2.5.2. Transfer matrices for multi-duct perforated elements When perforate tubes are present in the muffler, the extracted discrete volume segments, along with the corresponding surfacehierarchical equation forthat length segmentcontaining the perforated tube(s) contain information about which volume segments are connected through the perforated tube(s). This information is used to generate a 2p · 2p (where p is the number of interacting ducts) transfer matrix [20] relating the upstream and downstream state variables for all those interacting ducts. Then, the elements of this 2p · 2p transfer matrix are entered into proper location of the final system matrix using a similar logic as the one explained in the previous section (see Eqs. (7) and (8) for formulation).
2.5.3. Closed-end cavity impedance equations While traversing the data-structure for all the volume segments, whenever a closedended cavityis found,two more state variables getadded tothe system matrix. Inanearlier, related work by the authors [18], this case was being handled by introducing a hypothetical long tube with arbitrarily small area of cross-section between supposedly closed ends. The referred method was not very elegant, as with an additional tube the number of variables is increased. Moreover, the small cross-sections introduced tend to make the system matrix ill-conditioned. In the present scheme, however, that difficulty has been removed. The state variables at the open end of the cavity are related through the impedance: SV4ðn1Þ2mþ1 SV4ðn1Þ2mþ2 ¼jYn cotðknLnÞ; ð9Þ where the cavity under consideration is the nth element of the data-structure and Yn is the characteristic impedance of the nth element. This equation is entered into the system matrix by the following equations: SYSMATð4ðn1Þ2mþ1;4ðn1Þ2mþ1Þ¼1; SYSMATð4ðn1Þ2mþ1;4ðn1Þ2mþ2Þ¼jYn cotðknLnÞ: ð10Þ This finishes the process of equation writing for the branches. The next two steps deal with the connectivity of these volume elements for applying the final set of boundary conditions.
2.5.4. Pressure equilibrium equations The pressure equilibrium equations are written at every junction of the volume elements. At every junction, the pressure variable is set equal in all the inlet/outlet of the volume elements. The pressure equations are entered into the system matrix according to the following equations: SYSMATðEN;4ðn1Þ2mþrÞ¼1; SYSMATðEN;4ðp1Þ2qþsÞ¼ 1; ð11Þ where n is the serial number of the volume element, which is participating at any particular junction and p traverses over the serial numbers of all the other volume elements which are joined at that junction. Variables m and q denote the number of closed-ended cavities occurring before n and p in the data-structure, whereas r and s can take the value of either 1 or 3 for representing the pressure state variables at the inlet or the outlet of any participating volume element. The row number, EN, is chosen (by checking empty rows in the SYSMAT matrix) to fill-up the unused rows of the generated SYSMAT matrix in the