CN110610052A - PCB automatic wiring system and method - Google Patents

Abstract

The application provides a PCB automatic wiring system and method. The system includes: a schematic diagram module, a multiplexing processing module, an automatic rule module and an AI wiring module; the schematic diagram module is used to generate a netlist corresponding to a circuit schematic diagram; multiplexing processing The module is used to extract the module information corresponding to each functional module in the circuit schematic diagram from the netlist; the module information includes: module layout information, routing information and device parameters; the automatic rule module is used to according to the netlist and user preset rules, Generate the wiring rules corresponding to each functional module; the automatic rule module includes a rule generation model obtained after deep learning of the wiring rules of multiple circuit board diagrams; the AI wiring module is used to according to the module information and wiring rules corresponding to each functional module, Perform automatic wiring; the AI wiring module includes a wiring model obtained after deep learning of multiple functional modules and wiring rules. The application can improve the accuracy and efficiency of automatic wiring.

Description

PCB automatic wiring system and method

technical field

The present application relates to the technical field of integrated circuits, in particular to a PCB automatic wiring system and method.

Background technique

Wiring is a key step in PCB (Printed Circuit Board, printed circuit board) design, and its function is to connect components through wires to achieve specific functions. The traditional PCB design process takes a lot of time in the process of schematic design, PCB layout, etc. After the automatic wiring process, manual adjustments are required. The entire process is not highly automated, and most of the work is done manually, which is inefficient. For PCBs with complex structures, many layers, and many types of components, the results of automatic wiring are often difficult to meet the requirements of electrical performance.

The existing PCB design software is mainly composed of three parts, namely the schematic diagram part, the PCB part, and the simulation part, and these three parts are connected through a netlist. The PCB automatic routing part generally adopts traditional routing algorithms such as maze algorithm, line exploration method, etc. These algorithms occupy a large storage space when running, and have a large amount of calculation and low efficiency. It is difficult to achieve satisfactory results when processing PCBs with complex structures. .

Contents of the invention

The purpose of the present application is to provide a PCB automatic routing system and method to alleviate the technical problems of low routing efficiency and poor routing effect in the prior art.

The embodiment of the present application provides a PCB automatic wiring system, the system includes: a schematic diagram module, a multiplexing processing module, an automatic rule module and an AI wiring module;

The schematic diagram module is used to generate the netlist corresponding to the circuit schematic diagram; the netlist includes: device parameters, electrical connection relations, layout and rule information;

The multiplexing processing module is used to extract the module information corresponding to each functional module in the circuit schematic diagram from the netlist; the module information includes: module layout information, routing information and device parameters;

The automatic rule module is used to generate the wiring rules corresponding to each functional module according to the netlist and user preset rules; the automatic rule module includes a rule generation model obtained after deep learning of the wiring rules of multiple circuit board diagrams;

The AI wiring module is used to perform automatic wiring according to the module information and wiring rules corresponding to each functional module; the AI wiring module includes a wiring model obtained after deep learning of multiple functional modules and wiring rules.

Further, the automatic rule module is also used to extract rule information from the netlist, and input rule information and user preset rules into the rule generation model, so that the rule generation model outputs wiring rules corresponding to each functional module.

Further, the AI wiring module includes: a wiring model generation module, a wiring module and a post-processing module;

The AI wiring module is also used to generate the module through the wiring model, perform layout matching on the module information corresponding to each functional module, and perform rule matching on the wiring rules corresponding to each functional module, generate target wiring models corresponding to each functional module, and Call the routing module corresponding to each target routing model to perform routing operations, and output the routing results through the post-processing module.

Further, the system also includes: an evaluation module; the evaluation module includes an evaluation model generated by deep learning of multiple circuit wiring rules and layout information;

The AI wiring module is also used to send the current wiring information to the evaluation module in real time through the post-processing module;

The evaluation module is used to evaluate the current wiring information through the evaluation model, and return the evaluation information to the AI routing module, so that the AI routing module can modify the wiring model based on the evaluation information, and perform the wiring operation again until the evaluation information meets the predetermined requirements. set evaluation requirements.

Further, the system also includes: a simulation module;

The multiplexing processing module is also used to extract the simulation model and simulation parameters corresponding to each functional module from the netlist, and send the simulation model and simulation parameters corresponding to each functional module to the simulation module;

The AI routing module is also used to send the routing information to the simulation module after the automatic routing is completed; the routing information includes: line width, line length, position, via;

The simulation module, based on the simulation model, simulation parameters and wiring information corresponding to each functional module, simulates the PCB after the wiring is completed, and obtains the simulation result.

Further, the AI wiring module is also used to send the current wiring information to the evaluation module;

an automatic rule module, also used to send wiring rules to the evaluation module;

The simulation module is also used to send the simulation results to the evaluation module;

The evaluation module is also used to evaluate the current wiring information, wiring rules and simulation results through the evaluation model, and return the evaluation information to the AI wiring module, so that the AI wiring module can modify the wiring model based on the evaluation information and re-wire Operate until the evaluation information meets the preset evaluation requirements.

Further, the simulation module includes: an information processing module, a circuit analysis module and a display module;

The information processing module is used to perform comprehensive processing according to the simulation model, simulation parameters and wiring results, generate a global model of the PCB, and send the global model to the circuit analysis module;

The circuit analysis module is used to perform circuit analysis on the global model, and send the circuit analysis results to the display module;

The display module is used for displaying the circuit analysis results and outputting the simulation results.

Further, the circuit analysis module includes: a circuit dynamic analysis module and a circuit transient analysis module.

The embodiment of the present application also provides a PCB automatic wiring method, the method is applied in the PCB automatic wiring system, the PCB automatic wiring system includes: a schematic diagram module, a multiplexing processing module, an automatic rule module and an AI wiring module; the method includes:

The schematic diagram module is used to generate the netlist corresponding to the circuit schematic diagram; the netlist includes: device parameters, electrical connection relations, layout and rule information;

The multiplexing processing module is used to extract the module information corresponding to each functional module in the circuit schematic diagram from the netlist; the module information includes: module layout information, routing information and device parameters;

The automatic rule module is used to generate the wiring rules corresponding to each functional module according to the netlist and user preset rules; the automatic rule module includes a rule generation model obtained after deep learning of the wiring rules of multiple circuit board diagrams;

The AI wiring module is used to perform automatic wiring according to the module information and wiring rules corresponding to each functional module; the AI wiring module includes a wiring model obtained after deep learning of multiple functional modules and wiring rules.

Further, the PCB automatic routing system also includes: an evaluation module; the evaluation module includes an evaluation model generated by deep learning of multiple circuit routing rules and layout information; the method also includes:

The AI wiring module is also used to send the current wiring information to the evaluation module in real time;

The evaluation module is used to evaluate the current wiring information through the evaluation model, and return the evaluation information to the AI routing module, so that the AI routing module can modify the wiring model based on the evaluation information, and perform the wiring operation again until the evaluation information meets the predetermined requirements. set evaluation requirements.

The PCB automatic wiring system and method provided in the embodiments of the present application can realize a high-efficiency automatic wiring process through the joint action of the schematic diagram module, multiplexing processing module, automatic rule module and AI wiring module in the system. Among them, the schematic diagram module is used to generate the netlist corresponding to the circuit schematic diagram; the netlist includes: device parameters, electrical connection relations, layout and rule information; the multiplexing processing module is used to extract each functional module in the circuit schematic diagram from the netlist Corresponding module information; module information includes: module layout information, routing information and device parameters; automatic rule module is used to generate wiring rules corresponding to each functional module according to netlist and user preset rules; AI wiring module is used to The module information and wiring rules corresponding to each functional module are used for automatic wiring; among them, the automatic rule module includes a rule generation model obtained after deep learning of the wiring rules of multiple circuit board diagrams; the AI wiring module includes multiple functions The wiring model obtained after deep learning of modules and wiring rules, since both the automatic rule module and the AI wiring module include models obtained through deep learning, the accuracy and efficiency of automatic wiring can be improved.

Description of drawings

In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the accompanying drawings that need to be used in the description of the specific embodiments or prior art. Obviously, the accompanying drawings in the following description The drawings are some implementations of the present application, and those skilled in the art can obtain other drawings based on these drawings without creative work.

Fig. 1 is the flowchart of a kind of traditional PCB design in the prior art;

Fig. 2 is a kind of PCB design software block diagram in the prior art;

Fig. 3 is the block diagram of a kind of PCB automatic wiring system that the embodiment of the present application provides;

Fig. 4 is the block diagram of another kind of PCB automatic wiring system that the embodiment of the present application provides;

Fig. 5 is the working principle diagram of the automatic rule module in a kind of PCB automatic wiring system that the embodiment of the present application provides;

FIG. 6 is a working principle diagram of an AI wiring module in a PCB automatic wiring system provided in an embodiment of the present application;

FIG. 7 is a working principle diagram of a multiplexing processing module in a PCB automatic wiring system provided by an embodiment of the present application;

Fig. 8 is the working principle diagram of the simulation module in a kind of PCB automatic wiring system provided by the embodiment of the present application;

FIG. 9 is a working principle diagram of an evaluation module in a PCB automatic wiring system provided in an embodiment of the present application;

FIG. 10 is a layout diagram of PCB components in the prior art;

Figure 11 is a PCB component layout diagram provided by the embodiment of the present application;

Figure 12 is a functional block diagram of a PCB provided by the embodiment of the present application;

FIG. 13 is a flow chart of a PCB automatic wiring method provided by an embodiment of the present application.

Detailed ways

The technical solutions of the present application will be clearly and completely described below in conjunction with the embodiments. Apparently, the described embodiments are part of the embodiments of the present application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.

Wiring is a key step in PCB design, and its function is to connect components through wires to achieve specific functions. In PCB design, it is often desirable to use automatic routing to improve efficiency and save time. Figure 1 is the flow of traditional PCB design.

The traditional PCB design process takes a lot of time in the process of schematic design, PCB layout, etc. After the automatic wiring process, it needs to be adjusted manually according to the needs. The whole process is not highly automated, and most of the work is done manually. The efficiency low. For PCBs with complex structures, many layers, and many types of components, the results of automatic routing are often difficult to meet the requirements. At the same time, the electrical performance (such as frequency, electromagnetic compatibility, etc.) is not fully considered during the wiring process, which may make it difficult for the completed PCB to meet the requirements.

The block diagram of the existing PCB design software is shown in Figure 2. It is mainly composed of three parts, namely, the schematic part, the PCB part, and the simulation part. These three parts are connected through a netlist. The PCB automatic routing part generally adopts traditional routing algorithms such as maze algorithm, line exploration method, etc. These algorithms occupy a large storage space when running, and have a large amount of calculation and low efficiency. It is difficult to achieve satisfactory results when processing PCBs with complex structures. .

Based on this, an embodiment of the present application provides a PCB automatic routing system and method to alleviate the technical problems of low routing efficiency and poor routing effect in the prior art.

In order to facilitate the understanding of this embodiment, a PCB automatic routing system disclosed in this embodiment of the application is firstly introduced in detail.

FIG. 3 is a block diagram of a PCB automatic routing system provided by an embodiment of the present application. The system includes: a schematic diagram module 31, a multiplexing processing module 32, an automatic rule module 33 and an AI (Artificial Intelligence, artificial intelligence) routing module 34.

Wherein, the schematic diagram module 31 is respectively connected to the multiplexing processing module 32 and the automatic rule module 33 , and the multiplexing processing module 32 and the automatic rule module 33 are respectively connected to the AI wiring module 34 .

The schematic diagram module 31 is used to generate a netlist corresponding to the circuit schematic diagram; the netlist includes: device parameters, electrical connection relations, layout and rule information; the multiplexing processing module 32 is used to extract each element in the circuit schematic diagram from the netlist. The module information corresponding to the functional modules; the module information includes: module layout information, routing information and device parameters; the automatic rule module 33 is used to generate wiring rules corresponding to each functional module according to the netlist and user preset rules; the automatic rule The module includes a rule generation model obtained after deep learning of the wiring rules of multiple circuit board diagrams; the AI wiring module 34 is used to perform automatic wiring according to the module information and wiring rules corresponding to each functional module; the AI wiring module includes The wiring model obtained after deep learning of multiple functional modules and wiring rules.

During specific implementation, the main function of the above-mentioned schematic diagram module 31 is to draw a schematic circuit diagram and generate a netlist corresponding to the schematic diagram of the circuit. Unlike traditional netlists, in addition to including component attribute information and electrical connection relationships, The netlist also contains the layout and rule information of each functional module in the schematic diagram. In this embodiment, the function of the multiplexing processing module 32 is to extract module information corresponding to each functional module in the circuit schematic diagram from the netlist, such as: module layout information, routing information and device parameters.

Both the above-mentioned automatic rule module 33 and the AI wiring module 34 include models obtained through deep learning of samples, for example, the automatic rule module 33 includes a rule generation model obtained after deep learning of the wiring rules of multiple circuit board diagrams ; Specifically, the generation of the rule generation model is the result of deep learning, and its data source is a large number of circuit board diagrams. It uses unsupervised learning to learn and classify different circuit rules, and finally generates a model that can target each type of circuit. Generate a model of the corresponding rules. The AI wiring module 34 includes a wiring model obtained after deep learning of multiple functional modules and wiring rules. This wiring model is also the result of deep learning. The main process of training is similar to the rule generation model. The learning process is mainly based on the input modules and rules to train the best wiring model, through which the best wiring results can be achieved. . Furthermore, the automatic generation of rules and the automatic wiring process can be realized. For example, the automatic rule module 33 generates wiring rules corresponding to each functional module according to the netlist and user preset rules; the AI wiring module 34 generates wiring rules corresponding to each functional module according to the module information and wiring rules for automatic routing.

In the PCB automatic routing system provided by the embodiment of the present application, since both the automatic rule module and the AI routing module include models obtained through deep learning, the accuracy and routing efficiency of automatic routing can be improved.

In another preferred embodiment, a block diagram of another PCB automatic wiring system is provided, as shown in FIG. In addition to the module 43 and the AI wiring module 44 , it also includes: an evaluation module 45 and a simulation module 46 .

Wherein, the schematic diagram module 41 is respectively connected with the multiplexing processing module 42 and the automatic rule module 43, the multiplexing processing module 42 and the automatic rule module 43 are respectively connected with the AI wiring module 44, and the multiplexing processing module 42 is also connected with the simulation module 46 , the automatic rule module 43 is also connected to the evaluation module 45, the simulation module 46 is also connected to the evaluation module 45, and the simulation module 46 and the evaluation module 45 are respectively connected to the AI wiring module 44. The evaluation module 45 includes an evaluation model generated by performing deep learning on a plurality of circuit wiring rules and layout information. The generation of the evaluation model is also the result of deep learning. In the training process, a large number of circuit rules and layout information are analyzed to extract the corresponding evaluation model, which is optimized through testing and modification.

In this embodiment, the automatic rule module 43 is also used to extract rule information from the netlist, and input rule information and user preset rules into the rule generation model, so that the rule generation model outputs wiring rules corresponding to each functional module.

Specifically, as shown in FIG. 5 , the function of the automatic rule module 43 is to interpret the netlist, layout and some artificially set necessary rules input by the user, that is, after the above-mentioned user preset rules are interpreted, the specific input for this particular input is solved. Specific wiring rules. In addition to extracting rules from the netlist, users can also add custom rules for specific modules. After all the rules are processed by the automatic rule module 43, quantified routing rules are generated to provide to the AI routing module 44 for processing.

Further, the AI routing module 44 may specifically include: a routing model generation module, a routing module and a post-processing module. Among them, the AI wiring module 44 is also used to perform layout matching on the module information corresponding to each functional module through the wiring model generation module, and perform rule matching on the wiring rules corresponding to each functional module to generate target wiring corresponding to each functional module. model, and call the routing module corresponding to each target routing model to perform routing operations, and output the routing results through the post-processing module.

The above-mentioned AI wiring module 44 can also be used to send the current wiring information to the evaluation module 45 in real time through the post-processing module; the evaluation module 45 is used to evaluate the current wiring information through the evaluation model, and return the evaluation information to the AI wiring Module 44, so that the AI routing module 44 modifies the routing model based on the evaluation information, and re-performs the routing operation until the evaluation information meets the preset evaluation requirements.

The above-mentioned AI wiring module 44 can also send the wiring information to the simulation module 46 after the automatic wiring is completed; the wiring information includes: line width, line length, position, and via; the above-mentioned simulation module 46 can be based on each function module. The simulation model, simulation parameters and wiring information are used to simulate the PCB after the wiring is completed, and the simulation results are obtained.

In another preferred embodiment, the AI wiring module 44 can also send the current wiring information to the evaluation module 45; the automatic rule module 43 is also used to send the wiring rules to the evaluation module 45; the simulation module 46 is also used to The simulation result is sent to the evaluation module 45; the evaluation module 45 is also used to evaluate the current wiring information, wiring rules and simulation results through the evaluation model, and returns the evaluation information to the AI wiring module 44, so that the AI wiring module 44 is based on The evaluation information modifies the wiring model, and the wiring operation is performed again until the evaluation information meets the preset evaluation requirements.

In a preferred manner of this embodiment, the simulation module 46 includes: an information processing module, a circuit analysis module and a display module.

Among them, the information processing module is used to perform comprehensive processing according to the simulation model, simulation parameters and wiring results, generate a global model of the PCB, and send the global model to the circuit analysis module; the circuit analysis module is used to perform circuit analysis on the global model, And sending the circuit analysis result to the display module; the display module is used to display the circuit analysis result and output the simulation result. The above-mentioned circuit analysis module includes: a circuit dynamic analysis module and a circuit transient analysis module.

Specifically, FIG. 6 shows a working principle diagram of the AI routing module 44 provided by the embodiment of the present application. The AI routing module 44 uses an iterative method to route the PCB, and the unit of routing at this time is a module rather than a component. There are three input sources of the AI routing module 44 , which are respectively from the multiplex processing module 42 , the automatic rule module 43 and the evaluation module 45 . After the AI wiring module 44 performs layout matching and rule matching on the information of the input source, a wiring model is generated, and the wiring module is called to perform wiring. After each step or several steps, the post-processing module will output information to the evaluation module 45, and the evaluation module 45 will evaluate the current The wiring is evaluated, feedback information is obtained, and then sent back to the AI wiring module 44 to correct the wiring model, and so on, until all the wiring is completed.

In the case that the simulation module 46 is included in the system, the above-mentioned multiplexing processing module 42 may also include another function, which extracts simulation models and simulation parameters corresponding to each functional module from the netlist, and simulates the corresponding simulation parameters of each functional module. The model and simulation parameters are sent to the simulation module 46 .

As shown in FIG. 7 , the main function of the multiplexing processing module 42 is to extract the information of all functional modules from the netlist, and process the information as a whole to obtain the overall information of the PCB and the information of the simulation model. The overall PCB information includes the layout of the module, the wiring inside the module, the parameters of the components in the module, etc., and part of this information will be passed to the subsequent AI wiring module 44 as information input. The simulation model information includes the simulation model and simulation parameters of each independent module. These parameters usually contain various record results of the simulation and testing of the module in a complex environment. These parameters are necessary for the simulation module 46 .

As shown in FIG. 8 , the main function of the above-mentioned simulation module 46 is to simulate the PCB after wiring to check whether the design requirements are met. The input information of the simulation module 46 comes from the multiplex processing module 42 and the AI wiring module 44 respectively. The multiplexing processing module 42 provides the simulation model and simulation parameters inside the module, and the AI wiring module 44 provides wiring-related information such as line width, line length, position, vias, etc., and these information are synthesized by the information processing module to generate the layout of the PCB. The global model, passed to the circuit analysis section. The circuit analysis part includes transient analysis and steady state analysis. The analysis results are processed and displayed visually. Through simulation, the operating state of the entire circuit can be fully understood. For the parts that do not meet the design requirements, they can be fed back to the evaluation function. , pass the evaluation function to the wiring module to modify the current wiring, if the simulation result meets the requirements, the whole wiring process ends.

As shown in FIG. 9, the main function of the evaluation module 45 is to establish an evaluation model, perform global evaluation on the current wiring results, and output evaluation information to the AI wiring module 44, so as to improve the wiring results of the AI wiring module 44 until the design requirements are met. There are three input sources, namely, the input of the rules from the automatic rule module 43 , the simulation results from the simulation module 46 and the wiring information from the AI wiring module 44 .

Compared with the existing PCB wiring system, the PCB automatic wiring system provided by the embodiment of the present application has the following advantages:

1. High reusability. When designing a new PCB board, a large number of existing modules can be used for new PCB design;

2. High wiring efficiency. The wiring object of the wiring module is not a component, but a module, which greatly reduces the calculation scale and calculation time;

3. High reliability. It is possible to conduct a detailed study of each component constituting the module to confirm the optimal combination, which can improve efficiency and have high reliability at the same time.

The above-mentioned automatic rule module, AI wiring module and evaluation module adopt the idea of deep learning, realize the corresponding functions through training with a large amount of data, and ensure good electrical performance while wiring in the wiring process.

In addition, the embodiment of the present application also adopts the idea of modular PCB design, which is mainly reflected in the following two aspects:

1. Modularization of wiring system software

As mentioned above, the wiring system is designed as 6 modules, each module completes a specific function, and information is exchanged between modules through data flow.

2. Modularization of PCB layout

The traditional PCB layout takes components as the unit, as shown in Figure 10, if a group of components with specific functions is divided into a module, then the PCB board can be regarded as composed of many modules, which will greatly improve the Reduce the difficulty of layout, as shown in Figure 11.

For each functional module, it needs to be designed reasonably so that it can meet the requirements of modular PCB design, as shown in Figure 12: The PCB functional module is mainly composed of four parts, namely the schematic part, the PCB part, and the simulation part. Model section and rules section. The schematic part includes the schematic diagram and connection relationship of all components that make up the module, as well as the input and output interfaces; the PCB part includes the packaging and electrical connections of all the components that make up the module; the simulation model includes the internal wiring and electrical characteristics of the module The simulation model and data; the rule part contains the principles that the module should follow when connecting with other modules.

Based on the above-mentioned PCB automatic wiring system, the embodiment of the present application also provides a PCB automatic wiring method, which is applied to the PCB automatic wiring system. The PCB automatic wiring system includes: schematic diagram module, multiplex processing module, automatic rule module and AI Wiring module; The method comprises the following steps, as shown in Figure 13:

Step S101, the schematic diagram module is used to generate a netlist corresponding to the circuit schematic diagram; the netlist includes: device parameters, electrical connection relationships, layout and rule information;

Step S102, a multiplexing processing module, used to extract module information corresponding to each functional module in the circuit schematic diagram from the netlist; the module information includes: module layout information, routing information and device parameters;

Step S103, the automatic rule module is used to generate wiring rules corresponding to each functional module according to the netlist and user preset rules; the automatic rule module includes a rule generation model obtained after deep learning of the wiring rules of multiple circuit board diagrams ;

Step S104, the AI wiring module is used to perform automatic wiring according to the module information and wiring rules corresponding to each functional module; the AI wiring module includes a wiring model obtained after deep learning of multiple functional modules and wiring rules.

Further, the PCB automatic routing system also includes: an evaluation module; the evaluation module includes an evaluation model generated by deep learning of multiple circuit wiring rules and layout information; the above method also includes the following steps:

AI wiring module, used to send the current wiring information to the evaluation module in real time;

The evaluation module is used to evaluate the current wiring information through the evaluation model, and return the evaluation information to the AI routing module, so that the AI routing module can modify the wiring model based on the evaluation information, and perform the wiring operation again until the evaluation information meets the predetermined requirements. set evaluation requirements.

In another embodiment, the above-mentioned system also includes: a simulation module; the above-mentioned method also includes the following steps:

The multiplex processing module is used to extract the simulation model and simulation parameters corresponding to each functional module from the netlist, and send the simulation model and simulation parameters corresponding to each functional module to the simulation module;

The AI wiring module is used to send the wiring information to the simulation module after the automatic wiring is completed; the wiring information includes: line width, line length, position, via;

The simulation module is used to simulate the PCB after wiring based on the simulation model, simulation parameters and wiring information corresponding to each functional module, and obtain a simulation result.

In a preferred manner, the above-mentioned automatic rule module is also used to extract rule information from the netlist, and input the rule information and user preset rules into the rule generation model, so that the rule generation model outputs the wiring rules corresponding to each functional module .

The above-mentioned AI wiring module includes: a wiring model generation module, a wiring module and a post-processing module; the AI wiring module is also used to generate a module through the wiring model, perform layout matching on the module information corresponding to each functional module, and perform layout matching on each functional module. The wiring rules are matched according to the rules, and the target wiring models corresponding to each functional module are generated, and the wiring modules corresponding to each target wiring model are called to perform wiring operations, and the wiring results are output through the post-processing module.

The above method also includes:

The AI wiring module is also used to send the current wiring information to the evaluation module;

an automatic rule module, also used to send wiring rules to the evaluation module;

The simulation module is also used to send the simulation results to the evaluation module;

The evaluation module is also used to evaluate the current wiring information, wiring rules and simulation results through the evaluation model, and return the evaluation information to the AI wiring module, so that the AI wiring module can modify the wiring model based on the evaluation information and re-wire Operate until the evaluation information meets the preset evaluation requirements.

The above simulation module includes: an information processing module, a circuit analysis module and a display module; the information processing module is used to perform comprehensive processing according to the simulation model, simulation parameters and wiring results, generate a global model of the PCB, and send the global model to the circuit analysis module The circuit analysis module is used for circuit analysis on the global model, and the circuit analysis result is sent to the display module; the display module is used for displaying the circuit analysis result and outputting the simulation result. Wherein, the circuit analysis module includes: a circuit dynamic analysis module and a circuit transient analysis module.

For the implementation process of the method provided in the embodiment of the present application, reference may be made to the above-mentioned system embodiment, so details are not repeated here.

The PCB automatic wiring method provided by the embodiment of the present application can realize a high-efficiency automatic wiring process through the joint action of the schematic diagram module, the multiplexing processing module, the automatic rule module and the AI wiring module in the system. Among them, the schematic diagram module is used to generate the netlist corresponding to the circuit schematic diagram; the netlist includes: device parameters, electrical connection relations, layout and rule information; the multiplexing processing module is used to extract each functional module in the circuit schematic diagram from the netlist Corresponding module information; module information includes: module layout information, routing information and device parameters; automatic rule module is used to generate wiring rules corresponding to each functional module according to netlist and user preset rules; AI wiring module is used to The module information and wiring rules corresponding to each functional module are used for automatic wiring; among them, the automatic rule module includes a rule generation model obtained after deep learning of the wiring rules of multiple circuit board diagrams; the AI wiring module includes multiple functions The wiring model obtained after deep learning of modules and wiring rules, since both the automatic rule module and the AI wiring module include models obtained through deep learning, the accuracy and efficiency of automatic wiring can be improved.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit it; although the application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: It is still possible to modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the technical solutions of the various embodiments of the application. scope.

Claims (10)

1. An automatic routing system for a PCB, the system comprising: the device comprises a schematic diagram module, a multiplexing processing module, an automatic rule module and an AI wiring module;

the schematic diagram module is used for generating a netlist corresponding to the circuit schematic diagram; the netlist includes: device parameters, electrical connection relationships, layout and rule information;

the multiplexing processing module is used for extracting module information corresponding to each functional module in the circuit schematic diagram from the netlist; the module information includes: layout information, routing information and device parameters of the module;

the automatic rule module is used for generating a wiring rule corresponding to each functional module according to the netlist and a user preset rule; the automatic rule module comprises a rule generation model obtained by deep learning of wiring rules of a plurality of circuit board diagrams;

the AI wiring module is used for automatically wiring according to the module information and the wiring rule corresponding to each functional module; the AI wiring module comprises a wiring model obtained by deep learning of a plurality of functional modules and wiring rules.

2. The system of claim 1, wherein the automatic rule module is further configured to extract rule information from the netlist, and input the rule information and the user preset rule into the rule generation model, so that the rule generation model outputs a routing rule corresponding to each functional module.

3. The system of claim 1, wherein the AI routing module comprises: the system comprises a wiring model generation module, a wiring module and a post-processing module;

the AI wiring module is further configured to perform layout matching on the module information corresponding to each function module and perform rule matching on the wiring rules corresponding to each function module through the wiring model generation module to generate target wiring models corresponding to each function module, call the wiring module corresponding to each target wiring model to perform wiring operation, and output a wiring result through the post-processing module.

4. The system of claim 3, further comprising: an evaluation module; the evaluation module comprises an evaluation model generated by deep learning of a plurality of circuit wiring rules and layout information;

the AI wiring module is also used for sending the current wiring information to the evaluation module in real time through the post-processing module;

and the evaluation module is used for evaluating the current wiring information through the evaluation model and returning the evaluation information to the AI wiring module so that the AI wiring module modifies the wiring model based on the evaluation information and performs wiring operation again until the evaluation information meets the preset evaluation requirement.

5. The system of claim 4, further comprising: a simulation module;

the multiplexing processing module is further configured to extract a simulation model and a simulation parameter corresponding to each functional module from the netlist, and send the simulation model and the simulation parameter corresponding to each functional module to the simulation module;

the AI wiring module is also used for sending wiring information to the simulation module after the automatic wiring is finished; the wiring information includes: line width, line length, position, via hole;

and the simulation module is used for simulating the PCB after wiring is finished based on the simulation model and the simulation parameters corresponding to each functional module and the wiring information to obtain a simulation result.

6. The system of claim 5,

the AI wiring module is also used for sending the current wiring information to the evaluation module;

the automatic rule module is also used for sending the wiring rule to the evaluation module;

the simulation module is also used for sending a simulation result to the evaluation module;

the evaluation module is further configured to evaluate the current wiring information, the wiring rule, and the simulation result through the evaluation model, and return evaluation information to the AI wiring module, so that the AI wiring module modifies the wiring model based on the evaluation information and performs wiring operation again until the evaluation information meets a preset evaluation requirement.

7. The system of claim 5, wherein the simulation module comprises: the circuit comprises an information processing module, a circuit analysis module and a display module;

the information processing module is used for carrying out comprehensive processing according to the simulation model, the simulation parameters and the wiring result to generate a global model of the PCB and sending the global model to the circuit analysis module;

the circuit analysis module is used for carrying out circuit analysis on the global model and sending a circuit analysis result to the display module;

and the display module is used for displaying the circuit analysis result and outputting a simulation result.

8. The system of claim 7, wherein the circuit analysis module comprises: the circuit dynamic analysis module and the circuit transient analysis module.

9. A PCB automatic wiring method is characterized in that the method is applied to a PCB automatic wiring system, and the PCB automatic wiring system comprises: the device comprises a schematic diagram module, a multiplexing processing module, an automatic rule module and an AI wiring module; the method comprises the following steps:

the schematic diagram module is used for generating a netlist corresponding to the circuit schematic diagram; the netlist includes: device parameters, electrical connection relationships, layout and rule information;

the multiplexing processing module is used for extracting module information corresponding to each functional module in the circuit schematic diagram from the netlist; the module information includes: layout information, routing information and device parameters of the module;

the automatic rule module is used for generating a wiring rule corresponding to each functional module according to the netlist and a user preset rule; the automatic rule module comprises a rule generation model obtained by deep learning of wiring rules of a plurality of circuit board diagrams;

the AI wiring module is used for automatically wiring according to the module information and the wiring rule corresponding to each functional module; the AI wiring module comprises a wiring model obtained by deep learning of a plurality of functional modules and wiring rules.

10. The method of claim 9, wherein the PCB automated routing system further comprises: an evaluation module; the evaluation module comprises an evaluation model generated by deep learning of a plurality of circuit wiring rules and layout information; the method further comprises the following steps:

the AI wiring module is also used for sending the current wiring information to the evaluation module in real time;

and the evaluation module is used for evaluating the current wiring information through the evaluation model and returning the evaluation information to the AI wiring module so that the AI wiring module modifies the wiring model based on the evaluation information and performs wiring operation again until the evaluation information meets the preset evaluation requirement.