according to the height of the cabinet according to the input data. In the present invention, the cabinet is set to five nodes.

In this way, after calculating the concentrated mass numerical model set to the five nodes, the third step S120 of performing an impact test by the impact test apparatus 20 is performed.

In the third step (S120), as shown in Figs. 5 and 6, by performing an impact hammer test on the five nodes to obtain the response acceleration at each node, this response acceleration to the equation calculating unit 18 Input to estimate the dynamic state model.

Thus, the process of estimating the dynamic state equation through the impact test will be described in detail as follows.

First, the equivalent earthquake load transmitted to five materials of the cabinet can be expressed by the following equation.

---------수식 1

--------- Formula 1

u _g denotes the basic acceleration as the input earthquake motion, and " 1 " M is the mass matrix of the cabinet structure.

If the equivalent earthquake load represented by Equation 1 acts on a linear dynamic system such as a cabinet, the equation of motion for this system can be expressed as follows.

----수식 2

---- Formula 2

M is the mass of the cabinet structure, C is the damping, K is the rigid mattress, and u is the displacement response vector for F (t) , the force load vector.

From Equation 2, the acceleration response may be derived as follows.

----수식 3

---- Formula 3

Equation 3 may be rearranged in a matrix form as follows.

---수식 4

--- Equation 4

-----수식 5

----- Formula 5

Equations 4 and 5 are state equations in continuous time, and can be rearranged as follows.

----수식 6

---- Formula 6

In Equation 6, x (t) is a state value at time t , y (t) is an output response acceleration, and u (t) is an input force.

In order to obtain the discrete time spatial state equations described in Equation 14 described later from Equation 6, linear time invariant homogeneous equations are considered as follows.

----수식 7

---- Formula 7

The solution to the equation is always present and can be expressed as follows.

--------수식8

-------- Formula 8

The transition matrix is the solution of the matrix differential equation

---수식9

--- Equation 9

The above formula has an explicit form.

--수식 10

Equation 10

This is the Taylor series expansion and converges to all A's.

 The above equation is considered as follows.

-----수식 11

----- Formula 11

If u (t) is continuous for all t , then

---수식12

--- Equation 12

Therefore, the discrete time-space state equation can be summarized as follows.

--수식 13

Equation 13

For the time increment step, it can be rearranged as follows.

---수식 14

--- Formula 14

x (tk), y (tk), u (tk) is the time interval k (k = 1,2, ... n) Input force, response acceleration, and measured noise vector. n is a number of periods, and A, B, C, and D are system matrices.

Discrete-time spatial state equation (SSE) as shown in Equation 14 can evaluate the relationship between the response of the structure and the impact force.

In addition, various algorithms have been developed for the evaluation of discrete time-spatial state equations (SSE). In the present invention, the N4SID proposed by Moore and Overschee is used.

This method can be applied to various subspace evaluation methods and uses simple parameters.

In describing this evaluation relationship, the phase vector x (k) of Equation 14 may be defined as a linear synthesis of the input value and the output value.

---------수식 15

--------- Formula 15

From here,

---수식16

--- Equation 16

And p (k) is the "past" for sample k.

The dimension of the "past" is represented by the number of lags N. The phase vector x (k) is computed from the data and is not specified in the physical state of the system.

After J is computed (formula below), the phase vector can be evaluated by equation (15). State-space model mattresses can be evaluated through linear regression.

----수식 17

---- Equation 17

The operation of J separates the various subspace algorithms from each other.

In the N4SID approach, J is derived from a series of geometric arguments (or linear algebraic arguments) based on a set of matrix equations for the evolution of a linear system weighted by singular value decomposition as follows.

---수식 18

--- Equation 18

f is the "conditional future" evaluated by past observations,

p (k), U ₁ , U ₂ , V ₁ , V ₂ are orthonormal matrices UU ^T = U ^T U = I, VV ^T = V ^T V , S ₁ is a singular value matrix of high degree of freedom, S ₂ Is a singular value that is small enough to be ignored diagonally.

As such, after the third step S120 is performed, the fourth step is performed to obtain an internal response of the cabinet.

In the fourth step S130, the internal response of the cabinet is obtained by inputting the artificial earthquake data generated in the first step S100 into the dynamic state equation model of the cabinet estimated in the third step S120.

In more detail, after obtaining an accurate model of the cabinet through the estimation of the dynamic state equation model, the artificial earthquake value is inputted into the dynamic state equation, and the earthquake response at one point of the interior of the cabinet is analyzed by numerical analysis such as numerical integration. Can be obtained as a function.

8 illustrates an example of obtaining an internal response of the cabinet by inputting an artificial earthquake value generated in the first step S100 to the dynamic state equation model of the cabinet estimated in the third step S120.

After obtaining an accurate model of the cabinet through dynamic state equation model estimation, an accurate response to the load input simulated through numerical analysis can be obtained through the output unit 22.

As shown, the response accelerations at the third and fourth nodes of the five nodes are shown.

In the graph of FIG. 8, the horizontal axis represents the time axis and the vertical axis represents the response acceleration.

The graph 4 on one side shows a response acceleration in a time range of 0 to 30 seconds, and the graph 5 on the other side shows an enlarged time range of 1 to 5 seconds.

In addition, the graph predicting the response acceleration is shown by the solid line on the output unit 22, and the graph showing the actual simulation result is shown by the dotted line.

As a result, it was shown that the predicted response acceleration and the actual simulated response acceleration coincide somewhat.

As such, after the fourth step S130 is completed, a fifth step S140 of converting the cabinet internal response into the response spectrum is performed.

In a fifth step S140, the simulated and predicted response acceleration is converted into an internal response spectrum (ICRS) in order to prove the effect of the procedure proposed by the response spectrum calculator 12.

That is, as a seismic response of one point inside the cabinet, the recording of the time function obtained in the fourth step is numerically integrated again using software to convert to the internal response spectrum.

9 is a graph showing a state in which the cabinet internal response is converted into the cabinet response spectrum at each of the two nodes to the five nodes.

In the graphs at each node, the horizontal axis represents the oscillation period and the vertical axis represents the response acceleration.

The solid line represents the cabinet response spectrum curve in the case of simulation, and the dotted line represents the cabinet response spectrum curve in the case of prediction.

As shown, the cabinet response spectrum in the case of simulation and the cabinet response spectrum in the case of prediction are almost identical.

Therefore, it can be seen from the comparison between the predicted ICRS of the numerical model and the simulated ICRS that the proposed algorithm is effective for predicting the ICRS of the cabinet structure.

In addition, if necessary, the accuracy of the numerical model can be improved by repeating the previous steps.

10: calculator 12: response spectrum calculator
14: input unit 16: artificial earthquake generating unit
18: equation calculation unit 20: impact test device

Claims (4)

A first step of the computing unit calculating a target floor response spectrum of the cabinet and interlocking programs to generate an artificial earthquake corresponding to the floor response spectrum;
A second step of inputting a shape, dimensions, and weights of internal devices through the input unit, and separating the cabinet into a plurality of nodes according to heights, thereby idealizing a concentrated mass numerical model;
The impact hammer test is performed on each node of the concentrated mass numerical model calculated in the second step through an impact tester, the result is input through the input unit, and the calculation unit calculates the dynamic state equation model by calculating. Step 3;
A fourth step of obtaining an internal response of the cabinet by inputting the artificial earthquake value generated in the first step through an input unit and inputting the artificial earthquake value into the dynamic state equation model of the third step by the calculating unit; Wow; And
And a fifth step of converting the internal response of the cabinet obtained in the fourth step into a cabinet internal response spectrum by numerically integrating again by the calculation unit. The method of claim 1,
And a plurality of nodes of the lumped mass numerical model in the second step. The method of claim 1,
In the case of performing an impact hammer test by the impact tester, each of the plurality of nodes is equipped with a measuring instrument capable of measuring acceleration,
A method for creating an internal response spectrum of a cabinet, characterized by obtaining the response acceleration at each node by impacting each node with a hammer. The method of claim 1,
The process of calculating the dynamic state equation model
The equivalent earthquake loads transmitted to five materials of the cabinet are expressed by the following equation,

--------- Formula 1
( u _g is the input earthquake motion, the basic acceleration, "1" is the column vector is "1", and M is the mass matrix of the cabinet structure)
When applying Equation 1 to a linear dynamic system, it can be expressed by the equation of motion of Equation 2 below,

---- Formula 2
(M is the mass of the cabinet structure, C is the damping, K is the rigid mattress, and u is the displacement response vector for F (t) , the effective load vector).
From Equation 2, the acceleration response is derived as Equation 3 below,

---- Formula 3
Rearrange Equation 3 to matrix as below,

--- Equation 4

----- Formula 5
Rearrange Equations 4 and 5 as follows,

---- Formula 6
( x (t) is the state value at time t , y (t) is the output response acceleration, u (t) is the input force)
In order to obtain the discrete-time spatial state equation described in Equation 14 to be described later from Equation 6, a linear time invariant homogeneous equation is expressed as follows,

---- Formula 7
At this time, the solution to the formula is always present, can be expressed as follows,

-------- Formula 8
Where the transition matrix is the solution of the matrix differential equation

--- Equation 9
Equation 9 has an explicit form as shown in the following equation,

Equation 10
This equation is Taylor series expansion and converges to all A's.

----- Formula 11
At this time, if u (t) is a continuous interval for all t , can be summarized as follows,

--- Equation 12
The discrete time-space equation can be summarized as

Equation 13
Equation 13 is rearranged as follows for a time increment step.

--- Formula 14
( x (tk), y (tk), u (tk) is the time interval k (k = 1,2, ... n) Input force, response acceleration, and measured noise vector. n is the number of periods, and A, B, C, and D are system matrices)

Images