JP4676157B2 - Numerical calculation method - Google Patents
Numerical calculation method Download PDFInfo
- Publication number
- JP4676157B2 JP4676157B2 JP2004103817A JP2004103817A JP4676157B2 JP 4676157 B2 JP4676157 B2 JP 4676157B2 JP 2004103817 A JP2004103817 A JP 2004103817A JP 2004103817 A JP2004103817 A JP 2004103817A JP 4676157 B2 JP4676157 B2 JP 4676157B2
- Authority
- JP
- Japan
- Prior art keywords
- cell
- cells
- calculation
- electromagnetic field
- calculation method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Description
本発明は、科学技術計算およびコンピュータ支援設計に関する分野において、コンピュータを用いた数値計算の所要時間を著しく短縮することを可能にする数値計算法に関するものである。 The present invention relates to a numerical calculation method that makes it possible to significantly shorten the time required for numerical calculation using a computer in the fields of scientific and technical calculation and computer-aided design.
広範囲の大規模な電磁界分布を計算する際、部分的に高精度な計算結果が必要とされる場合がある。例えば図１のように部屋の中に発信源があり、（ア）アンテナ近傍の高精度な電磁界分布、及び（イ）部屋内全体の電磁界分布（但し（ア）ほどの精度は必要ない）の両者を求め、なおかつできるだけ短時間で計算しなければならない場合などである。
このような場合の最も一般的な計算方法は、アンテナ近傍の要求精度を満たす十分に細かいセルで部屋全体の領域を分割して計算することである。この計算方法によれば、前記（ア）、（イ）両者の要求を満たすことができる。しかしながら、結果の算出に多大な計算時間を要するという問題がある。
そこで、従来、前記電磁界分布の計算にはサブグリッド法（例えば、非特許文献１参照）と呼ばれる計算方法が採用されている。このサブグリッド法は、高精度の結果が必要な部分は細かいセルを設定し、それ以外の領域はそれよりも粗いセルを設定することで全セル数を少なくして、全体の計算時間を短縮するものである。
When calculating a large-scale electromagnetic field distribution over a wide range, a partially highly accurate calculation result may be required. For example, as shown in FIG. 1, there is a transmission source in the room, (a) a highly accurate electromagnetic field distribution near the antenna, and (b) an entire electromagnetic field distribution in the room (however, accuracy as high as (a) is not required. ) Both of which must be calculated in as short a time as possible.
In such a case, the most common calculation method is to divide and calculate the area of the entire room with sufficiently fine cells that satisfy the required accuracy near the antenna. According to this calculation method, both the requirements (a) and (b) can be satisfied. However, there is a problem that it takes a lot of calculation time to calculate the result.
Therefore, conventionally, a calculation method called a subgrid method (see, for example, Non-Patent Document 1) has been adopted for the calculation of the electromagnetic field distribution. This sub-grid method reduces the total calculation time by setting fine cells in areas where high-precision results are required, and setting coarser cells in other areas, thereby reducing the total number of cells. To do.
しかしながら、このサブグリッド法には、（ａ）一つの計算プログラム中で大きさの異なる２つのセルを取り扱わなければならないためプログラムそのものが複雑で長くなる、（ｂ）細かいグリッドと粗いグリッドの界面では電磁界が不連続になるために報告されているサブグリッド法ではいずれも２万ステップ程度計算すると電磁界が発散し有効な計算結果が得られない、という２つの大きな問題がある。
即ち、アンテナ構造の計算精度と全体の計算時間を短縮するためにサブグリッド法を用いると、プログラムが複雑になることに加えて一定時間以上計算すると解が発散するというもので、このことは現在まで解決されていない。
However, in this subgrid method, (a) the program itself is complicated and long because two cells having different sizes must be handled in one calculation program, and (b) at the interface between a fine grid and a coarse grid. The reported subgrid methods due to the discontinuity of the electromagnetic field have two major problems in that if the calculation is about 20,000 steps, the electromagnetic field diverges and an effective calculation result cannot be obtained.
In other words, if the subgrid method is used to reduce the calculation accuracy of the antenna structure and the overall calculation time, the program will become complicated, and the solution will diverge if calculated over a certain time. It is not solved until.
本発明は、前記サブグリッド法の短所である計算精度と全体の計算時間の短縮という相反する用件を満足しながら、それと同時に、長時間電磁界計算を続けても解が発散しない計算方法を提供するものである。 The present invention satisfies the contradictory requirements of calculation accuracy and overall calculation time, which are disadvantages of the subgrid method, and at the same time, a calculation method in which the solution does not diverge even if the electromagnetic field calculation is continued for a long time. It is to provide.
請求項１記載の発明は、コンピュータが、格子状に配置された節点によって空間内を同一寸法の矩形状のセルに分割して、各セルに電界と磁界が割り振られるＹｅｅ格子を用いて、空間内の電磁界分布を求める数値計算方法に於いて、前記コンピュータのＣＰＵは、時間領域差分法を用いて給電点を含む領域を第１のセルに分割して電磁界分布を計算する１回目の計算を行い、前記第１のセルから奇数ｎ個おきに抽出したセルの各時間ステップの電界と磁界の値をファイルに保存し、前記第１のセルによって分割された領域を含む領域に、前記第１のセルと相似で前記第１のセルの（ｎ＋１）倍の寸法を有する第２のセルを設定し、各時間ステップにおいて、前記第２のセルの内、前記奇数ｎ個おきに抽出したセルを内包するセルには前記ファイルに保存された同じ時間ステップの電界と磁界の値を代入して、時間領域差分法を用いて前記第２のセルによって分割された空間の電磁界分布を計算する２回目の計算を行うことを特徴とする数値計算方法である。
According to the first aspect of the present invention, the computer divides the space into rectangular cells of the same size by nodes arranged in a grid, and uses a Yee grid in which an electric field and a magnetic field are allocated to each cell. In the numerical calculation method for calculating the electromagnetic field distribution in the computer, the CPU of the computer calculates the electromagnetic field distribution by dividing the region including the feeding point into the first cells by using the time domain difference method. Performing calculation, storing the electric field and magnetic field values of each time step of the cells extracted from the first cell every odd number n in a file, and including the region divided by the first cell in the region A second cell similar to the first cell and having a size (n + 1) times that of the first cell is set, and every second odd number of the second cells is extracted at each time step . The cell that contains the cell By substituting the electric and magnetic fields the values of the same time step stored in yl, perform computations second for calculating the electromagnetic field distribution of the divided space by the second cell by using the FDTD method Is a numerical calculation method characterized by
請求項２記載の発明は、コンピュータが、格子状に配置された節点によって空間内を同一寸法の矩形状のセルに分割して、各セルに電界と磁界が割り振られるＹｅｅ格子を用いて、空間内の電磁界分布を求める数値計算方法に於いて、前記コンピュータのＣＰＵは、時間領域差分法を用いて給電点を含む領域を第１のセルに分割して電磁界分布を計算する１回目の計算を行い、前記第１のセルの各時間ステップの電界と磁界の値をファイルに保存し、前記第１のセルによって分割された領域に含まれる領域を、前記第１のセルと相似で前記第１のセルの１／ｍ（ｍは偶数）の寸法を有する第２のセルに分割し、各時間ステップにおいて、前記第２のセルからｍ−１個おきに抽出したセルに、前記抽出したセルが含まれる前記第１のセルに係る前記ファイルに保存された同じ時間ステップの電界と磁界の値を代入して、時間領域差分法を用いて前記第２のセルによって分割された空間の電磁界分布を計算する２回目の計算を行うことを特徴とする数値計算方法である。
According to the second aspect of the present invention, the computer divides the space into rectangular cells of the same size by the nodes arranged in a grid, and uses a Yee grid in which an electric field and a magnetic field are allocated to each cell. In the numerical calculation method for calculating the electromagnetic field distribution in the computer, the CPU of the computer calculates the electromagnetic field distribution by dividing the region including the feeding point into the first cells by using the time domain difference method. Performing calculation, storing the electric field and magnetic field values of each time step of the first cell in a file, and the region included in the region divided by the first cell is similar to the first cell The first cell is divided into second cells having dimensions of 1 / m (m is an even number), and the cells are extracted into cells extracted every m-1 from the second cells at each time step . Related to the first cell in which the cell is included By substituting the electric and magnetic fields the values of the same time step stored in the serial file, perform a second calculation for calculating the electromagnetic field distribution of the divided space by the second cell by using the FDTD method It is the numerical calculation method characterized by this.
請求項３記載の発明は、請求項１に記載の数値計算方法または請求項２に記載の数値計算方法のいずれかを実行し、アンテナ近傍の電磁界分布及び全体の電磁界分布を計算することを特徴とする数値計算方法である。
The invention according to claim 3 executes either the numerical calculation method according to claim 1 or the numerical calculation method according to claim 2 to calculate the electromagnetic field distribution near the antenna and the entire electromagnetic field distribution. Is a numerical calculation method characterized by
請求項４記載の発明は、前記Ｙｅｅ格子は、或る時刻で各セルに割り振られた電界と磁界が存在する場合、奇数ｎ個おきにセルを選び、それらのセルの間に隣接２項間漸化式を適用して計算しても結果が変わらない性質を有することを特徴とする請求項１乃至請求項３のいずれかに記載の数値計算方法である。
According to a fourth aspect of the present invention, when there is an electric field and a magnetic field allocated to each cell at a certain time, the Yee lattice selects cells every odd number n, and between the adjacent two terms between the cells. The numerical calculation method according to claim 1, wherein even if calculation is performed by applying a recurrence formula, the result does not change .
電磁界分布の計算などの数値計算において、部分的に高い精度を実現しつつ、大規模範囲の計算を従来のサブグリッド法に比べて簡潔、且つ発散させずに短時間で計算することができるようになり、工業上顕著な効果を奏するものである。 In numerical calculations such as the calculation of electromagnetic field distribution, it is possible to calculate a large-scale range more concisely and in less time than the conventional subgrid method, while partially achieving high accuracy. As a result, there are significant industrial effects.
本発明の特徴とするところは、時間領域差分法において広く使用されるＹｅｅ格子の性質に着目し、電磁界分布を計算する際に用いる数式１〜数式６の６つの隣接２項間漸化式を給電開始時刻ｔ＝０［ｓｅｃ］からΔｔ［ｓｅｃ］毎に時間を進めながら、繰り返し計算して各時刻の電界および磁界の分布を計算するもので、全ての電界Ｅ（Ｅｘ、Ｅｙ、Ｅｚ）と磁界Ｈ（Ｈｘ、Ｈｙ、Ｈｚ）が互いの差分によって表されていることにある。 The feature of the present invention is that it pays attention to the property of the Yee lattice widely used in the time domain difference method, and the recurrence formula between six adjacent binomials of Formula 1 to Formula 6 used when calculating the electromagnetic field distribution. Are calculated repeatedly every time Δt [sec] from the power supply start time t = 0 [sec] to calculate the electric field and magnetic field distribution at each time, and all electric fields E (Ex, Ey, Ez ) And the magnetic field H (Hx, Hy, Hz) are represented by the difference between each other.
図２の「（ａ）セルサイズの拡大」において、本発明に係る計算手順の一つが示されている。本発明に係る計算手順の一つは、（１）給電点を含む細かいセルを対象として１回目の計算を行う。（２）例えば１個おきにセル値を抽出する。（３）等価な粗いセルを対象として２回目の計算を行う。一般には、給電点を含む領域を第１のセルに分割して電磁界分布を計算した後、前記第１のセルから奇数ｎ個おきに抽出したセルの計算結果を、前記第１のセルと相似で前記第１のセルの（ｎ＋１）倍の寸法を有し、前記奇数ｎ個おきに抽出したセルが複数のセルにまたがらないように配置された第２のセルを設定し、前記第２のセルの内、前記奇数ｎ個おきに抽出したセルを内包するセルには前記奇数ｎ個おきに抽出したセルの計算結果を代入して、時間領域差分法を用いて前記第２のセルによって分割された空間の電磁界分布を計算する。
また、図２の「（ｂ）セルサイズの縮小」において、本発明に係る計算手順の一つが示されている。本発明に係る計算手順の一つは、（１）粗いセルを対象として１回目の計算を行う。（２）セルを偶数ｍ個に分割する。（３）ｍ−１個おきに１回目の計算値を設定して給電点とし、細かいセルを対象として２回目の計算を行う。一般には、給電点を含む領域を第１のセルに分割して電磁界分布を計算した後、前記第１のセルを、前記第１のセルと相似で前記第１のセルの１／ｍ（ｍは偶数）の寸法を有する第２のセルに分割し、前記第２のセルからｍ−１個おきに抽出したセルに、前記抽出したセルが含まれる前記第１のセルの計算結果を代入して、時間領域差分法を用いて前記第２のセルによって分割された空間の電磁界分布を計算する。
これは、或る時刻で各セルに割り振られた電界Ｅと磁界Ｈが存在する場合、奇数ｎ個飛びにセルを選び、それらのセルの間に上記の数式１〜数式６の漸化式を適用して計算しても結果は変わらないという性質によるものである。即ち、数式１〜数式３によって磁界Ｈの差分から計算される電界Ｅの値を持つセルは、数式４〜数式６で磁界を計算するために使われる電界の値を持つセルに一致するからである。但し、ｎ個飛びにセルを選んだ場合にはΔｘ、Δｙ、Δｚを（ｎ+１）倍するものとする。
なお、前記性質は偶数ｎ個飛びにセルを抽出した場合には、差分の値に一致する値を持つセルが存在しないため物理的に意味の無い計算になる。
In “(a) Expansion of cell size” in FIG. 2, one of the calculation procedures according to the present invention is shown. One of the calculation procedures according to the present invention is (1) performing a first calculation for a fine cell including a feeding point. (2) For example, every other cell value is extracted. (3) The second calculation is performed on an equivalent coarse cell. In general, after dividing the region including the feeding point into the first cells and calculating the electromagnetic field distribution, the calculation result of the cells extracted every odd number n from the first cell is calculated as the first cell. A second cell that is similar and has a size (n + 1) times that of the first cell, and is arranged so that the odd-numbered n extracted cells do not span a plurality of cells; Among the two cells, the calculation result of the odd-numbered n cells is substituted into a cell containing the odd-numbered n-th extracted cell, and the second cell is calculated using a time domain difference method. The electromagnetic field distribution of the space divided by is calculated.
Also, one of the calculation procedures according to the present invention is shown in “(b) Reduction of cell size” in FIG. One of the calculation procedures according to the present invention is (1) performing a first calculation for a rough cell. (2) Divide the cell into an even number m. (3) The first calculation value is set every m−1 and used as a feeding point, and the second calculation is performed for a fine cell. In general, after calculating the electromagnetic field distribution by dividing the region including the feeding point into the first cells, the first cell is similar to the first cell and is 1 / m ( m is an even number)), and the calculation result of the first cell including the extracted cell is substituted for every m−1 cells extracted from the second cell. Then, the electromagnetic field distribution of the space divided by the second cell is calculated using the time domain difference method.
This means that if there is an electric field E and a magnetic field H allocated to each cell at a certain time, an odd number n cells are selected and the recurrence formulas of the above formulas 1 to 6 are set between these cells. This is because the result does not change even if applied and calculated. That is, the cell having the value of the electric field E calculated from the difference of the magnetic field H according to Equations 1 to 3 matches the cell having the value of the electric field used for calculating the magnetic field according to Equations 4 to 6. is there. However, when n cells are skipped, Δx, Δy, and Δz are multiplied by (n + 1).
Note that, when the cell is extracted in an even number n, the above property is a physically meaningless calculation because there is no cell having a value that matches the difference value.
前記理由から、１回目の計算結果を用い、２回目により大きなセルを用いて同じ問題の電磁界分布を再現することができる。本発明はこの性質を用い計算時間の短縮を図るもので、計算技術上の３つの利点を有する。
先ず第１に、大きな計算領域を取り扱う場合、計算時間が（ｎ＋１）−３のオーダーで短縮できる。即ち、時間領域差分法の計算時間はセル数に比例することから、２回目の計算では（ｎ＋１）倍のセルで計算するので、たとえ１回目と同じ大きさの領域を計算するのであってもセル数は３乗分の１になる。これは、数値計算にとって非常に大きな時間短縮が図れることを意味している。
For the above reason, the electromagnetic field distribution of the same problem can be reproduced by using the calculation result of the first time and using a larger cell for the second time. The present invention uses this property to shorten the calculation time, and has three advantages in calculation technology.
First, when a large calculation area is handled, the calculation time can be shortened on the order of (n + 1) −3 . That is, since the calculation time of the time domain difference method is proportional to the number of cells, the calculation is performed with (n + 1) times as many cells in the second calculation, so even if an area having the same size as the first calculation is calculated. The number of cells is 1/3. This means that the time can be greatly reduced for numerical calculations.
第２に、計算プログラムを簡潔にでき、その作成が容易である。このことは、１回目、２回目どちらの計算もそれぞれの回で取り扱っているセルの大きさは1種類だけであることからきており、従来のサブグリッド法のように同じ計算プログラムの中で大きさの異なるセルを取り扱いながら、界面での電磁界の連続性を考慮して計算するという複雑なアルゴリズムは必要なく、計算プログラムを非常に簡潔にできる。したがって、計算の信頼性は向上し、保守も容易である。なお、セルの大きさの変更は単なるパラメータの書き換えのみであるので容易である。 Secondly, the calculation program can be simplified and the creation thereof is easy. This is because both the first and second calculations are handled in only one type of cell, and it is large in the same calculation program as in the conventional subgrid method. A complicated algorithm of calculating in consideration of the continuity of the electromagnetic field at the interface while handling different cells is not necessary, and the calculation program can be made very simple. Therefore, the calculation reliability is improved and maintenance is easy. Note that the cell size can be easily changed by simply rewriting the parameters.
第３に、電磁界計算が発散しない。本発明では２回目の計算において、サイズの異なるセルの界面が存在しないため、本質的に電磁界の発散は起こり得ない。従来のサブグリッド法のように高々２万ステップでの発散を恐れることなく、必要とあらばどのような長時間でもシミュレーションを続けることができる。 Third, electromagnetic field calculations do not diverge. In the present invention, in the second calculation, since there is no interface between cells having different sizes, electromagnetic field divergence essentially cannot occur. The simulation can be continued for any long time if necessary without fear of divergence in 20,000 steps at most as in the conventional subgrid method.
以上は、時間領域差分法を用いて計算する場合について説明したが、本発明は時間領域差分法を用いた計算に限定されるものではなく、１回目の計算方法として有限要素法や境界要素法、モーメント法等の格子状に分割して電磁界を出力する計算方法ならばどのような計算方法とも組み合わせて計算することができる。 In the above, the case where the calculation is performed using the time domain difference method has been described, but the present invention is not limited to the calculation using the time domain difference method, and the finite element method and the boundary element method are used as the first calculation method. Any calculation method can be used in combination with any calculation method that outputs an electromagnetic field divided into a grid, such as the moment method.
以下に本発明を２次元問題の電磁界分布の解析に適用した実施例を用いて説明する。
図３に示す横１７．６ｍ、高さ１０．０ｍの部屋の中央で高さ４．４ｍの位置に１．２ＧＨｚで動作するダイポールアンテナを設置する。このときのダイポールアンテナ近傍の２．０ｍ四方領域の電界と部屋全体の電界分布を計算する。計算に使用したのは、動作周波数２．０ＧＨｚのＣＰＵ、１ＧＢのメモリーを実装したパーソナルコンピュータを５台並列に接続したＰＣクラスターを用い、１回目、２回目共に時間領域差分法で計算を行った。
The present invention will be described below with reference to an embodiment in which the present invention is applied to analysis of electromagnetic field distribution of a two-dimensional problem.
A dipole antenna that operates at 1.2 GHz is installed at a position of 4.4 m in the center of a room having a width of 17.6 m and a height of 10.0 m shown in FIG. At this time, the electric field in the 2.0 m square area near the dipole antenna and the electric field distribution of the entire room are calculated. For the calculation, we used a PC cluster in which five personal computers equipped with a CPU with an operating frequency of 2.0 GHz and a 1 GB memory were connected in parallel. .
先ず１回目の近傍電界の計算にはΔｘ＝Δｙ＝１０ｍｍの細かいセルを用いて２．０ｍ四方の領域を分割する。1回目の計算を行いながら、セル３個おきに電界と磁界の値をファイルに保存する。
図４は近傍電界の計算結果である。電界強度の絶対値に対して等高線表示してある。アンテナ近傍の詳細な電界分布が得られていることがわかる。1回目の計算時間は１．５時間であった。
First, in the first calculation of the near electric field, a 2.0 m square region is divided using a fine cell of Δx = Δy = 10 mm. While performing the first calculation, the electric field and magnetic field values are stored in a file every three cells.
FIG. 4 shows the calculation result of the near electric field. Contour lines are displayed with respect to the absolute value of the electric field intensity. It can be seen that a detailed electric field distribution in the vicinity of the antenna is obtained. The first calculation time was 1.5 hours.
２回目の計算では、部屋全体の領域を１回目より大きいΔｘ＝Δｙ＝４０ｍｍのセルで分割する。そして、各時間ステップにおいて１回目の計算で作成したファイルから電界と磁界の値を読み出して、ダイポールアンテナが設置された領域のセルに書き込み、これらのセルを給電点として２回目の計算を行う。その計算結果を図５に示す。
近傍領域だけ見ていては分からなかったダイポールアンテナの双峰性の放射特性が明瞭に現れている。ただし、図５は図を分かり易くするために、図４に比べて等高線の間隔を大きくした。２回目の計算時間は、３時間であった。
In the second calculation, the region of the entire room is divided by cells of Δx = Δy = 40 mm, which is larger than the first time. Then, the electric field and magnetic field values are read from the file created in the first calculation at each time step, written in the cells in the area where the dipole antenna is installed, and the second calculation is performed using these cells as feed points. The calculation result is shown in FIG.
The bimodal radiation characteristics of the dipole antenna, which was not understood by looking at only the nearby region, clearly appear. However, in FIG. 5, the interval between the contour lines is made larger than that of FIG. The second calculation time was 3 hours.
本発明との比較として、２回目で計算したのと同じ領域を1回目の計算の分割幅Δｘ＝Δｙ＝１０ｍｍで計算したところ、計算時間は約５０時間であった。
この問題の場合２次元問題であり、セルの大きさは４倍違うので、本発明によって期待される計算時間の短縮は（１/４）の２乗で１／１６となる。実際には５０時間かるところを３時間で終わっているので、ほぼ予想されたとおりの計算時間の短縮が実現していることが確認された。
As a comparison with the present invention, when the same region as that calculated in the second time was calculated with the division width Δx = Δy = 10 mm in the first calculation, the calculation time was about 50 hours.
This problem is a two-dimensional problem, and the cell size is four times different. Therefore, the reduction in calculation time expected by the present invention is 1/16 of the square of (1/4). Actually, it took 3 hours from 50 hours, so it was confirmed that the calculation time was reduced as expected.
Claims (4)
前記コンピュータのＣＰＵは、
時間領域差分法を用いて給電点を含む領域を第１のセルに分割して電磁界分布を計算する１回目の計算を行い、
前記第１のセルから奇数ｎ個おきに抽出したセルの各時間ステップの電界と磁界の値をファイルに保存し、
前記第１のセルによって分割された領域を含む領域に、前記第１のセルと相似で前記第１のセルの（ｎ＋１）倍の寸法を有する第２のセルを設定し、
各時間ステップにおいて、前記第２のセルの内、前記奇数ｎ個おきに抽出したセルを内包するセルには前記ファイルに保存された同じ時間ステップの電界と磁界の値を代入して、時間領域差分法を用いて前記第２のセルによって分割された空間の電磁界分布を計算する２回目の計算を行うことを特徴とする数値計算方法。 A numerical value by which a computer divides a space into rectangular cells of the same size by nodes arranged in a grid and obtains an electromagnetic field distribution in the space using a Yee grid in which an electric field and a magnetic field are allocated to each cell. In the calculation method,
The CPU of the computer is
Perform a first calculation to calculate the electromagnetic field distribution by dividing the region including the feeding point into the first cells using the time domain difference method,
The electric field and magnetic field values of each time step of the cells extracted from the first cell every odd number n are stored in a file,
A second cell having a size similar to the first cell and having a size (n + 1) times the first cell is set in a region including the region divided by the first cell;
At each time step, the electric field and magnetic field values of the same time step stored in the file are substituted into the cells containing the odd-numbered n extracted cells among the second cells, and the time domain A numerical calculation method comprising performing a second calculation for calculating an electromagnetic field distribution in a space divided by the second cell using a difference method.
前記コンピュータのＣＰＵは、
時間領域差分法を用いて給電点を含む領域を第１のセルに分割して電磁界分布を計算する１回目の計算を行い、
前記第１のセルの各時間ステップの電界と磁界の値をファイルに保存し、
前記第１のセルによって分割された領域に含まれる領域を、前記第１のセルと相似で前記第１のセルの１／ｍ（ｍは偶数）の寸法を有する第２のセルに分割し、
各時間ステップにおいて、前記第２のセルからｍ−１個おきに抽出したセルに、前記抽出したセルが含まれる前記第１のセルに係る前記ファイルに保存された同じ時間ステップの電界と磁界の値を代入して、時間領域差分法を用いて前記第２のセルによって分割された空間の電磁界分布を計算する２回目の計算を行うことを特徴とする数値計算方法。 A numerical value by which a computer divides a space into rectangular cells of the same size by nodes arranged in a grid and obtains an electromagnetic field distribution in the space using a Yee grid in which an electric field and a magnetic field are allocated to each cell. In the calculation method,
The CPU of the computer is
Perform a first calculation to calculate the electromagnetic field distribution by dividing the region including the feeding point into the first cells using the time domain difference method,
Storing the electric and magnetic field values of each time step of the first cell in a file;
A region included in the region divided by the first cell is divided into second cells having a size of 1 / m (m is an even number) of the first cell, similar to the first cell;
At each time step, every m−1 cells extracted from the second cell, the electric field and magnetic field of the same time step stored in the file related to the first cell including the extracted cell. numerical calculation method and performing by assigning a value, the calculation of the second time to compute the electromagnetic field distribution of the divided space by the second cell by using the FDTD method.
The Yee lattice is calculated by selecting an odd-numbered n-th cell and applying an adjacent binomial recurrence formula between these cells when there is an electric field and a magnetic field allocated to each cell at a certain time. The numerical calculation method according to any one of claims 1 to 3, wherein the result does not change even when the result is obtained.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2004103817A JP4676157B2 (en) | 2004-03-31 | 2004-03-31 | Numerical calculation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2004103817A JP4676157B2 (en) | 2004-03-31 | 2004-03-31 | Numerical calculation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| JP2005292950A JP2005292950A (en) | 2005-10-20 |
| JP4676157B2 true JP4676157B2 (en) | 2011-04-27 |
Family
ID=35325849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| JP2004103817A Expired - Fee Related JP4676157B2 (en) | 2004-03-31 | 2004-03-31 | Numerical calculation method |
Country Status (1)
| Country | Link |
|---|---|
| JP (1) | JP4676157B2 (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104280621A (en) * | 2014-10-29 | 2015-01-14 | 中国电子科技集团公司第二十九研究所 | Regional electromagnetic environment ray propagation path parallel search method |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105760615A (en) * | 2016-03-04 | 2016-07-13 | 重庆交通大学 | System for carrying out secondary development on TransCAD based on GISDK platform |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004038774A (en) * | 2002-07-05 | 2004-02-05 | Japan Science & Technology Corp | Method and program for analyzing electromagnetic environment and recording medium recorded with the program |
-
2004
- 2004-03-31 JP JP2004103817A patent/JP4676157B2/en not_active Expired - Fee Related
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2004038774A (en) * | 2002-07-05 | 2004-02-05 | Japan Science & Technology Corp | Method and program for analyzing electromagnetic environment and recording medium recorded with the program |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104280621A (en) * | 2014-10-29 | 2015-01-14 | 中国电子科技集团公司第二十九研究所 | Regional electromagnetic environment ray propagation path parallel search method |
| CN104280621B (en) * | 2014-10-29 | 2017-03-15 | 中国电子科技集团公司第二十九研究所 | A kind of region electromagnetic environment ray propagation path parallel search method |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2005292950A (en) | 2005-10-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101029278B1 (en) | Simulation techniques | |
| JP6360192B2 (en) | Generation of unconstrained Voronoi lattices in regions containing complex internal boundaries | |
| Song et al. | Phase field simulations of coupled microstructure solidification problems via the strong form particle difference method | |
| JP2009169558A (en) | Multi-purpose optimization design support device, method and program under consideration of manufacture variation | |
| Chen et al. | Triangulated manifold meshing method preserving molecular surface topology | |
| Wingen et al. | Connection between plasma response and resonant magnetic perturbation (RMP) edge localized mode (ELM) suppression in DIII-D | |
| JP4499781B2 (en) | Electromagnetic field strength calculation method, electromagnetic field strength calculation device, control program | |
| Colella et al. | Performance and scaling of locally-structured grid methods for partial differential equations | |
| Guthaus et al. | High-performance clock mesh optimization | |
| KR20090054934A (en) | Simulation techniques | |
| EP2287757A1 (en) | Multilevel-Multigrid simulation techniques | |
| KR20180088595A (en) | Power grid healing techniques | |
| Sieglin et al. | Progress in extrapolating divertor heat fluxes towards large fusion devices | |
| JP4676157B2 (en) | Numerical calculation method | |
| Wowk et al. | An adaptive remeshing technique for predicting the growth of irregular crack fronts using p-version finite element analysis | |
| Shterenlikht et al. | Modelling fracture in heterogeneous materials on HPC systems using a hybrid MPI/Fortran coarray multi-scale CAFE framework | |
| Einaudi et al. | Energetics and 3D structure of elementary events in solar coronal heating | |
| US9773083B1 (en) | Post-placement and pre-routing processing of critical paths in a circuit design | |
| Sidiropoulos et al. | Jitpr: A framework for supporting fast application's implementation onto fpgas | |
| Luo et al. | Large-scale fixed-outline floorplanning design using convex optimization techniques | |
| US8694940B2 (en) | System and method for integrated circuit design and implementation using mixed cell libraries | |
| CN105988301B (en) | For the method and optical adjacent correction method of the coverage rate for checking test graphics library | |
| JP6108343B2 (en) | Physical quantity simulation method and physical quantity simulation system using the same | |
| Enakoutsa | An improved nonlocal Gurson model for plastic porous solids, with an application to the simulation of ductile rupture tests | |
| Cebamanos et al. | Scaling hybrid coarray/MPI miniapps on Archer |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A621 | Written request for application examination |
Free format text: JAPANESE INTERMEDIATE CODE: A621 Effective date: 20070301 |
|
| A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20100218 |
|
| A521 | Written amendment |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20100416 |
|
| RD02 | Notification of acceptance of power of attorney |
Free format text: JAPANESE INTERMEDIATE CODE: A7422 Effective date: 20100416 |
|
| RD13 | Notification of appointment of power of sub attorney |
Free format text: JAPANESE INTERMEDIATE CODE: A7433 Effective date: 20101105 |
|
| A521 | Written amendment |
Free format text: JAPANESE INTERMEDIATE CODE: A821 Effective date: 20101105 |
|
| A131 | Notification of reasons for refusal |
Free format text: JAPANESE INTERMEDIATE CODE: A131 Effective date: 20101214 |
|
| A521 | Written amendment |
Free format text: JAPANESE INTERMEDIATE CODE: A523 Effective date: 20101222 |
|
| TRDD | Decision of grant or rejection written | ||
| A01 | Written decision to grant a patent or to grant a registration (utility model) |
Free format text: JAPANESE INTERMEDIATE CODE: A01 Effective date: 20110118 |
|
| A01 | Written decision to grant a patent or to grant a registration (utility model) |
Free format text: JAPANESE INTERMEDIATE CODE: A01 |
|
| A61 | First payment of annual fees (during grant procedure) |
Free format text: JAPANESE INTERMEDIATE CODE: A61 Effective date: 20110127 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20140204 Year of fee payment: 3 |
|
| FPAY | Renewal fee payment (event date is renewal date of database) |
Free format text: PAYMENT UNTIL: 20140204 Year of fee payment: 3 |
|
| LAPS | Cancellation because of no payment of annual fees |