Order reduction method of circuit model and related equipment

By lowering the interconnect matrix in the circuit model, the problems of low efficiency and high density of the circuit model in the prior art are solved, and the method of efficient calculation of output signals is realized, and the sparseness of the circuit model and structure.

CN119026558APending Publication Date: 2024-11-26HUAWEI TECH CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202310589361.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2023-05-23
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

The existing circuit model reduction technology is inefficient when there are many input ports, and the reduced model has high density, which cannot maintain the structure and sparsity of the original circuit model, resulting in low efficiency in calculating the output signal.

Method used

By determining the initial coupling capacitance matrix, resistance matrix and capacitance matrix of interconnected lines in the target circuit, the projection matrix of each interconnected line is calculated and the matrix is ​​lowered to obtain the reduced order circuit model and maintain the matrix The sparsity and structure of the two input ports are allocated to the corresponding matrix of each interconnect line.

Benefits of technology

The efficiency of calculating the output signal is improved and the calculation amount is reduced. The obtained down-order model has good sparsity, high down-order accuracy, small error, and maintains the structure and sparsity of the circuit model.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119026558A_ABST
    Figure CN119026558A_ABST
Patent Text Reader

Abstract

The embodiment of the invention provides an order reduction method of a circuit model and related equipment. The method comprises the steps of determining at least one initial coupling capacitance matrix and an initial resistance matrix and an initial capacitance matrix of each interconnection line according to a plurality of interconnection lines in a target circuit; determining a projection matrix of each interconnection line according to the initial resistance matrix of each interconnection line; and according to the projection matrix of each interconnection line, performing order reduction on the at least one initial coupling capacitance matrix and the initial resistance matrix and the initial capacitance matrix of each interconnection line, and determining a circuit model of the target circuit after order reduction. According to the method, order reduction can be performed on the block matrix corresponding to each interconnection line in the circuit model before order reduction of the target circuit, so that the structure of the matrix in the circuit model before order reduction and after order reduction can be kept unchanged, the calculation amount during calculation of the output signal can be reduced, the efficiency of calculation of the output signal can be improved, and the calculation cost can be reduced. And the efficiency of reducing the order of the circuit model is higher.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Embodiments of the present application relate to the field of computers, and more specifically, to a circuit model order reduction method, a computing device, a computing equipment, a chip system, and a computer-readable medium. Background Art

[0002] As multi-core and multi-functional integration of integrated circuits becomes a development trend, the scale of circuits is increasing rapidly. For a circuit with a large scale and multiple input ports, the circuit model determined according to the circuit is relatively complex. Therefore, according to the input signal of each input port and the circuit model, the complexity of calculating the output signal of each output port of the circuit is high and takes a long time. Model order reduction (MOR) technology can convert a larger circuit model into an approximate smaller circuit model, thereby reducing the amount of calculation of the circuit model when calculating the output signal. However, the existing MOR technology has a low efficiency of order reduction when the number of input ports in the circuit is large, that is, it takes a long time to obtain the reduced circuit model according to the original circuit model. In addition, the density of the reduced circuit model obtained is high, and the structure and sparsity of the original circuit model cannot be maintained, so the amount of calculation when calculating the output signal is still large, and the efficiency of calculating the output signal is low.

[0003] Therefore, how to obtain a reduced-order circuit model and improve the efficiency of calculating the output signal becomes an urgent problem to be solved. Summary of the invention

[0004] The embodiments of the present application provide a circuit model reduction method, a computing device, a computing equipment, a chip system, and a computer-readable medium, which can obtain a reduced-order circuit model and improve the efficiency of calculating output signals.

[0005] In a first aspect, a circuit model reduction method is provided. The method comprises: determining at least one initial coupling capacitance matrix, an initial resistance matrix and an initial capacitance matrix of each interconnection line according to a plurality of interconnections in a target circuit; determining a projection matrix of each interconnection line according to the initial resistance matrix of each interconnection line; and reducing the at least one initial coupling capacitance matrix, the initial resistance matrix and the initial capacitance matrix of each interconnection line according to the projection matrix of each interconnection line to determine a reduced circuit model of the target circuit.

[0006] Each interconnection line includes a plurality of nodes. Each initial coupling capacitance matrix is ​​used to indicate the coupling capacitance value between the nodes of two interconnections. The initial resistance matrix is ​​used to indicate the resistance value between the nodes in the interconnection lines corresponding to the initial resistance matrix. The initial capacitance matrix is ​​used to indicate the capacitance value between the nodes in the interconnection lines corresponding to the initial capacitance matrix. The target circuit is a resistor-capacitance (RC) circuit. The projection matrix of each interconnection line is used to reduce the order of one or more initial coupling capacitance matrices, the initial resistance matrix of the interconnection line, and the initial capacitance matrix.

[0007] In an embodiment of the present application, the computing device can determine the projection matrix of each interconnection line in units of interconnections, thereby reducing the order of the initial resistance matrix, the initial capacitance matrix, and the initial coupling capacitance matrix between the interconnections corresponding to each interconnection line, and then obtaining the reduced-order circuit model of the target circuit. In other words, the computing device can reduce the order of the block matrix corresponding to each interconnection line in the circuit model of the target circuit before the reduction, without reducing the order of the entire matrix in the circuit model before the reduction, so that the structure of the matrix in the circuit model before and after the reduction can remain unchanged. That is, the method in the embodiment of the present application can keep the sparsity of the matrix in the circuit model before and after the reduction unchanged, so that the amount of calculation when calculating the output signal can be reduced, and the efficiency of calculating the output signal can be improved. In addition, the computing device reduces the order in units of interconnections, and the reduced-order complexity caused by multiple input ports of the target circuit can be apportioned to the matrix corresponding to each interconnection line, and each interconnection line includes at least one input port, so that the efficiency of reducing the order of the circuit model is higher.

[0008] In combination with the first aspect, in certain implementations of the first aspect, the circuit model of the target circuit before reduction includes a capacitance matrix before reduction, and the circuit model of the target circuit after reduction includes a capacitance matrix after reduction, and the capacitance matrix before reduction and the capacitance matrix after reduction are block bordered diagonal structure (BBD) matrices.

[0009] In the embodiment of the present application, the capacitor matrix of the target circuit after the order reduction can maintain the BBD structure, so that the sparsity of the capacitor matrix after the order reduction is better. Since the capacitor matrix of the target circuit after the order reduction has good sparsity, the calculation cost required when using the capacitor matrix after the order reduction to calculate the output signal of the output port of the target circuit is low and the efficiency is high.

[0010] In combination with the first aspect, in certain implementations of the first aspect, the circuit model of the target circuit before order reduction includes a resistance matrix before order reduction, and the circuit model of the target circuit after order reduction includes a resistance matrix after order reduction, and the resistance matrix before order reduction and the resistance matrix after order reduction are block diagonal matrices.

[0011] In the embodiment of the present application, the resistance matrix of the target circuit after the order reduction can maintain a block diagonal structure, so that the sparsity of the resistance matrix after the order reduction is better. Since the resistance matrix of the target circuit after the order reduction has good sparsity, the calculation cost required when using the resistance matrix after the order reduction to calculate the output signal of the output port of the target circuit is low and the efficiency is high.

[0012] In combination with the first aspect, in certain implementations of the first aspect, the initial resistance matrix of each interconnect line is reduced in order according to the projection matrix of each interconnect line to determine the reduced-order resistance matrix of each interconnect line; based on the reduced-order resistance matrix of each interconnect line, a reduced-order resistance matrix is ​​determined, and the reduced-order resistance matrix of each interconnect line is the diagonal block elements in the reduced-order resistance matrix.

[0013] In the embodiment of the present application, since the initial resistance matrix of each interconnection line is small in scale, the amount of calculation required when reducing the order of the initial resistance matrix of each interconnection line is small, and the speed of obtaining the reduced-order resistance matrix of each interconnection line is fast. In addition, the computing device can directly splice the reduced-order resistance matrix of each interconnection line to determine the reduced-order resistance matrix of the target circuit without reducing the order of the larger-scale resistance matrix of the target circuit as a whole, thereby reducing the complexity of obtaining the reduced-order resistance matrix and improving the efficiency of reduction.

[0014] In combination with the first aspect, in certain implementations of the first aspect, the initial capacitance matrix of each interconnect line is reduced in order according to the projection matrix of each interconnect line to determine a reduced-order capacitance matrix of each interconnect line; at least one initial coupling capacitance matrix is ​​reduced in order according to the projection matrices of multiple interconnect lines to determine at least one reduced-order coupling capacitance matrix, and each reduced-order coupling capacitance matrix corresponds to an initial coupling capacitance matrix; and a reduced-order capacitance matrix is ​​determined according to the at least one reduced-order coupling capacitance matrix and the reduced-order capacitance matrix of each interconnect line.

[0015] In the embodiment of the present application, since the scale of each initial coupling capacitance matrix and the initial capacitance matrix is ​​small, the amount of calculation required when reducing the order of each initial coupling capacitance matrix and the initial capacitance matrix is ​​small, and the speed of obtaining the reduced-order coupling capacitance matrix and the reduced-order capacitance matrix is ​​fast. In addition, the computing device can directly splice one or more reduced-order coupling capacitance matrices and each reduced-order capacitance matrix to determine the reduced-order capacitance matrix of the target circuit without reducing the order of the larger-scale capacitance matrix of the target circuit as a whole, thereby reducing the complexity of obtaining the reduced-order capacitance matrix and improving the efficiency of reduction.

[0016] In combination with the first aspect, in some implementations of the first aspect, the first initial coupling capacitance matrix is ​​reduced in order according to the projection matrix of the first interconnect line and the projection matrix of the second interconnect line to determine the first reduced-order coupling capacitance matrix, wherein the first interconnect line and the second interconnect line are different interconnect lines among the plurality of interconnect lines, and the first initial coupling capacitance matrix is ​​used to indicate the coupling capacitance values ​​between the nodes in the first interconnect line and the second interconnect line.

[0017] In the embodiment of the present application, the computing device can reuse the projection matrices of the two interconnects to reduce the order of an initial coupling capacitance matrix, thereby obtaining a reduced-order coupling capacitance matrix. Since the computing device can reuse the projection matrix of each interconnect without recalculating the projection matrix corresponding to the entire capacitance matrix before reduction of the target circuit containing different initial coupling capacitance matrices, the cost of calculating the projection matrix can be reduced, thereby increasing the speed of reduction.

[0018] In combination with the first aspect, in certain implementations of the first aspect, the reduced-order capacitance matrix of each interconnect line is used as a diagonal block element in the reduced-order capacitance matrix; and one or more second reduced-order coupling capacitance matrices are used as block elements in the first region of the reduced-order capacitance matrix.

[0019] The second reduced-order coupling capacitor matrix corresponds to the third interconnection line and the fourth interconnection line. The third interconnection line is the interconnection line corresponding to the first diagonal block in the reduced-order capacitor matrix, and the fourth interconnection line is an interconnection line different from the third interconnection line among the plurality of interconnections. The first region is the row region or column region where the first diagonal block is located. The reduced-order capacitor matrix is ​​a symmetric matrix.

[0020] In an embodiment of the present application, the computing device can directly splice the reduced-order capacitance matrix of each interconnect line and one or more reduced-order coupling capacitance matrices to obtain a reduced-order capacitance matrix, thereby increasing the reduction speed.

[0021] In combination with the first aspect, in some implementations of the first aspect, a projection matrix of each interconnect line is determined according to an initial resistance matrix of each interconnect line; or, a projection matrix of each interconnect line is determined according to an initial resistance matrix and an initial capacitance matrix of each interconnect line; or, a projection matrix of each interconnect line is determined according to an initial input port matrix, an initial resistance matrix, and an initial capacitance matrix of each interconnect line. The initial input port matrix is ​​used to indicate an input port in the interconnect line corresponding to the initial input port matrix.

[0022] In the embodiment of the present application, the computing device can determine the projection matrix of each interconnection line in a variety of ways, so as to facilitate the subsequent reduction of the circuit model of the target circuit. The computing device can determine the projection matrix of each interconnection line according to the initial resistance matrix of each interconnection line, so that the speed of obtaining the projection matrix is ​​faster, thereby improving the speed of reduction. The computing device can also determine the projection matrix of each interconnection line according to the initial input port matrix, initial resistance matrix and initial capacitance matrix of each interconnection line, so as to improve the accuracy of reduction when using the projection matrix for reduction.

[0023] In combination with the first aspect, in certain implementations of the first aspect, an initial input port matrix and an initial output port matrix of each interconnection line are determined according to a plurality of interconnections; the initial input port matrix and the initial output port matrix of each interconnection line are respectively reduced in order according to the projection matrix of each interconnection line to determine a reduced-order input port matrix and a reduced-order output port matrix of each interconnection line; a reduced-order input port matrix is ​​determined according to the reduced-order input port matrix of each interconnection line; and a reduced-order output port matrix is ​​determined according to the reduced-order output port matrix of each interconnection line. The initial input port matrix is ​​used to indicate the input ports in the interconnection line corresponding to the initial input port matrix, and the initial output port matrix is ​​used to indicate the output ports in the interconnection line corresponding to the initial output port matrix.

[0024] In the embodiment of the present application, the computing device can reduce the order of the initial input port matrix and the initial output port matrix of each interconnection line according to the projection matrix of each interconnection line, thereby obtaining a reduced-order input port matrix and a reduced-order output port matrix. The computing device can also directly splice each reduced-order input port matrix to obtain a reduced-order input port matrix of the target circuit. The computing device can also directly splice each reduced-order output port matrix to obtain a reduced-order output port matrix of the target circuit.

[0025] In combination with the first aspect, in certain implementations of the first aspect, an output signal of each output port of the target circuit is determined based on an input signal of each input port in the target circuit and a reduced-order circuit model, and the target circuit includes at least one input port and at least one output port.

[0026] In the embodiment of the present application, the computing device can also calculate the output signal of each output port of the target circuit according to the reduced-order circuit model of the target circuit. Since the reduced-order circuit model obtained according to the method in the embodiment of the present application has good sparsity and high reduced-order accuracy, the efficiency of obtaining the output signal is high, and the error of the obtained output signal is small.

[0027] In a second aspect, an embodiment of the present application provides a computing device, which includes a unit for implementing the first aspect or any possible implementation manner of the first aspect.

[0028] In a third aspect, an embodiment of the present application provides a computing device, which includes a processor, wherein the processor is used to couple with a memory, read and execute instructions and / or program codes in the memory to execute the first aspect or any possible implementation of the first aspect.

[0029] In a fourth aspect, an embodiment of the present application provides a chip system, which includes a logic circuit, which is used to couple with an input / output interface and transmit data through the input / output interface to execute the first aspect or any possible implementation method of the first aspect.

[0030] In a fifth aspect, an embodiment of the present application provides a computer-readable medium, which stores a program code. When the program code runs on a computer, the computer executes the first aspect or any possible implementation of the first aspect.

[0031] In a sixth aspect, an embodiment of the present application provides a computer program product, which includes: a computer program code, which, when executed on a computer, enables the computer to execute the first aspect or any possible implementation of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a schematic structural block diagram of a circuit model reduction system 100 according to an embodiment of the present application.

[0033] Figure 2 It is a schematic flow chart of a circuit model reduction method according to an embodiment of the present application.

[0034] Figure 3 is a circuit diagram according to an embodiment of the present application.

[0035] Figure 4 is a schematic flowchart of a circuit model reduction method according to an embodiment of the present application.

[0036] Figure 5 is a waveform diagram of an input signal according to an embodiment of the present application.

[0037] Figure 6 is a waveform diagram of an output signal according to an embodiment of the present application.

[0038] Figure 7 Schematic diagram of the distribution of non-zero elements of the resistance matrix and the capacitance matrix of the target circuit before order reduction.

[0039] Figure 8 It is a schematic diagram of the non-zero element distribution of the reduced-order resistance matrix and capacitance matrix of the target circuit.

[0040] Fig. 9 It is a schematic diagram of the non-zero element distribution of the reduced-order resistance matrix and capacitance matrix of the target circuit.

[0041] Fig.10 It is a schematic diagram of the distribution of non-zero elements of the reduced-order resistance matrix and capacitance matrix of the target circuit according to one embodiment of the present application.

[0042] Fig.11 It is a schematic structural block diagram of a computing device 1100 according to an embodiment of the present application.

[0043] Fig.12 It is a schematic structural block diagram of a computing device 1200 according to an embodiment of the present application. DETAILED DESCRIPTION

[0044] The technical solutions in the embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0045] The technical solution of the embodiment of the present application can be applied to various computing devices, such as servers, desktop computers, laptop computers, handheld computers, portable computing devices, etc., which are not limited in the embodiment of the present application. The server can be, for example, a cloud server or a local server, etc., which are not limited in the embodiment of the present application.

[0046] Figure 1 It is a schematic structural block diagram of a circuit model reduction system 100 provided in an embodiment of the present application. Figure 1 The system 100 in FIG. 1 includes a determining component 110 and a processing component 120 .

[0047] The determination component 110 can determine at least one initial coupling capacitance matrix, an initial resistance matrix and an initial capacitance matrix of each interconnection line according to multiple interconnections in the target circuit. Among them, the target circuit includes multiple interconnections, each interconnection line includes multiple nodes, and each node is used to separate electronic components in the circuit, such as resistors or capacitors. The target circuit can be an RC circuit. Every two nodes in the interconnection line are connected by a resistor, that is, the interconnection line is a full resistance path. The two interconnections of the target circuit can be connected by a capacitor, and the capacitor is a coupling capacitor. The initial coupling capacitance matrix is ​​used to indicate the coupling capacitance value between the nodes of the two interconnections. That is, according to the coupling capacitance value between the nodes of the two interconnections, an initial coupling capacitance matrix can be determined. The initial resistance matrix is ​​used to indicate the resistance value between the nodes in the interconnection line corresponding to the initial resistance matrix, and the initial capacitance matrix is ​​used to indicate the capacitance value between the nodes in the interconnection line corresponding to the initial capacitance matrix.

[0048] In some embodiments, the determination component 110 may also determine an initial input port matrix and an initial output port matrix for each interconnection line according to multiple interconnections in the target circuit. The initial input port matrix is ​​used to indicate the input ports in the interconnection lines corresponding to the initial input port matrix, and the initial output port matrix is ​​used to indicate the output ports in the interconnection lines corresponding to the initial output port matrix.

[0049] In some embodiments, the circuit model of the target circuit before order reduction includes a resistance matrix of the target circuit before order reduction, and the resistance matrix before order reduction is a block diagonal matrix. The resistance matrix before order reduction includes an initial resistance matrix of each interconnect line, and the initial resistance matrix of each interconnect line is a diagonal block element of the resistance matrix before order reduction.

[0050] In some embodiments, the circuit model of the target circuit before reduction also includes a capacitance matrix of the target circuit before reduction, and the capacitance matrix before reduction is a block diagonal plus edge BBD matrix. The capacitance matrix before reduction includes the initial capacitance matrix of each interconnection line and part or all of the initial coupling capacitance matrices in at least one initial coupling capacitance matrix. The initial capacitance matrix of each interconnection line is a diagonal block element of the capacitance matrix before reduction. Each initial coupling capacitance matrix in the part or all of the initial coupling capacitance matrices is a block element in the first region of the capacitance matrix before reduction, and the first region is a row region or a column region where the first diagonal block in the capacitance matrix before reduction is located. The capacitance matrix before reduction is a symmetric matrix.

[0051] In some embodiments, the circuit model of the target circuit before order reduction further includes an input port matrix of the target circuit before order reduction, and the input port matrix before order reduction includes an initial input port matrix of each interconnection line.

[0052] In some embodiments, the circuit model of the target circuit before order reduction further includes an output port matrix of the target circuit before order reduction, and the output port matrix before order reduction includes an initial output port matrix of each interconnection line.

[0053] The processing component 120 may determine the projection matrix of each interconnect line according to the initial resistance matrix of each interconnect line. The projection matrix of each interconnect line is used to reduce the order of one or more initial coupling capacitance matrices, the initial resistance matrix of the interconnect line, and the initial capacitance matrix. The processing component 120 may also reduce the order of at least one initial coupling capacitance matrix, the initial resistance matrix of each interconnect line, and the initial capacitance matrix according to the projection matrix of each interconnect line to determine the reduced-order circuit model of the target circuit.

[0054] In some embodiments, the processing component 120 can reduce the order of the initial resistance matrix of each interconnect line according to the projection matrix of each interconnect line to determine the reduced order resistance matrix of each interconnect line. The processing component 120 can also determine the reduced order resistance matrix according to the reduced order resistance matrix of each interconnect line. The reduced order resistance matrix of each interconnect line is the diagonal block elements in the reduced order resistance matrix.

[0055] In some embodiments, the reduced-order circuit model of the target circuit includes a reduced-order resistance matrix of the target circuit, and the reduced-order resistance matrix is ​​a block diagonal matrix. The reduced-order resistance matrix includes a reduced-order resistance matrix of each interconnect line, and the reduced-order resistance matrix of each interconnect line is a diagonal block element of the reduced-order resistance matrix.

[0056] In some embodiments, the processing component 120 can reduce the order of the initial capacitance matrix of each interconnect line according to the projection matrix of each interconnect line to determine the reduced order capacitance matrix of each interconnect line. The processing component 120 can also reduce the order of at least one initial coupling capacitance matrix according to the projection matrices of multiple interconnect lines to determine at least one reduced order coupling capacitance matrix, and each reduced order coupling capacitance matrix corresponds to an initial coupling capacitance matrix. The processing component 120 can also determine the reduced order capacitance matrix according to the at least one reduced order coupling capacitance matrix and the reduced order capacitance matrix of each interconnect line.

[0057] In some embodiments, the processing component 120 can reduce the order of the first initial coupling capacitance matrix according to the projection matrix of the first interconnection line and the projection matrix of the second interconnection line to determine the first reduced-order coupling capacitance matrix. The first interconnection line and the second interconnection line are different interconnections among the plurality of interconnections of the target circuit. The first initial coupling capacitance matrix is ​​used to indicate the coupling capacitance values ​​between the nodes in the first interconnection line and the second interconnection line.

[0058] In some embodiments, the processing component 120 can use the reduced-order capacitance matrix of each interconnect as a diagonal block element in the reduced-order capacitance matrix, and can also use one or more second reduced-order coupling capacitance matrices as block elements in the first region of the reduced-order capacitance matrix. The second reduced-order coupling capacitance matrix corresponds to the third interconnect and the fourth interconnect, the third interconnect is the interconnect corresponding to the first diagonal block in the reduced-order capacitance matrix, and the fourth interconnect is an interconnect different from the third interconnect among the plurality of interconnects. The first region is a row region or a column region where the first diagonal block is located. The reduced-order capacitance matrix is ​​a symmetric matrix.

[0059] In some embodiments, the reduced-order circuit model of the target circuit also includes a reduced-order capacitance matrix of the target circuit, and the reduced-order capacitance matrix is ​​a BBD matrix. The reduced-order capacitance matrix includes a reduced-order capacitance matrix of each interconnect line and part or all of the reduced-order coupling capacitance matrices in at least one reduced-order coupling capacitance matrix. The reduced-order capacitance matrix of each interconnect line is a diagonal block element of the reduced-order capacitance matrix. Each reduced-order coupling capacitance matrix in the part or all of the reduced-order coupling capacitance matrices is a block element in a first region in the reduced-order capacitance matrix, and the first region is a row region or a column region where a first diagonal block in the reduced-order capacitance matrix is ​​located. The reduced-order capacitance matrix is ​​a symmetric matrix.

[0060] In some embodiments, the processing component 120 may determine the projection matrix of each interconnect line according to the initial resistance matrix of each interconnect line. Alternatively, the processing component 120 may determine the projection matrix of each interconnect line according to the initial resistance matrix and the initial capacitance matrix of each interconnect line. Alternatively, the processing component 120 may determine the projection matrix of each interconnect line according to the initial input port matrix, the initial resistance matrix, and the initial capacitance matrix of each interconnect line.

[0061] In some embodiments, the processing component 120 can also reduce the initial input port matrix and the initial output port matrix of each interconnect line according to the projection matrix of each interconnect line, and determine the reduced-order input port matrix and the reduced-order output port matrix of each interconnect line. The processing component 120 can also determine the reduced-order input port matrix according to the reduced-order input port matrix of each interconnect line. The processing component 120 can also determine the reduced-order output port matrix according to the reduced-order output port matrix of each interconnect line.

[0062] In some embodiments, the reduced-order circuit model of the target circuit includes a reduced-order input port matrix of the target circuit, the reduced-order input port matrix including a reduced-order input port matrix for each interconnect line.

[0063] In some embodiments, the reduced-order circuit model of the target circuit includes a reduced-order output port matrix of the target circuit, the reduced-order output port matrix including a reduced-order output port matrix of each interconnect line.

[0064] In some embodiments, the processing component 120 may also determine the output signal of each output port of the target circuit according to the input signal of each input port in the target circuit and the reduced-order circuit model. The target circuit includes at least one input port and at least one output port.

[0065] The system 100 can determine the projection matrix of each interconnection line respectively in units of interconnections, thereby reducing the order of the initial resistance matrix, initial capacitance matrix and initial coupling capacitance matrix between interconnections corresponding to each interconnection line, and then obtaining the reduced-order circuit model of the target circuit. In other words, the system 100 can reduce the order of the block matrix corresponding to each interconnection line in the circuit model of the target circuit before reduction, without reducing the order of the entire matrix in the circuit model before reduction, so that the structure of the matrix in the circuit model before and after reduction can remain unchanged. That is, the system 100 can keep the sparsity of the matrix in the circuit model before and after reduction unchanged, so as to reduce the amount of calculation when calculating the output signal and improve the efficiency of calculating the output signal. In addition, by reducing the order in units of interconnections, the reduced-order complexity caused by multiple input ports of the target circuit can be apportioned to the matrix corresponding to each interconnection line, and each interconnection line includes at least one input port, so that the efficiency of reducing the order of the circuit model is higher.

[0066] Figure 2 is a schematic flow chart of a circuit model reduction method provided in an embodiment of the present application, Figure 2 The following steps are included.

[0067] S210 , determining at least one initial coupling capacitance matrix, an initial resistance matrix and an initial capacitance matrix of each interconnection line according to a plurality of interconnections in a target circuit.

[0068] The target circuit may include multiple nodes, and the multiple nodes are interconnected. The nodes in the target circuit are used to separate electronic components in the target circuit, such as resistors or capacitors. The target circuit may be an RC circuit. If there is a path in the target circuit in which each node is connected to each other through a resistor, then the path is an interconnection line. In other words, every two nodes in the interconnection line are connected through a resistor, that is, the interconnection line is a full resistance path.

[0069] For example, suppose the target circuit includes Figure 311 nodes are shown. Among them, the 11 nodes are nodes 1 to 11. Specifically, node 1 is connected to node 2 through resistor 1. Node 2 is connected to node 3 through resistor 2. Node 3 is connected to node 4 in parallel through resistor 3 and capacitor 1. Node 4 is connected to grounded capacitor 7. Node 4 is connected to node 5 through resistor 4. Node 5 and node 6 are connected in parallel through resistor 5 and capacitor 2. Node 6 and node 7 are not connected through any electronic components. Node 7 and node 8 are connected through resistor 6. Node 8 and node 9 are connected in parallel through resistor 7 and capacitor 3. Node 9 is connected to grounded capacitor 8. Node 9 and node 10 are connected in parallel through resistor 8 and capacitor 4. Node 10 and node 11 are connected through resistor 9. Node 2 and node 8 are connected through capacitor 5, and node 5 and node 10 are connected through capacitor 6.

[0070] Depend on Figure 3 It can be seen that since nodes 1, 2, 3, 4, 5 and 6 can be connected through a full resistance path, nodes 1, 2, 3, 4, 5 and 6 belong to interconnection line 1. Similarly, since nodes 7, 8, 9, 10 and 11 can be connected through a full resistance path, nodes 7, 8, 9, 10 and 11 belong to interconnection line 2.

[0071] The computing device may determine at least one initial coupling capacitance matrix, an initial resistance matrix of each interconnection line, and an initial capacitance matrix according to each interconnection line in the target circuit. The initial coupling capacitance matrix is ​​used to indicate the coupling capacitance value between the nodes of two interconnections. The initial resistance matrix is ​​used to indicate the resistance value between the nodes in the interconnection lines corresponding to the initial resistance matrix. The initial capacitance matrix is ​​used to indicate the capacitance value between the nodes in the interconnection lines corresponding to the initial capacitance matrix.

[0072] Optionally, the computing device may determine the size of the initial resistance matrix and the initial capacitance matrix of each interconnect line according to the number of nodes in each interconnect line. The number of rows and columns of the initial resistance matrix of each interconnect line is the same as the number of nodes included in each interconnect line.

[0073] For example, Figure 3 For example, since interconnection line 1 includes 6 nodes, the initial resistance matrix of interconnection line 1 is a 6×6 matrix, and the initial capacitance matrix is ​​also a 6×6 matrix. Since interconnection line 2 includes 5 nodes, the initial resistance matrix of interconnection line 2 is a 5×5 matrix, and the initial capacitance matrix is ​​also a 5×5 matrix.

[0074] Optionally, the computing device may determine the values ​​of the elements in the initial resistance matrix of the interconnect according to the resistance value between each node in the interconnect. For example, the computing device may determine the values ​​of the elements in the xth row and yth column and the values ​​of the elements in the yth row and xth column in the initial resistance matrix of the interconnect according to the resistance value between the xth node and the yth node in the interconnect. The xth node and the yth node are different nodes in the interconnect. The value of the element in the xth row and yth column is the same as the value of the element in the yth row and xth column. The computing device may also determine the values ​​of the elements in the xth row and xth column in the initial resistance matrix of the interconnect according to the resistance value between the xth node in the interconnect and other nodes in the interconnect except the xth node.

[0075] In some embodiments, the values ​​of the elements in the initial resistance matrix of the interconnect are related to the connection mode between each node. For example, if the xth node and the yth node in the interconnect are indirectly connected through multiple resistors, the values ​​of the xth row and yth column elements and the yth row and xth column elements in the initial resistance matrix of the interconnect are 0. If the xth node and the yth node in the interconnect are directly connected through a resistor, the values ​​of the xth row and yth column elements and the yth row and xth column elements in the initial resistance matrix of the interconnect are determined according to the resistance value of the resistor.

[0076] When the xth node and the yth node in the interconnection line are directly connected through a resistor, the value of the element in the xth row and yth column may be the inverse of the reciprocal of the resistance value of the resistor. The value of the element in the xth row and xth column may be the sum of the reciprocals of the resistance values ​​between the xth node and other nodes in the interconnection line except the xth node. When the xth node is connected to a grounding resistor, the value of the element in the xth row and xth column may be the sum of the reciprocals of the resistance values ​​between the xth node and other nodes in the interconnection line except the xth node, and the reciprocal of the resistance value of the grounding resistor.

[0077] It should be understood that the serial numbers of the nodes in the interconnection lines are only used to distinguish different nodes and have no practical significance. The embodiment of the present application does not limit the method of determining the serial numbers of the nodes.

[0078] For example, Figure 3 For example, assume that the resistance value of resistor 1 is R1, the resistance value of resistor 2 is R2, the resistance value of resistor 3 is R3, the resistance value of resistor 4 is R4, and the resistance value of resistor 5 is R5. Figure 3 For interconnect line 1 in the example, the initial resistance matrix of interconnect line 1 can be determined as shown in the following formula (1):

[0079]

[0080] Assume that the resistance value of resistor 6 is R6, the resistance value of resistor 7 is R7, the resistance value of resistor 8 is R8, and the resistance value of resistor 9 is R9. Figure 3 For interconnect line 2 in FIG, the initial resistance matrix of interconnect line 2 can be determined as shown in the following formula (2):

[0081]

[0082] Optionally, the computing device may determine the value of the element in the initial capacitance matrix of the interconnect according to the capacitance value between each node in the interconnect. For example, the computing device may determine the value of the element in the xth row and yth column and the value of the element in the yth row and xth column in the initial capacitance matrix of the interconnect according to the capacitance value between the xth node and the yth node in the interconnect. The xth node and the yth node are different nodes in the interconnect. The value of the element in the xth row and yth column is the same as the value of the element in the yth row and xth column. The computing device may also determine the value of the element in the xth row and xth column in the initial capacitance matrix of the interconnect according to the capacitance value between the xth node in the interconnect and other nodes in the interconnect except the xth node. Alternatively, when there is a coupling capacitance between the interconnect and other interconnects in the target circuit except the interconnect, the computing device may determine the value of the element in the xth row and xth column in the initial capacitance matrix of the interconnect according to the coupling capacitance value between the xth node in the interconnect and other nodes in the target circuit except the xth node.

[0083] In some embodiments, the values ​​of the elements in the initial capacitance matrix of the interconnect are related to the connection mode between each node. For example, if the xth node and the yth node in the interconnect are indirectly connected through multiple capacitors or not connected through capacitors, the values ​​of the xth row and yth column elements and the yth row and xth column elements in the initial capacitance matrix of the interconnect are 0. If the xth node and the yth node in the interconnect are directly connected through a capacitor, the values ​​of the xth row and yth column elements and the yth row and xth column elements in the initial capacitance matrix of the interconnect are determined according to the capacitance value of the capacitor.

[0084] When the xth node and the yth node in the interconnect are directly connected by a capacitor, the value of the xth row and yth column element can be the opposite number of the capacitance value of the capacitor. The value of the xth row and xth column element can be the sum of the capacitance values ​​between the xth node and other nodes in the interconnect except the xth node. When the xth node is connected to a grounding capacitor, the value of the xth row and xth column element can be the sum of the capacitance values ​​between the xth node and other nodes in the interconnect except the xth node, and the capacitance value of the grounding capacitor. When there is a coupling capacitance between the xth node of the interconnect and other nodes in the target circuit except the xth node, the value of the xth row and xth column element can be the sum of the capacitance values ​​between the xth node and other nodes in the interconnect except the xth node, and the coupling capacitance values ​​between the xth node and nodes in the target circuit except the xth node. When there is a coupling capacitance between the xth node of the interconnect line and other nodes in the target circuit except the xth node, and the xth node is connected to a grounding capacitor, the value of the element in the xth row and xth column may be the sum of the capacitance value between the xth node and other nodes in the interconnect line except the xth node, the coupling capacitance value between the xth node and nodes in the target circuit except the xth node, and the capacitance value of the grounding capacitor.

[0085] For example, Figure 3 For example, assume that the capacitance value of capacitor 1 is c1, the capacitance value of capacitor 2 is c2, the capacitance value of capacitor 5 is c5, the capacitance value of capacitor 6 is c6, and the capacitance value of capacitor 7 is c7. Figure 3 The initial capacitance matrix of interconnection line 1 can be determined as shown in equation (3):

[0086]

[0087] Assume that the capacitance value of capacitor 3 is c3, the capacitance value of capacitor 4 is c4, and the capacitance value of capacitor 8 is c8. Figure 3 The interconnection line 2 in the figure can be determined as follows:

[0088]

[0089] Optionally, the computing device may determine the size of the initial coupling capacitance matrix corresponding to the two interconnects according to the number of nodes in each of the two interconnects. For example, the computing device may determine the size of the first initial coupling capacitance matrix according to the number of nodes in the first interconnect and the second interconnect. The first interconnect and the second interconnect are different interconnects among the plurality of interconnects of the target circuit. The number of rows of the first initial coupling capacitance matrix is ​​the same as the number of nodes in the first interconnect, and the number of columns is the same as the number of nodes in the second interconnect. Alternatively, the number of rows of the first initial coupling capacitance matrix is ​​the same as the number of nodes in the second interconnect, and the number of columns is the same as the number of nodes in the first interconnect.

[0090] For example, Figure 3 For example, since interconnection line 1 includes 6 nodes and interconnection line 2 includes 5 nodes, the initial coupling capacitance matrix corresponding to interconnection line 1 and interconnection line 2 is a 6×5 matrix or a 5×6 matrix.

[0091] Optionally, the computing device may determine the values ​​of the elements in the initial coupling capacitance matrix corresponding to the two interconnects according to the capacitance value between the nodes of the two interconnects. For example, the computing device may determine the values ​​of the elements in the x1th row and x2th column and the values ​​of the elements in the x2th row and x1th column in the first initial coupling capacitance matrix according to the coupling capacitance value between the x1th node in the first interconnect and the x2th node in the second interconnect. The x1th node is any node in the first interconnect, and the x2th node is any node in the second interconnect. The value of the element in the x1th row and x2th column is the same as the value of the element in the x2th row and x1th column.

[0092] In some embodiments, the values ​​of the elements in the initial coupling capacitor matrix are related to the connection mode between the nodes in the two interconnects. For example, if there is no coupling capacitor between the x1th node in the first interconnect and the x2th node in the second interconnect, or they are indirectly connected through multiple coupling capacitors, the values ​​of the x1th row and x2th column elements and the x2th row and x1th column elements in the first initial coupling capacitor matrix are 0. If the x1th node in the first interconnect and the x2th node in the second interconnect are directly connected through a coupling capacitor, the values ​​of the x1th row and x2th column elements and the x2th row and x1th column elements in the first initial coupling capacitor matrix are determined according to the capacitance value of the coupling capacitor.

[0093] When the x1th node in the first interconnection line and the x2th node in the second interconnection line are directly connected via a coupling capacitor, the value of the element in the x1th row and x2th column may be the opposite of the capacitance value of the coupling capacitor.

[0094] For example, Figure 3For example, assume that the capacitance value of capacitor 5 is c5 and the capacitance value of capacitor 6 is c6. Figure 3 The interconnection line 1 and the interconnection line 2 in FIG. 1 can determine the initial coupling capacitance matrix corresponding to the interconnection line 1 and the interconnection line 2 as shown in the following formula (5) or the following formula (6):

[0095]

[0096]

[0097] Optionally, the computing device may also determine an initial input port matrix of each interconnection line according to multiple interconnections in the target circuit. The initial input port matrix is ​​used to indicate input ports in the interconnection lines corresponding to the initial input port matrix. The input port is a node in the target circuit to which a signal can be input.

[0098] Optionally, the computing device may determine the size of the initial input port matrix of each interconnect line according to the number of input ports in the target circuit and the number of nodes in each interconnect line. For example, the number of rows of the initial input port matrix of each interconnect line is the same as the number of nodes in the interconnect line, and the number of columns of the initial input port matrix of each interconnect line is the same as the number of input ports in the target circuit.

[0099] For example, Figure 3 For example, assuming that the target circuit includes 2 input port nodes, since interconnection line 1 includes 6 nodes, the size of the initial input port matrix of interconnection line 1 is 6×2. Since interconnection line 2 includes 5 nodes, the size of the initial input port matrix of interconnection line 2 is 5×2.

[0100] Optionally, the computing device can determine the values ​​of the elements in the initial input port matrix of the interconnect line according to the input port nodes in the interconnect line. For example, if the xth node in the interconnect line is an input port node, and the xth node in the interconnect line is the pth input port node in the target circuit, then the value of the element in the xth row and pth column in the initial input port matrix of the interconnect line is 1. Wherein, p=1, ..., P, P is the number of input ports in the target circuit. If the yth node in the interconnect line is not an input port node, the value of the element in the yth row in the initial input port matrix of the interconnect line is 0. The xth node and the yth node are any points in the interconnect line.

[0101] For example, Figure 3 As an example, assume that the target circuit includes two input ports, namely node 1 and node 7. Figure 3 The interconnection line 1 and the interconnection line 2 in the example can be determined as follows: the initial input port matrix of the interconnection line 1 is as shown in the following equation (7), and the initial input port matrix of the interconnection line 2 is as shown in the following equation (8).

[0102]

[0103]

[0104] Optionally, the computing device may also determine an initial output port matrix for each interconnection line according to multiple interconnections in the target circuit. The initial output port matrix is ​​used to indicate output ports in the interconnection lines corresponding to the initial output port matrix. The output port is a node in the target circuit that can detect an output signal.

[0105] Optionally, the computing device may determine the size of the initial output port matrix of each interconnect line according to the number of output ports in the target circuit and the number of nodes in each interconnect line. For example, the number of rows of the initial output port matrix of each interconnect line is the same as the number of nodes in the interconnect line, and the number of columns of the initial output port matrix of each interconnect line is the same as the number of output ports in the target circuit.

[0106] For example, Figure 3 For example, assuming that the target circuit includes 2 output port nodes, since interconnection line 1 includes 6 nodes, the size of the initial output port matrix of interconnection line 1 is 6×2. Since interconnection line 2 includes 5 nodes, the size of the initial output port matrix of interconnection line 2 is 5×2.

[0107] Optionally, the computing device can determine the values ​​of the elements in the initial output port matrix of the interconnect line according to the output port nodes in the interconnect line. For example, if the xth node in the interconnect line is an output port node, and the xth node in the interconnect line is the qth output port node in the target circuit, then the value of the element in the xth row and qth column in the initial output port matrix of the interconnect line is 1. Wherein, q=1, ..., Q, Q is the number of output ports in the target circuit. If the yth node in the interconnect line is not an output port node, the value of the element in the yth row in the initial output port matrix of the interconnect line is 0. The xth node and the yth node are any points in the interconnect line.

[0108] For example, Figure 3 As an example, assume that the target circuit includes two output ports, namely node 6 and node 11. Figure 3 The interconnection line 1 and the interconnection line 2 in the example can be determined as follows: the initial output port matrix of the interconnection line 1 is as shown in the following equation (9), and the initial output port matrix of the interconnection line 2 is as shown in the following equation (10).

[0109]

[0110]

[0111] In some embodiments, the computing device may determine at least one initial coupling capacitance matrix, an initial resistance matrix and an initial capacitance matrix of each interconnect in the target circuit according to a node analysis method, a modified node analysis method (MNA), etc. The embodiment of the present application does not limit the specific method for determining the initial coupling capacitance matrix, the initial resistance matrix and the initial capacitance matrix.

[0112] S220 , determining a projection matrix of each interconnection line according to the initial resistance matrix of each interconnection line.

[0113] The computing device may determine a projection matrix of each interconnect line according to an initial resistance matrix of each interconnect line. The projection matrix of each interconnect line is used to reduce the order of one or more initial coupling capacitance matrices, the initial resistance matrix and the initial capacitance matrix of the interconnect line.

[0114] Optionally, the computing device may directly determine the projection matrix of each interconnection line according to the initial resistance matrix of each interconnection line.

[0115] Exemplarily, when determining the initial resistance matrix of the nth interconnection line, the computing device may sort the nodes in the nth interconnection line so that the initial resistance matrix G of the nth interconnection line is n It is expressed as shown in formula (11). Where n=1, ..., N, and N is the number of interconnection lines in the target circuit.

[0116]

[0117] Among them, G a1 It is used to indicate the resistance value between the first type of nodes in the nth interconnection line, and the first type of nodes are input port nodes or output port nodes in the nth interconnection line. a2 , G a3 It is used to indicate the resistance value between the first type node and the second type node in the nth interconnection line, and the second type node is the node other than the first type node in the nth interconnection line. a2 G a3 The transposed matrix of G a4 Used to indicate the resistance value between the second-type nodes in the nth interconnection line.

[0118] In some embodiments, after determining the initial resistance matrix of each interconnect line, the computing device may perform elementary transformation on the initial resistance matrix of each interconnect line, so that the initial resistance matrix of each interconnect line after elementary transformation is expressed in the form of the above formula (11).

[0119] For example, Figure 3Assume that node 1 is the input port node of interconnection line 1 and node 6 is the output port node of interconnection line 1. After elementary transformation of the above formula (1), the initial resistance matrix of interconnection line 1 can be shown as the following formula (12).

[0120]

[0121] Exemplarily, the computing device determines G in the form shown in formula (11): n After that, we can determine G n G a1 Is it a singular matrix? a1 non-singular, the computing device can determine the projection matrix V of the interconnect n As shown in the following formula (13).

[0122]

[0123] in, G a1 I1 represents the unit matrix, and the number of rows and columns of I1 is the same as the number of rows of the initial resistance matrix.

[0124] Optionally, the computing device may determine the projection matrix of each interconnect line according to an initial resistance matrix and an initial capacitance matrix of each interconnect line.

[0125] For example, if G a1 is a singular matrix, then the computing device can n Updated to G n +s0C n , the G n +s0C n is a non-singular matrix. Where s0 is a preset value. C n is the initial capacitance matrix of the nth interconnect, C n It can be expressed as shown in the following formula (14).

[0126]

[0127] Among them, C n With G n The sorting order of the nodes in C is consistent. n The element in the xth row and yth column in the nth interconnection line represents the capacitance between the xth node and the yth node. n The element in the xth row and yth column in represents the resistance value between the xth node and the yth node in the nth interconnection line. The xth node and the yth node are different nodes in the nth interconnection line. In other words, C a1 It is used to indicate the capacitance value between the first type of nodes in the nth interconnection line.a2 , C a3 It is used to indicate the capacitance value between the first type node and the second type node in the nth interconnection line. a2 C a3 The transposed matrix of C a4 Used to indicate the capacitance value between the second type of nodes in the nth interconnection line.

[0128] In some embodiments, after determining the initial capacitance matrix of each interconnect line, the computing device may perform an elementary transformation on the initial capacitance matrix of each interconnect line, so that the initial capacitance matrix of each interconnect line after the elementary transformation is expressed in the form of the above formula (14).

[0129] For example, Figure 3 Assume that node 1 is the input port node of interconnection line 1 and node 6 is the output port node of interconnection line 1. After performing elementary transformation on the above equation (3), the initial capacitance matrix of interconnection line 1 can be shown as the following equation (15).

[0130]

[0131] For example, the computing device converts G n Updated to G n +s0C n After that, the projection matrix of the interconnect line can be determined as shown in the following formula (16).

[0132]

[0133] in, G a5 The inverse matrix, G a5 =G a1 +s0C a1 , G a6 =G a2 +s0C a2 .

[0134] Optionally, the computing device may determine the projection matrix of each interconnection line according to an initial input port matrix, an initial resistance matrix, and an initial capacitance matrix of each interconnection line.

[0135] For example, in the initial resistance matrix G of the nth interconnection line n When non-singular, the computing device can calculate the initial resistance matrix G of the nth interconnection line according to the initial resistance matrix G of the nth interconnection line. n and the initial input port matrix B n , determine the first intermediate matrix M of the nth interconnection line n Among them, G n M n =B nThe computing device can also calculate the first intermediate matrix M n Perform orthogonal triangular decomposition to obtain matrix X0 and matrix K0, that is, M n =X0K0. The number of rows of matrix X0 is the same as the number of rows of the first intermediate matrix M n The number of rows of matrix X0 is the same as that of the first intermediate matrix M. n The matrix K0 is a square matrix, and the number of rows of the matrix K0 is the same as the number of rows of the matrix X0. The computing device can also calculate the initial capacitance matrix C of the nth interconnection line according to the initial capacitance matrix C of the nth interconnection line. n and matrix X0, determine the projection matrix V of the nth interconnection line n .

[0136] For example, the computing device may be configured to: i =C n X i-1 , determine the matrix W i . Where i = 1, ..., r, represents rounding down. P is the number of input ports in the target circuit, and m is the number of rows of the resistance matrix after the target circuit is reduced in order. m is a preset threshold value, and the specific value of m is not limited in the embodiment of the present application. The computing device can also be based on Determine the matrix in, Represents the matrix X in the 0th loop i The computing device can also be based on Determine the matrix H j . Wherein, j = 1, ..., i. For X i-j The transposed matrix of Represents the matrix X in the (j-1)th loop i The computing device can also be based on Determine the matrix in, Represents the matrix X in the jth loop i , Represents the matrix X in the i-th loop i The computing device can also perform matrix Perform orthogonal triangular decomposition to obtain the matrix X i ,Right now The computing device can also calculate the matrix X i , determine the projection matrix V of the nth interconnect line n Among them, V n =[X0,X1,......,X r-1 ].

[0137] For example, in the initial resistance matrix G of the nth interconnection linen When the singularity occurs, the computing device can n Updated to G n +s0C n , the G n +s0C n is a non-singular matrix. Where s0 is a preset value. The computing device can convert G n Updated to G n +s0C n After that, according to C n , B n and the updated G n , determine the projection matrix V of the nth interconnect line n The specific implementation method is the same as that of G n The method for non-singular cases is similar and will not be described here.

[0138] S230, reducing the order of at least one initial coupling capacitance matrix, the initial resistance matrix and the initial capacitance matrix of each interconnection line according to the projection matrix of each interconnection line, and determining a reduced-order circuit model of the target circuit.

[0139] Optionally, the computing device may reduce the order of the initial resistance matrix of each interconnect line according to the projection matrix of each interconnect line to determine the reduced-order resistance matrix of each interconnect line. The computing device may also determine the reduced-order resistance matrix according to the reduced-order resistance matrix of each interconnect line. The reduced-order resistance matrix of each interconnect line is the diagonal block elements in the reduced-order resistance matrix.

[0140] In some embodiments, the resistance matrix of the target circuit before order reduction includes an initial resistance matrix of each interconnection line, and the initial resistance matrix of each interconnection line is a diagonal block element of the resistance matrix before order reduction. That is, the initial resistance matrix G of the nth interconnection line n is the nth diagonal block element in the resistance matrix of the target circuit before order reduction, and the values ​​of the remaining elements in the resistance matrix before order reduction except the diagonal block elements are 0. The resistance matrix G before order reduction is shown in the following formula (17).

[0141]

[0142] In some embodiments, the computing device may calculate the projection matrix V of the nth interconnect line according to n , the initial resistance matrix G of the nth interconnection line n Perform order reduction to determine the reduced-order resistance matrix of the nth interconnect line For example, the computing device may Determine the reduced-order resistor matrix in, Indicates V n The transposed matrix of .

[0143] In some embodiments, the computing device may determine the reduced-order resistance matrix of the target circuit based on the reduced-order resistance matrix of each interconnect line, as shown in the following equation (18).

[0144]

[0145] Among them, the reduced-order resistance matrix of the nth interconnect line is is the reduced-order resistance matrix of the target circuit The nth diagonal block element in . The reduced-order resistance matrix of the target circuit The values ​​of all elements except the diagonal elements are 0.

[0146] It can be seen from the above formula (18) that the reduced-order resistance matrix obtained by the method in the embodiment of the present application is a block diagonal matrix. Compared with the resistance matrix before the reduced-order resistance matrix, the reduced-order resistance matrix can also maintain good sparsity.

[0147] Optionally, the computing device may reduce the order of the initial capacitance matrix of each interconnect line according to the projection matrix of each interconnect line to determine the reduced order capacitance matrix of each interconnect line. The computing device may also reduce the order of at least one initial coupling capacitance matrix according to the projection matrices of multiple interconnect lines to determine at least one reduced order coupling capacitance matrix, wherein each reduced order coupling capacitance matrix corresponds to an initial coupling capacitance matrix. The computing device may also determine a reduced order capacitance matrix according to at least one reduced order coupling capacitance matrix and the reduced order capacitance matrix of each interconnect line.

[0148] In some embodiments, the capacitance matrix of the target circuit before order reduction includes the initial capacitance matrix of each interconnect line and part or all of the initial coupling capacitance matrices in at least one initial coupling capacitance matrix. The initial capacitance matrix of each interconnect line is a diagonal block element of the capacitance matrix before order reduction. Each initial coupling capacitance matrix in part or all of the initial coupling capacitance matrices is a block element in the first region of the capacitance matrix before order reduction, and the first region is the row region or column region where the first diagonal block in the capacitance matrix before order reduction is located. The capacitance matrix of the target circuit before order reduction is a symmetric matrix. That is, the initial capacitance matrix C of the nth interconnect line n is the nth diagonal block element in the capacitance matrix C of the target circuit before order reduction. The initial coupling capacitance matrix corresponding to the n1th interconnection line and each interconnection line in the target circuit except the n1th interconnection line is a block element in the first region of the capacitance matrix C before order reduction. The n1th interconnection line is the interconnection line corresponding to the first diagonal block element in the capacitance matrix C before order reduction. The values ​​of the remaining elements in the capacitance matrix C before order reduction except the diagonal block elements and the first region are 0. The capacitance matrix C before order reduction is shown in the following formula (19).

[0149]

[0150] The first region is the column region where the first diagonal block element in the capacitance matrix before order reduction is located. 21 represents the initial coupling capacitance matrix corresponding to the second interconnection line and the first interconnection line, that is, C 21 Used to indicate the coupling capacitance value between the node of the second interconnection line and the first interconnection line. C 21 The transposed matrix of C. N1 represents the initial coupling capacitance matrix corresponding to the Nth interconnection line and the first interconnection line, that is, C N1 Used to indicate the coupling capacitance value between the node of the Nth interconnection line and the 1st interconnection line. C N1 The transposed matrix of . In the above formula (19), except for C n , C n1 and The values ​​of the remaining elements outside are 0, n=1,...,N.

[0151] It should be understood that the serial number of the interconnection line is only used to distinguish different interconnections and has no practical significance. The embodiment of the present application does not limit the method of determining the serial number of the interconnection line.

[0152] Alternatively, the above formula (19) can be expressed in the form of formula (20).

[0153]

[0154] The first region is the row region where the first diagonal block element in the capacitance matrix before order reduction is located. It should be understood that C 21 and C 12 Both are used to indicate the initial coupling capacitance matrix corresponding to the second interconnection line and the first interconnection line.

[0155] In some embodiments, the computing device may calculate the projection matrix V of the nth interconnect line according to n , the initial capacitance matrix C of the nth interconnection line n Perform order reduction to determine the reduced-order capacitance matrix of the nth interconnect line For example, the computing device may Determine the reduced-order resistor matrix in, Indicates V n The transposed matrix of .

[0156] In some embodiments, the computing device may use the projection matrix V of the first interconnect line d and the projection matrix V of the second interconnect e , for the first initial coupling capacitance matrix C dePerform order reduction to determine the first reduced-order coupling capacitance matrix The first interconnection line and the second interconnection line are different interconnections among a plurality of interconnections of the target circuit. The first initial coupling capacitance matrix C de For indicating the coupling capacitance value between the nodes in the first interconnection line and the second interconnection line. Determine the first reduced-order coupling capacitance matrix in, Indicates V d The transposed matrix of .

[0157] In some embodiments, the computing device may determine a reduced-order capacitance matrix of the target circuit according to the reduced-order capacitance matrix of each interconnect line and one or more reduced-order coupling capacitance matrices. As shown in the following formula (21).

[0158]

[0159] Among them, the reduced-order capacitance matrix of the nth interconnect line is is the reduced-order capacitance matrix of the target circuit The second reduced-order coupling capacitance matrix corresponding to the third interconnection line and the fourth interconnection line is the reduced-order capacitance matrix The third interconnection line is a reduced-order capacitance matrix. The first region is a row region or a column region where the first diagonal block is located. The reduced-order capacitance matrix is a symmetric matrix. The reduced-order capacitance matrix of the target circuit is Except for the diagonal block elements and the first region, the values ​​of the remaining elements in are 0. That is, the reduced-order capacitance matrix The method includes a reduced-order capacitance matrix for each interconnect line and one or more second reduced-order coupling capacitance matrices.

[0160] It can be seen from the above formula (21) that the reduced-order capacitance matrix obtained by the method in the embodiment of the present application is a BBD matrix. Compared with the capacitance matrix before reduction, the reduced-order capacitance matrix can also maintain good sparsity.

[0161] When the target circuit includes two interconnects, the computing device may determine an initial capacitance matrix and an initial coupling capacitance matrix of each interconnect according to the two interconnects in the target circuit, thereby determining a capacitance matrix before order reduction of the target circuit. The capacitance matrix before order reduction includes the initial capacitance matrix of each interconnect and the initial coupling capacitance matrix.

[0162] When the target circuit includes more than two interconnects, the computing device may determine an initial capacitance matrix and multiple initial coupling capacitance matrices for each interconnect according to multiple interconnects in the target circuit, thereby determining multiple capacitance matrices before reduction of the target circuit. The multiple capacitance matrices before reduction include multiple capacitance matrices before reduction that use each interconnect as a first diagonal block. The initial coupling capacitance matrices included in each capacitance matrix before reduction in the multiple capacitance matrices before reduction are not completely the same.

[0163] For example, assuming that the target circuit includes three interconnects (i.e., interconnect 1, interconnect 2, and interconnect 3), the target circuit may have three capacitance matrices before order reduction, namely, capacitance matrix 1 before order reduction, capacitance matrix 2 before order reduction, and capacitance matrix 3 before order reduction. Among them, capacitance matrix 1 before order reduction includes C1, C2, C3, C 12 , C 13 , C 12 The transposed matrix, C 13 The transposed matrix of . Where C1 is the initial capacitance matrix of interconnection line 1, C2 is the initial capacitance matrix of interconnection line 2, and C3 is the initial capacitance matrix of interconnection line 3. 12 is the initial coupling capacitance matrix corresponding to interconnection line 1 and interconnection line 2, C 13 is the initial coupling capacitance matrix corresponding to interconnection line 1 and interconnection line 3. The capacitance matrix 2 before order reduction includes C1, C2, C3, C 12 , C 23 , C 12 The transposed matrix, C 23 The transposed matrix of . 23 is the initial coupling capacitance matrix corresponding to interconnection line 2 and interconnection line 3. The capacitance matrix 3 before reduction includes C1, C2, C3, C 13 , C 23 , C 13 The transposed matrix, C 23 That is, the initial coupling capacitance matrix contained in each capacitance matrix before reduction is not the same.

[0164] When the target circuit includes more than two interconnects, when each capacitance matrix of the target circuit before reduction is reduced in order according to the method in the embodiment of the present application, the projection matrix of each interconnect can be reused, and at least one initial coupling capacitance matrix included in each capacitance matrix before reduction and the initial capacitance matrix of each interconnect can be reduced in order, without calculating the projection matrix corresponding to each capacitance matrix before reduction as a whole. Therefore, the computational cost of reducing the order of the target circuit can be saved, thereby improving the efficiency of reducing the order of the target circuit.

[0165] Optionally, the computing device may determine a reduced-order circuit model of the target circuit according to the reduced-order resistance matrix and the reduced-order capacitance matrix. For example, the reduced-order circuit model includes the above equations (18) and (21).

[0166] Optionally, the computing device may reduce the order of the initial input port matrix of each interconnect line according to the projection matrix of each interconnect line to determine the reduced-order input port matrix of each interconnect line. The computing device may also determine the reduced-order input port matrix according to the reduced-order input port matrix of each interconnect line.

[0167] In some embodiments, the input port matrix of the target circuit before order reduction includes the initial input port matrix of each interconnection line. The input port matrix B before order reduction is shown in the following equation (22).

[0168]

[0169] Among them, B n is the initial input port matrix of the nth interconnect line.

[0170] In some embodiments, the computing device may calculate the projection matrix V of the nth interconnect line according to n , for the initial input port matrix B of the nth interconnection line n Perform order reduction to determine the reduced-order input port matrix of the nth interconnect line For example, the computing device may Determine the reduced-order input port matrix in, Indicates V n The transposed matrix of .

[0171] In some embodiments, the computing device may determine the reduced-order input port matrix of the target circuit according to the reduced-order input port matrix of each interconnect line. As shown in the following formula (23).

[0172]

[0173] Optionally, the computing device may reduce the order of the initial output port matrix of each interconnect line according to the projection matrix of each interconnect line to determine the reduced-order output port matrix of each interconnect line. The computing device may also determine the reduced-order output port matrix according to the reduced-order output port matrix of each interconnect line.

[0174] In some embodiments, the output port matrix of the target circuit before order reduction includes an initial output port matrix of each interconnection line. The output port matrix L before order reduction is shown in the following equation (24).

[0175]

[0176] Among them, L n is the initial output port matrix of the nth interconnect line.

[0177] In some embodiments, the computing device may calculate the projection matrix V of the nth interconnect line according to n , for the initial output port matrix L of the nth interconnection line n Perform order reduction to determine the reduced-order output port matrix of the nth interconnect line For example, the computing device may Determine the reduced-order output port matrix in, Indicates V n The transposed matrix of .

[0178] In some embodiments, the computing device may determine the reduced-order output port matrix of the target circuit according to the reduced-order output port matrix of each interconnect line. As shown in the following formula (25).

[0179]

[0180] Optionally, the computing device may calculate the reduced-order resistance matrix Reduced-order capacitance matrix Reduced input port matrix and the reduced output port matrix Determine a reduced-order circuit model of the target circuit. For example, the reduced-order circuit model includes the above equations (18), (21), (23) and (25).

[0181] Optionally, after determining the reduced-order circuit model of the target circuit, the computing device may further execute step S240 .

[0182] S240, determining an output signal of each output port of the target circuit according to an input signal of each input port in the target circuit and the reduced-order circuit model.

[0183] For example, the computing device can calculate the reduced-order resistance matrix Reduced-order capacitance matrix Reduced input port matrix And the following formula (26) to determine x(t).

[0184]

[0185] Wherein, h is a preset discrete step length, and the specific size of h is not limited in the embodiment of the present application, and h is a fixed value greater than 0. u(t) represents the input signal of each input port in the target circuit, and the input signal of each input port changes with time t.

[0186] Exemplarily, the computing device may also calculate the output port matrix after the reduction in order. The output signal y(t) of each output port is determined by the following equation (27). The output signal of each output port changes with time t.

[0187]

[0188] in, for The transposed matrix of .

[0189] In some embodiments, the computing device may Will Replace with in, K n When h is fixed, the reduced-order resistance matrix and reduced-order capacitance matrix of each interconnect only need to be calculated once. n and stored, so that when the output signal of each output port of the target circuit is subsequently calculated, the calculation overhead can be reduced, thereby improving the time for calculating the output signal.

[0190] The computing device can determine the projection matrix of each interconnection line in units of interconnections, thereby reducing the order of the initial resistance matrix, the initial capacitance matrix, and the initial coupling capacitance matrix between the interconnections corresponding to each interconnection line, and then obtaining the reduced-order circuit model of the target circuit. In other words, the computing device can reduce the order of the block matrix corresponding to each interconnection line in the circuit model of the target circuit before the reduction, without reducing the order of the entire matrix in the circuit model before the reduction, so that the structure of the matrix in the circuit model before and after the reduction can remain unchanged. That is, the method in the embodiment of the present application can make the sparsity of the matrix in the circuit model before and after the reduction better, so that the amount of calculation when calculating the output signal can be reduced, and the efficiency of calculating the output signal can be improved. In addition, the computing device reduces the order in units of interconnections, and the reduced-order complexity caused by multiple input ports of the target circuit can be apportioned to the matrix corresponding to each interconnection line, and each interconnection line includes at least one input port, so that the efficiency of reducing the order of the circuit model is higher.

[0191] At the same time, since the resistance matrix of the target circuit before and after order reduction, and the capacitance matrix before and after order reduction are both symmetric semi-positive definite matrices, the reduced-order circuit model obtained according to the method in the embodiment of the present application can maintain passivity, symmetry and stability.

[0192] Figure 4 is a schematic flow chart of a circuit model reduction method provided in an embodiment of the present application, Figure 4 The following steps are included.

[0193] S410, according to multiple interconnects in the target circuit, determine at least one initial coupling capacitance matrix and an initial resistance matrix, an initial capacitance matrix, an initial input port matrix and an initial output port matrix of each interconnect. For details, please refer to the description in step S210, which will not be repeated here.

[0194] S420, determining a projection matrix of the nth interconnection line according to the initial resistance matrix, initial capacitance matrix and initial input port matrix of the nth interconnection line. For details, please refer to the description in step S220, which will not be repeated here.

[0195] S430, according to the projection matrix of the nth interconnection line, respectively reduce the order of one or more initial coupling capacitance matrices, the initial resistance matrix of the nth interconnection line, the initial capacitance matrix, the initial input port matrix, and the initial output port matrix.

[0196] Exemplarily, the computing device may reduce the order of the initial coupling capacitance matrix corresponding to the nth interconnection and each interconnection except the nth interconnection in the target circuit according to the projection matrix of the nth interconnection, thereby determining the reduced-order coupling capacitance matrix corresponding to each initial coupling capacitance matrix. The computing device may also reduce the order of the initial resistance matrix of the nth interconnection according to the projection matrix of the nth interconnection, and determine the reduced-order resistance matrix of the nth interconnection. The computing device may also reduce the order of the initial capacitance matrix of the nth interconnection according to the projection matrix of the nth interconnection, and determine the reduced-order capacitance matrix of the nth interconnection. The computing device may also reduce the order of the initial input port matrix of the nth interconnection according to the projection matrix of the nth interconnection, and determine the reduced-order input port matrix of the nth interconnection. The computing device may also reduce the order of the initial output port matrix of the nth interconnection according to the projection matrix of the nth interconnection, and determine the reduced-order output port matrix of the nth interconnection. The specific implementation method can refer to the description in step S230, which will not be repeated here.

[0197] S440, determining a reduced-order resistance matrix according to the reduced-order resistance matrix of each interconnect line.

[0198] The computing device may use the reduced-order resistance matrix of each interconnect as a diagonal block element in the reduced-order resistance matrix, and set the elements in the reduced-order resistance matrix other than the diagonal block elements to 0, thereby determining the reduced-order resistance matrix, as shown in equation (18). The reduced-order resistance matrix is ​​a block diagonal matrix. For details, see the description in step S230.

[0199] S450 , determining a reduced-order capacitance matrix according to the reduced-order capacitance matrix of each interconnect line and one or more reduced-order coupling capacitance matrices.

[0200] The computing device may use the reduced-order capacitance matrix of each interconnect as a diagonal block element in the reduced-order capacitance matrix, and use one or more second reduced-order coupling capacitance matrices as block elements in the first region of the reduced-order capacitance matrix. The second reduced-order coupling capacitance matrix corresponds to the third interconnect and the fourth interconnect, the third interconnect is the interconnect corresponding to the first diagonal block in the reduced-order capacitance matrix, and the fourth interconnect is an interconnect different from the third interconnect among the plurality of interconnects. The first region is the row region or column region where the first diagonal block is located. The reduced-order capacitance matrix is ​​a symmetric matrix.

[0201] The computing device may also set the values ​​of elements other than the diagonal block elements and the block elements in the first region to 0, thereby determining a reduced-order capacitance matrix, as shown in equation (21). The reduced-order capacitance matrix is ​​a BBD matrix. For details, see the description in step S230.

[0202] S460, determining a reduced-order input port matrix according to the reduced-order input port matrix of each interconnection line.

[0203] The computing device may arrange the reduced-order input port matrix of each interconnection line in a column matrix form, thereby obtaining a reduced-order input port matrix, as shown in equation (23). For details, please refer to the description in step S230.

[0204] S470 , determining a reduced-order output port matrix according to the reduced-order output port matrix of each interconnect line.

[0205] The computing device may arrange the reduced-order output port matrix of each interconnection line in a column matrix form, thereby obtaining a reduced-order output port matrix, as shown in equation (25). For details, please refer to the description in step S230.

[0206] S480, determining the output signal of each output port according to the reduced-order resistance matrix, the reduced-order capacitance matrix, the reduced-order input port matrix, the reduced-order output port matrix and the input signal of each input port. For details, please refer to the description in step S240.

[0207] Assume that the specific information of the six target circuits is shown in Table 1.

[0208] Table 1 Target circuit information table

[0209]

[0210] As shown in Table 1, the target circuit 1 includes 2421 nodes, 113 interconnects, 113 input ports, and 2 output ports. The number of non-zero elements in the resistance matrix G of the target circuit 1 before order reduction is 7223, and the number of non-zero elements in the capacitance matrix C of the target circuit 1 before order reduction is 2667.

[0211] Input a square wave signal to each input port of the six target circuits. The square wave signal input to one input port of one target circuit is as follows: Figure 5 shown. Figure 5 It is a schematic diagram of the waveform of an input signal of an input port provided in an embodiment of the present application. Figure 5 The horizontal axis is the input time, in milliseconds (ms). Figure 5 The ordinate in is the current in amperes. The square wave signals input to each input port in each target circuit may be the same or different, and the present application embodiment does not limit this. The difference in square wave signals may be, for example, at least one of the peak value, rise time, fall time, and peak duration of the square wave signal.

[0212] According to the passive reduced-order interconnect macromodeling algorithm (PRIMA), sparse implicit projection (SIP) and the method in the embodiment of the present application, the 6 target circuits in Table 1 are reduced in order respectively, and the relative error of each output port of each target circuit is calculated. The error data that can be obtained is shown in Table 2. Among them, the relative error of the signal output by the p-th output port in a target circuit is er p It is determined by the following formula (28): p=1, ..., P, where P is the number of output ports in the target circuit.

[0213]

[0214] Wherein, y(p,b) represents the output signal value of the p-th output port at the b-th time determined according to the circuit model before order reduction. It represents the output signal value of the p-th output port at the b-th time determined according to the reduced-order circuit model.

[0215] Table 2 Error data of different order reduction methods

[0216]

[0217]

[0218] In Table 2, for the PRIMA method and the SIP method, if the resistance matrix G of the target circuit before order reduction is a singular matrix, G is updated to G+s1C. Where C is the capacitance matrix of the target circuit before order reduction, and the value of s1 is 1e5. For the method in the embodiment of the present application, the value of s0 is 1e5, and s0 refers to the description in step S220. For the PRIMA method and the method in the embodiment of the present application, the value of r is 2. Where, represents rounding down, and P is the number of input ports in the target circuit. For the PRIMA method, m is the number of rows of the reduced-order resistance matrix of the target circuit. For the method in the embodiment of the present application, m is the number of rows of the reduced-order resistance matrix of the nth interconnect. It should be understood that the maximum relative error, average relative error and the above 1e5 in Table 2 are data expressed in scientific notation.

[0219] As shown in Table 2, after the circuit model of the target circuit 1 is reduced in order according to the PRIMA method, the maximum relative error of the signal output by each output port of the target circuit 1 is 4.8541e-10, and the average relative error is 3.3357e-10. After the circuit model of the target circuit 1 is reduced in order according to the SIP method, the maximum relative error of the signal output by each output port of the target circuit 1 is 6.3135e-9, and the average relative error is 4.2566e-9. After the circuit model of the target circuit 1 is reduced in order according to the method in the embodiment of the present application, the maximum relative error of the signal output by each output port of the target circuit 1 is 1.5467e-9, and the average relative error is 1.0593e-9.

[0220] It can be seen from Table 2 that according to the method in the embodiment of the present application, the maximum relative error and the average relative error of the signal output by the output port of each target circuit are both small, so a higher order reduction accuracy can be achieved. That is, the circuit model after order reduction obtained according to the method in the embodiment of the present application has a higher similarity to the circuit model before order reduction.

[0221] Figure 6 It is a waveform diagram of an output signal of an output port provided in an embodiment of the present application. Figure 6 The horizontal axis is time, in ms. Figure 6 The vertical axis is voltage, in volts. Figure 6 The output signal 1 and the output signal 2 are included. The output signal 1 is the signal output by the output port determined according to the circuit model before order reduction. The output signal 2 is the signal output by the output port determined according to the circuit model after order reduction. Figure 6 It can be seen from the figure that the waveforms of output signal 1 and output signal 2 are almost completely overlapped. That is to say, the method in the embodiment of the present application can achieve a higher order reduction accuracy.

[0222] According to PRIMA, SIP and the method in the embodiments of the present application, the circuit models of the six target circuits in Table 1 are respectively reduced in order, and the time data that can be obtained is shown in Table 3.

[0223] Table 3 Reduction schedule of different reduction methods

[0224]

[0225]

[0226] As shown in Table 3, according to the PRIMA method, the circuit model of the target circuit 1 before reduction is reduced, and the number of rows and columns of the reduced resistance matrix and the reduced capacitance matrix are both 226. The time t0 required for simulating the circuit model of the target circuit 1 before reduction is 9.2 seconds. The time t0 required for simulating the reduced circuit model of the target circuit 1 obtained according to the PRIMA method is 9.2 seconds. r is 0.62 seconds. For the PRIMA method, the speedup ratio is 9.2 / 0.62≈15. After the circuit model of the target circuit 1 is reduced according to the PRIMA method, the maximum relative error of the signal output by each output port of the target circuit 1 is 4.8541e-10. The calculation method of the relative error can refer to the above formula (28).

[0227] It should be understood that the time required for simulating the circuit model before order reduction refers to the time required for inputting a signal to each input port in the target circuit and obtaining a signal output by each output port of the target circuit according to the circuit model before order reduction. The time required for simulating the circuit model after order reduction refers to the time required for inputting a signal to each input port in the target circuit and obtaining a signal output by each output port of the target circuit according to the circuit model after order reduction.

[0228] As shown in Table 3, according to the SIP method, the circuit model of the target circuit 1 before the order reduction is reduced, and the number of rows and columns of the reduced-order resistance matrix and the reduced-order capacitance matrix are both 113. The time t required to simulate the reduced-order circuit model of the target circuit 1 obtained according to the SIP method is ris 0.06 seconds. For the SIP method, the acceleration ratio is 9.2 / 0.06≈153. After the circuit model of the target circuit 1 is reduced in order according to the SIP method, the maximum relative error of the signal output by each output port of the target circuit 1 is 6.3135e-9. According to the method in the embodiment of the present application, the circuit model of the target circuit 1 before reduction is reduced in order, and the number of rows and columns of the reduced-order resistance matrix and the reduced-order capacitance matrix are both 115. The time t required to simulate the reduced-order circuit model of the target circuit 1 obtained according to the method in the embodiment of the present application is r The speedup ratio of the method in the embodiment of the present application is 9.2 / 0.06≈153. After the circuit model of the target circuit 1 is reduced in order according to the method in the embodiment of the present application, the maximum relative error of the signal output by each output port of the target circuit 1 is 1.5467e-9.

[0229] As can be seen from Table 3, compared with the PRIMA method, the speedup ratio of the method in the embodiment of the present application is greater. That is, the time required for calculating the output signal is shorter according to the reduced-order circuit model obtained by the method in the embodiment of the present application. It can also be seen from Tables 2 and 3 that, compared with the SIP method, although the speedup ratio advantage of the method in the embodiment of the present application is smaller, the reduction accuracy of the method in the embodiment of the present application is higher. In other words, the reduction efficiency of the method in the embodiment of the present application is higher.

[0230] Figure 7 Schematic diagram of the distribution of non-zero elements in the resistance matrix and capacitance matrix of the target circuit before order reduction. Figure 7 It can be seen from the figure that the resistance matrix of the target circuit before order reduction is a block diagonal matrix, and all elements except the diagonal block elements in the resistance matrix before order reduction are zero. Figure 7 It can also be seen that the capacitance matrix of the target circuit before order reduction is a BBD matrix, and the elements of the capacitance matrix before order reduction are all zero except the diagonal block elements, the row region where the first diagonal block is located, and the column region. That is, the sparsity of the resistance matrix before order reduction and the capacitance matrix before order reduction are both good.

[0231] Figure 8 Schematic diagram of the non-zero element distribution in the reduced-order resistance matrix and capacitance matrix obtained according to the PRIMA method. Figure 8 It can be seen that almost all the elements in the reduced-order resistance matrix obtained by the PRIMA method are non-zero elements, and almost all the elements in the reduced-order capacitance matrix obtained by the PRIMA method are non-zero elements. That is, the sparsity of the reduced-order resistance matrix and the reduced-order capacitance matrix obtained by the PRIMA method is poor.

[0232] Fig. 9Schematic diagram of the non-zero element distribution in the reduced-order resistance matrix and capacitance matrix obtained according to the SIP method. Fig. 9 It can be seen that almost all of the reduced-order resistance matrix obtained by the SIP method are non-zero elements, and almost all of the reduced-order capacitance matrix obtained by the SIP method are non-zero elements. That is, the reduced-order resistance matrix and the reduced-order capacitance matrix obtained by the SIP method have poor sparsity.

[0233] Fig.10 Schematic diagram of the distribution of non-zero elements in the reduced-order resistance matrix and capacitance matrix obtained according to the method in the embodiment of the present application. Fig.10 It can be seen that, in the reduced-order resistance matrix obtained according to the method in the embodiment of the present application, all elements except the diagonal block elements are zero. In the reduced-order capacitance matrix obtained according to the method in the embodiment of the present application, all elements except the diagonal block elements, the row region and the column region where the first diagonal block is located are zero. That is, the reduced-order resistance matrix and the reduced-order capacitance matrix obtained according to the method in the embodiment of the present application have good sparsity. Moreover, according to the method in the embodiment of the present application, the structure of the resistance matrix before and after the reduction can be kept unchanged, and the structure of the capacitance matrix before and after the reduction can also be kept unchanged.

[0234] Fig.11 It is a schematic structural block diagram of the computing device 1100 provided in an embodiment of the present application. Fig.11 The computing device 1100 includes a determination module 1110 and a processing module 1120.

[0235] The determination module 1110 may determine at least one initial coupling capacitance matrix, an initial resistance matrix and an initial capacitance matrix of each interconnection line according to a plurality of interconnections in the target circuit. The determination module 1110 may perform Figure 2 Step S210 and Figure 4 Step S410 in .

[0236] The processing module 1120 may determine the projection matrix of each interconnect line according to the initial resistance matrix of each interconnect line. The processing module 1120 may also reduce the order of at least one initial coupling capacitance matrix, the initial resistance matrix of each interconnect line, and the initial capacitance matrix according to the projection matrix of each interconnect line to determine the reduced-order circuit model of the target circuit. The processing module 1120 may perform Figure 2 Steps S220, S230, Figure 4 Steps S420 to S480 in .

[0237] Fig.12 It is a schematic structural block diagram of a computing device 1200 provided according to an embodiment of the present application. Fig.12The computing device 1200 shown includes: a processor 1201 , a memory 1202 , and a communication interface 1203 . The processor 1201 , the memory 1202 , and the communication interface 1203 communicate with each other via a bus 1204 .

[0238] In some embodiments, the computing device 1200 may further include a receiver and / or a transmitter. The receiver is used to receive information or data from other devices, and the transmitter is used to send information or data stored in the memory 1202 to other devices. The other devices are devices connected to the computing device 1200.

[0239] The method disclosed in the above embodiment of the present invention can be applied to the processor 1201, or implemented by the processor 1201. The processor 1201 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit in the processor 1201 or the instruction in the form of software. The processor 1201 can implement or execute the methods, steps and logic block diagrams disclosed in the embodiment of the present invention. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in the embodiment of the present invention can be directly embodied as a hardware decoding processor to execute, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in the memory 1202. The processor 1201 reads the instructions in the memory 1202 and completes the steps of the above method in combination with its hardware.

[0240] The memory 1202 may store instructions for executing the method executed by the computing device in the above-mentioned embodiment. The memory 1202 may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct rambus DRAM (DRDRAM). It should be noted that the memory of the system and method described herein is intended to include but is not limited to these and any other suitable types of memory. The processor 1201 can execute the instructions stored in the memory 1202 and complete the steps of the computing device in the above embodiment in combination with other hardware. The specific working process and beneficial effects can refer to the description in the above embodiment.

[0241] In addition to the data bus, the bus 1204 may also include a power bus, a control bus, a status signal bus, etc. However, for the sake of clarity, various buses are labeled as the bus 1204 in the figure.

[0242] An embodiment of the present application further provides a computer storage medium in which program instructions are stored. When the program is executed, the program may include some or all of the steps of the method in the above embodiment.

[0243] An embodiment of the present application also provides a chip system, which includes a logic circuit, which is used to couple with an input / output interface and transmit data through the input / output interface to execute the various steps included in the above embodiment.

[0244] According to the method provided in the embodiments of the present application, the present application also provides a computer program product, which includes: a computer program code, when the computer program code is run on a computer, the computer executes each step in the above embodiments.

[0245] According to the method provided in the embodiments of the present application, the present application also provides a computer-readable medium, which stores a program code. When the program code runs on a computer, the computer executes each step in the above embodiments.

[0246] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0247] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0248] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0249] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0250] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0251] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0252] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A circuit model reduction method, characterized in that: include: According to a plurality of interconnecting lines in a target circuit, at least one initial coupling capacitance matrix, an initial resistance matrix and an initial capacitance matrix of each interconnecting line are determined, each of the interconnecting lines includes a plurality of nodes, the initial coupling capacitance matrix is ​​used to indicate the coupling capacitance value between the nodes of two interconnecting lines, the initial resistance matrix is ​​used to indicate the resistance value between the nodes in the interconnecting lines corresponding to the initial resistance matrix, the initial capacitance matrix is ​​used to indicate the capacitance value between the nodes in the interconnecting lines corresponding to the initial capacitance matrix, and the target circuit is a resistance-capacitance RC circuit; Determine a projection matrix of each interconnect line according to an initial resistance matrix of each interconnect line, wherein the projection matrix of each interconnect line is used to reduce the order of one or more initial coupling capacitance matrices, initial resistance matrices and initial capacitance matrices of each interconnect line; According to the projection matrix of each interconnection line, the at least one initial coupling capacitance matrix, the initial resistance matrix and the initial capacitance matrix of each interconnection line are reduced in order to determine a reduced-order circuit model of the target circuit.

2. The method according to claim 1, characterized in that The circuit model of the target circuit before order reduction includes a capacitance matrix before order reduction, and the circuit model of the target circuit after order reduction includes a capacitance matrix after order reduction. The capacitance matrix before order reduction and the capacitance matrix after order reduction are block diagonal plus edge BBD matrices.

3. The method according to claim 1 or 2, characterized in that: The circuit model of the target circuit before order reduction includes a resistance matrix before order reduction, and the circuit model of the target circuit after order reduction includes a resistance matrix after order reduction. The resistance matrix before order reduction and the resistance matrix after order reduction are block diagonal matrices.

4. The method according to any one of claims 1 to 3, characterized in that The step of reducing the order of the at least one initial coupling capacitance matrix, the initial resistance matrix and the initial capacitance matrix of each of the interconnecting lines according to the projection matrix of each of the interconnecting lines to determine the reduced-order circuit model of the target circuit comprises: According to the projection matrix of each interconnection line, the initial resistance matrix of each interconnection line is reduced to determine the reduced-order resistance matrix of each interconnection line; A reduced-order resistance matrix is ​​determined according to the reduced-order resistance matrix of each interconnect line, wherein the reduced-order resistance matrix of each interconnect line is a diagonal block element in the reduced-order resistance matrix.

5. The method according to any one of claims 1 to 4, characterized in that The step of reducing the order of the at least one initial coupling capacitance matrix, the initial resistance matrix and the initial capacitance matrix of each of the interconnecting lines according to the projection matrix of each of the interconnecting lines to determine the reduced-order circuit model of the target circuit comprises: According to the projection matrix of each interconnection line, the initial capacitance matrix of each interconnection line is reduced to determine the reduced capacitance matrix of each interconnection line; According to the projection matrices of the plurality of interconnects, the at least one initial coupling capacitance matrix is ​​reduced in order to determine at least one reduced-order coupling capacitance matrix, each reduced-order coupling capacitance matrix corresponding to an initial coupling capacitance matrix; A reduced-order capacitance matrix is ​​determined according to the at least one reduced-order coupling capacitance matrix and the reduced-order capacitance matrix of each interconnection line.

6. The method according to claim 5, characterized in that The step of reducing the order of the at least one initial coupling capacitance matrix according to the projection matrix of the plurality of interconnects to determine at least one reduced-order coupling capacitance matrix comprises: A first initial coupling capacitance matrix is ​​reduced in order according to a projection matrix of a first interconnect line and a projection matrix of a second interconnect line to determine a first reduced-order coupling capacitance matrix, wherein the first interconnect line and the second interconnect line are different interconnect lines among the plurality of interconnect lines, and the first initial coupling capacitance matrix is ​​used to indicate coupling capacitance values ​​between nodes in the first interconnect line and the second interconnect line.

7. The method according to claim 5 or 6, characterized in that: The step of determining a reduced-order capacitance matrix according to the at least one reduced-order coupling capacitance matrix and the reduced-order capacitance matrix of each interconnect line comprises: Using the reduced-order capacitance matrix of each interconnect line as a diagonal block element in the reduced-order capacitance matrix; One or more second reduced-order coupling capacitor matrices are used as block elements in a first region in the reduced-order capacitor matrix, the second reduced-order coupling capacitor matrix corresponds to a third interconnect line and a fourth interconnect line, the third interconnect line is an interconnect line corresponding to a first diagonal block in the reduced-order capacitor matrix, the fourth interconnect line is an interconnect line among the multiple interconnect lines that is different from the third interconnect line, the first region is a row region or a column region where the first diagonal block is located, and the reduced-order capacitor matrix is ​​a symmetric matrix.

8. The method according to any one of claims 1 to 7, characterized in that Determining the projection matrix of each interconnection line according to the initial resistance matrix of each interconnection line comprises: Determine a projection matrix of each interconnection line according to an initial resistance matrix of each interconnection line; or, Determine a projection matrix of each interconnect line according to an initial resistance matrix and an initial capacitance matrix of each interconnect line; or, A projection matrix of each interconnection line is determined according to an initial input port matrix, an initial resistance matrix and an initial capacitance matrix of each interconnection line, wherein the initial input port matrix is ​​used to indicate input ports in the interconnection line corresponding to the initial input port matrix.

9. The method according to any one of claims 1 to 8, characterized in that The reduced-order circuit model of the target circuit further includes a reduced-order input port matrix and a reduced-order output port matrix, and the method further includes: Determine, according to the plurality of interconnections, an initial input port matrix and an initial output port matrix for each of the interconnections, wherein the initial input port matrix is ​​used to indicate input ports in the interconnections corresponding to the initial input port matrix, and the initial output port matrix is ​​used to indicate output ports in the interconnections corresponding to the initial output port matrix; According to the projection matrix of each interconnection line, respectively reduce the order of the initial input port matrix and the initial output port matrix of each interconnection line to determine the reduced order input port matrix and the reduced order output port matrix of each interconnection line; Determining the reduced-order input port matrix according to the reduced-order input port matrix of each interconnection line; The reduced-order output port matrix is ​​determined according to the reduced-order output port matrix of each interconnection line.

10. The method according to any one of claims 1 to 9, characterized in that The method further comprises: According to the input signal of each input port in the target circuit and the reduced-order circuit model, an output signal of each output port of the target circuit is determined, wherein the target circuit includes at least one input port and at least one output port.

11. A computing device, characterized in that: include: A determination module, configured to determine at least one initial coupling capacitance matrix, an initial resistance matrix and an initial capacitance matrix of each interconnection line according to a plurality of interconnections in a target circuit, wherein each interconnection line includes a plurality of nodes, the initial coupling capacitance matrix is ​​used to indicate a coupling capacitance value between nodes of two interconnections, the initial resistance matrix is ​​used to indicate a resistance value between nodes in the interconnections corresponding to the initial resistance matrix, and the initial capacitance matrix is ​​used to indicate a capacitance value between nodes in the interconnections corresponding to the initial capacitance matrix, wherein the target circuit is a resistor-capacitor RC circuit; A processing module, configured to determine a projection matrix of each interconnect line according to an initial resistance matrix of each interconnect line, wherein the projection matrix of each interconnect line is used to reduce the order of one or more initial coupling capacitance matrices, initial resistance matrices and initial capacitance matrices of each interconnect line; The processing module is further used to reduce the order of the at least one initial coupling capacitance matrix, the initial resistance matrix and the initial capacitance matrix of each of the interconnecting lines according to the projection matrix of each of the interconnecting lines, and determine the reduced-order circuit model of the target circuit.

12. The device according to claim 11, characterized in that The circuit model of the target circuit before order reduction includes a capacitance matrix before order reduction, and the circuit model of the target circuit after order reduction includes a capacitance matrix after order reduction. The capacitance matrix before order reduction and the capacitance matrix after order reduction are block diagonal plus edge BBD matrices.

13. The device according to claim 11 or 12, characterized in that The circuit model of the target circuit before order reduction includes a resistance matrix before order reduction, and the circuit model of the target circuit after order reduction includes a resistance matrix after order reduction. The resistance matrix before order reduction and the resistance matrix after order reduction are block diagonal matrices.

14. The device according to any one of claims 11 to 13, characterized in that The processing module is specifically used for: According to the projection matrix of each interconnection line, the initial resistance matrix of each interconnection line is reduced to determine the reduced-order resistance matrix of each interconnection line; A reduced-order resistance matrix is ​​determined according to the reduced-order resistance matrix of each interconnect line, wherein the reduced-order resistance matrix of each interconnect line is a diagonal block element in the reduced-order resistance matrix.

15. The device according to any one of claims 11 to 14, characterized in that The processing module is specifically used for: According to the projection matrix of each interconnection line, the initial capacitance matrix of each interconnection line is reduced to determine the reduced capacitance matrix of each interconnection line; According to the projection matrices of the plurality of interconnects, the at least one initial coupling capacitance matrix is ​​reduced in order to determine at least one reduced-order coupling capacitance matrix, each reduced-order coupling capacitance matrix corresponding to an initial coupling capacitance matrix; A reduced-order capacitance matrix is ​​determined according to the at least one reduced-order coupling capacitance matrix and the reduced-order capacitance matrix of each interconnection line.

16. The device according to claim 15, characterized in that The processing module is specifically used for: A first initial coupling capacitance matrix is ​​reduced in order according to a projection matrix of a first interconnect line and a projection matrix of a second interconnect line to determine a first reduced-order coupling capacitance matrix, wherein the first interconnect line and the second interconnect line are different interconnect lines among the plurality of interconnect lines, and the first initial coupling capacitance matrix is ​​used to indicate coupling capacitance values ​​between nodes in the first interconnect line and the second interconnect line.

17. The device according to claim 15 or 16, characterized in that The processing module is specifically used for: Using the reduced-order capacitance matrix of each interconnect line as a diagonal block element in the reduced-order capacitance matrix; One or more second reduced-order coupling capacitor matrices are used as block elements in a first region in the reduced-order capacitor matrix, the second reduced-order coupling capacitor matrix corresponds to a third interconnect line and a fourth interconnect line, the third interconnect line is an interconnect line corresponding to a first diagonal block in the reduced-order capacitor matrix, the fourth interconnect line is an interconnect line among the multiple interconnect lines that is different from the third interconnect line, the first region is a row region or a column region where the first diagonal block is located, and the reduced-order capacitor matrix is ​​a symmetric matrix.

18. The device according to any one of claims 11 to 17, characterized in that The processing module is specifically used for: Determine a projection matrix of each interconnection line according to an initial resistance matrix of each interconnection line; or, Determining a projection matrix of each interconnection line according to an initial resistance matrix and an initial capacitance matrix of each interconnection line; or, A projection matrix of each interconnection line is determined according to an initial input port matrix, an initial resistance matrix and an initial capacitance matrix of each interconnection line, wherein the initial input port matrix is ​​used to indicate input ports in the interconnection line corresponding to the initial input port matrix.

19. The device according to any one of claims 11 to 18, characterized in that The reduced-order circuit model of the target circuit also includes a reduced-order input port matrix and a reduced-order output port matrix. The determination module is further used to determine, based on the plurality of interconnections, an initial input port matrix and an initial output port matrix of each interconnection line, wherein the initial input port matrix is ​​used to indicate input ports in the interconnection line corresponding to the initial input port matrix, and the initial output port matrix is ​​used to indicate output ports in the interconnection line corresponding to the initial output port matrix; The processing module is further used to reduce the order of the initial input port matrix and the initial output port matrix of each interconnection line according to the projection matrix of each interconnection line, and determine the reduced order input port matrix and the reduced order output port matrix of each interconnection line; The processing module is further used to determine the reduced-order input port matrix according to the reduced-order input port matrix of each interconnection line; The processing module is further used to determine the reduced-order output port matrix according to the reduced-order output port matrix of each of the interconnection lines.

20. The device according to any one of claims 11 to 19, characterized in that The processing module is also used to determine the output signal of each output port of the target circuit according to the input signal of each input port in the target circuit and the reduced-order circuit model, wherein the target circuit includes at least one input port and at least one output port.

21. A computing device, characterized in that include: A processor, wherein the processor is configured to be coupled to a memory, read and execute instructions and / or program codes in the memory, so as to perform the method according to any one of claims 1 to 10.

22. A chip system, characterized in that: include: A logic circuit, the logic circuit is used to be coupled to an input / output interface, and transmit data through the input / output interface to execute the method according to any one of claims 1 to 10.

23. A computer readable medium, characterized in that The computer-readable medium stores a program code, and when the program code is executed on a computer, the computer is caused to execute the method according to any one of claims 1 to 10.