CN108595815B - Artificial intelligence body training system and passive circuit optimization design system and method - Google Patents
Artificial intelligence body training system and passive circuit optimization design system and method Download PDFInfo
- Publication number
- CN108595815B CN108595815B CN201810350393.7A CN201810350393A CN108595815B CN 108595815 B CN108595815 B CN 108595815B CN 201810350393 A CN201810350393 A CN 201810350393A CN 108595815 B CN108595815 B CN 108595815B
- Authority
- CN
- China
- Prior art keywords
- circuit
- design
- model
- optimization
- artificial intelligence
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/39—Circuit design at the physical level
- G06F30/398—Design verification or optimisation, e.g. using design rule check [DRC], layout versus schematics [LVS] or finite element methods [FEM]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
- G06F30/36—Circuit design at the analogue level
- G06F30/367—Design verification, e.g. using simulation, simulation program with integrated circuit emphasis [SPICE], direct methods or relaxation methods
Landscapes
- Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Design And Manufacture Of Integrated Circuits (AREA)
Abstract
The embodiment of the application provides an artificial intelligence body training system and a passive circuit optimal design system and method, by inputting a circuit model to be optimized and required circuit model design indexes, the artificial intelligence body can execute circuit optimal design processing aiming at the circuit model according to the circuit model design indexes, and automatically generate a circuit optimal model according with the circuit model design indexes, so that the sharing of different types of passive circuit design experiences can be realized, the reusability of circuit design is improved, and the system and the method have the advantages of high design efficiency and low design cost.
Description
Technical Field
The embodiment of the application relates to a passive circuit design and development technology, in particular to an artificial intelligence training system, a passive circuit optimization design system based on artificial intelligence and a method thereof.
Background
Currently, passive circuits are widely used in the electronic field, and the passive circuits are circuits composed of only basic elements such as RCL (i.e., a resistor R, an inductor L, and a capacitor C). The existing passive circuit design has the following problems:
firstly, the design of the passive circuit needs a designer with rich experience, and the whole design process mainly comprises the steps of establishing a simulation model in a computer simulation design tool, generating a calculation result, debugging and optimizing the circuit. However, the design flow of current passive circuits is scattered and all is achieved manually.
Secondly, since the passive circuits have a large variety, they have a high design threshold, and for designers with little or no design experience, the trouble that it is difficult to perform circuit design due to limited design experience is likely to occur.
Moreover, the passive circuit can generate highly relevant data in the process of design and optimization, but at present, no system or method can automatically learn the rule of the data generated in the circuit design, so that the working efficiency of the current circuit design is low, and the development period is long.
In summary, since different types of passive circuits in the prior art have common design and debug characteristics, but none of the systems and methods can indicate how to perform migration debugging of different types of circuits, so that the reusability of the prior circuit design method is low, and how to improve the above problem is the technical subject to be solved by the present application.
Disclosure of Invention
In order to solve the defects in the prior art, the invention mainly aims to provide an artificial intelligence training system, a passive circuit optimization design system and a passive circuit optimization design method, which can enable the artificial intelligence to have the automatic design function of a passive circuit so as to realize the automatic optimization design of the passive circuit by utilizing deep reinforcement learning.
Another objective of the present invention is to provide a system and a method for artificial intelligence training and passive circuit optimization design, which have the advantages of high design efficiency, high reusability and low design cost.
To achieve the above and other related objects, a first embodiment of the present application provides an artificial intelligence training system, applied to an artificial intelligence, for training a passive circuit of the artificial intelligence to optimize design performance, comprising: a receiving module for receiving the circuit model; the setting module is used for providing design indexes of the set circuit model; the training optimization design module is used for executing circuit optimization design processing aiming at the circuit model according to the circuit model design index and generating a circuit optimization model; and the calculation analysis module is used for calculating and analyzing whether the simulation operation result of the circuit optimization model accords with the circuit model design index so as to output an analysis result, so that the training optimization design module trains and updates the circuit optimization design processing executed by the training optimization design module according to the analysis result.
Optionally, in any embodiment of the present application, the circuit model design index includes a dielectric material type parameter, a circuit element type and size parameter, a conductor type and size parameter, an input interface parameter, and an output interface parameter.
Optionally, in any embodiment of the present application, the artificial intelligence training system further comprises a database storing initialization parameters for circuit elements constituting the circuit model.
Optionally, in any embodiment of the present application, the training optimization design module performs at least one circuit optimization design process on the circuit model according to the initialization parameters of the circuit elements in the library and the circuit model design indexes.
Optionally, in any embodiment of the application, the calculation analysis module further includes a step of performing, by the training optimization design module, the circuit optimization model generated by the circuit optimization design processing each time, so as to generate at least one simulation operation result, and determining whether the simulation operation result of the circuit optimization model generated by the training optimization design module at present is closer to the circuit model design index than the simulation operation result of the circuit optimization model generated at the previous time, if the analysis result is yes, outputting a positive feedback signal, and otherwise, outputting a negative feedback signal.
Optionally, in any embodiment of the application, the training optimization design module further includes a step of recording, according to the positive feedback signal or the negative feedback signal output by the calculation and analysis module, historical data of an influence of adjustment of each circuit element on the circuit model design index in each circuit optimization design process and real-time data of an influence of adjustment of each circuit element on the circuit model design index in a simulation operation of the circuit optimization model.
Optionally, in any embodiment of the application, the training optimization design module further includes an operation of determining, according to the positive feedback signal or the negative feedback signal output by the calculation and analysis module, an operation of next circuit optimization adjustment processing performed on the circuit model, so that the probability of the calculation and analysis module outputting the positive feedback signal is continuously increased as an optimization design reference.
A second embodiment of the present application further provides a passive circuit optimization design system, implemented based on artificial intelligence, including: a receiving unit for receiving a circuit model to be optimized; the setting unit is used for providing a design index of a set circuit model; and an artificial agent trained by the artificial agent training system of the first embodiment; and the artificial intelligence body carries out at least one time of optimization design treatment on the received circuit model to be optimized according to the set circuit model design index so as to generate a circuit optimization model meeting the circuit model design index.
A second embodiment of the present application further provides a passive circuit optimal design method for performing optimal design of a passive circuit by an artificial intelligence, the method including: inputting a circuit model to be optimized and inputting a circuit model design index into the artificial intelligence body; enabling the artificial intelligence body to execute circuit optimization design processing aiming at the circuit model according to the circuit model design index so as to generate a circuit optimization model; simulating and operating the circuit optimization model generated by the artificial intelligent agent to generate a simulation operation result, analyzing whether the simulation operation result meets the circuit model design index or not, and outputting an analysis result; and enabling the artificial intelligence body to receive the analysis, repeatedly executing the circuit optimization design processing to generate a new circuit optimization model when the analysis result is that the simulation operation result does not accord with the circuit model design index, and ending the execution of the circuit optimization design processing when the analysis result is that the simulation operation result accords with the circuit model design index.
Optionally, in any embodiment of the present application, the circuit model design index includes a dielectric material type parameter, a circuit element type and size parameter, a conductor type and size parameter, an input interface parameter, and an output interface parameter.
Optionally, in any embodiment of the present application, the artificial intelligence body further stores initialization parameters of circuit elements constituting the circuit model, and the artificial intelligence body performs at least one circuit optimization design process on the circuit model according to the stored initialization parameters of the circuit elements and the circuit model design index.
Optionally, in any embodiment of the present application, the method further includes: simulating and operating the currently generated circuit optimization model to generate a simulation operation result; enabling the artificial intelligence body to judge whether the simulation operation result of the circuit optimization model generated at present is closer to the circuit model design index than the simulation operation result of the circuit optimization model generated at the previous time, if the simulation operation result is yes, outputting a positive feedback signal, otherwise, outputting a negative feedback signal; and enabling the artificial intelligence body to continuously increase the probability of the output positive feedback signal as an optimization design reference according to the output positive feedback signal or negative feedback signal, judging an operation scheme of next circuit optimization design processing to be executed aiming at the circuit model, and executing the next circuit optimization design processing aiming at the circuit model so as to generate a new circuit optimization model.
Optionally, in any embodiment of the present application, the method further includes causing the artificial intelligence body to record historical data of an influence of the adjustment of each circuit element on the circuit model design index during each circuit optimization design process and real-time data of an influence of the adjustment of each circuit element on the circuit model design index during simulation runtime of the circuit optimization model.
Therefore, the artificial intelligence training system provided by the application has the function of automatically optimizing and designing the passive circuit by acquiring the design experience of a passive circuit designer, historical data of the circuit optimization process, real-time data of the passive circuit in the simulation operation process and other information.
Moreover, the application also provides a passive circuit optimization design system and a passive circuit optimization design method, which can realize automatic optimization design of the passive circuit based on artificial intelligence and have the advantages of high design efficiency, high reusability and low design cost.
Drawings
Some specific embodiments of the present application will be described in detail hereinafter by way of illustration and not limitation with reference to the accompanying drawings. The same reference numbers in the drawings identify the same or similar elements or components. Those skilled in the art will appreciate that the drawings are not necessarily drawn to scale. In the drawings:
FIG. 1 is a schematic diagram of a basic architecture of a human agent training system according to an embodiment of the present application;
FIG. 2 is a schematic diagram showing various embodiments of the human agent training system of the present application;
FIG. 3 is a schematic diagram of a basic architecture of a passive circuit optimization design system according to another embodiment of the present application;
FIG. 4 is a schematic diagram of a basic flow of a passive circuit optimization design method according to another embodiment of the present application; and
FIG. 5 is a flowchart showing an embodiment of the method of the passive circuit optimization design of FIG. 4.
Detailed Description
It is not necessary for any particular embodiment of the invention to achieve all of the above advantages at the same time.
In order to make those skilled in the art better understand the technical solutions in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be described clearly and completely with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments obtained by a person skilled in the art based on the embodiments of the present invention shall fall within the scope of the protection of the embodiments of the present invention.
The following further describes specific implementation of the embodiments of the present invention with reference to the drawings.
Fig. 1 is a schematic diagram of a basic architecture of a human body agent training system according to an embodiment of the present application. As shown in the figure, the artificial intelligence body training system 10 of the present application is applied to an artificial intelligence body 1, and is mainly used for training the passive circuit optimization design function of the artificial intelligence body 1, and mainly includes a receiving module 11, a setting module 12, a training optimization design module 13, and a calculation analysis module 14.
The receiving module 11 is used for receiving the circuit model. In the embodiments of the present application, the received circuit model refers to a circuit model to be optimally designed, which does not meet the design criteria of the circuit model.
The setting module 12 is used for providing a circuit designer with a desired circuit model design index. In the embodiment of the present application, the circuit model design criteria input by the circuit designer include dielectric material type parameters, circuit element type and dimension parameters, conductor type and dimension parameters, input interface parameters, and output interface parameters.
The training optimization design module 13 is configured to perform a circuit optimization design process on the input circuit model according to the circuit model design index to generate a corresponding circuit optimization model.
Referring to fig. 2, in the embodiment of the present application, the artificial intelligence training system 10 further includes a database 15 storing initialization parameters of various types of circuit elements used to form a circuit model, and the training optimization design module 13 is capable of performing at least one circuit optimization design process on the circuit model according to the initialization parameters of various types of circuit elements stored in the database 15 and the input circuit model design indexes.
The calculation and analysis module 14 is configured to analyze whether a simulation operation result of the circuit optimization model generated by the training and optimization design module 13 meets a circuit model design index received by the receiving module 11, and accordingly output an analysis result, so that the training and optimization design module 13 trains and updates the circuit optimization design process executed by the training and optimization design module according to the analysis result, and executes the next circuit optimization design process.
In this embodiment, the calculation and analysis module 14 is configured to simulate and run the circuit optimization model generated by the training and optimization design module 13 to generate a corresponding simulation running result, and analyze whether the simulation running result meets the circuit model design index received by the receiving module 11, and when the analysis result is that the simulation running result does not meet the circuit model design index, the training and optimization design module 13 will repeatedly execute the operation of the circuit optimization design process until the calculation and analysis module 14 determines that the simulation running result of the generated circuit optimization model can meet the circuit model design index.
Specifically, the calculation analysis module 14 is used to simulate and run a circuit optimization model generated by the training optimization design module 13 each time the circuit optimization design process is executed, to generate at least one simulation operation result, and compare whether the simulation operation result of the circuit optimization model generated by the training optimization design module 13 at present is closer to the circuit model design index set by the circuit designer than the simulation operation result of the circuit optimization model generated at the previous time, if the analysis result is yes, the operation scheme representing the circuit optimal design process currently performed by the training optimal design module 13 is valid, a positive feedback signal (e.g., a reward signal) is output, and if the analysis result is negative, a negative feedback signal (e.g., a penalty signal) is output, which represents that the operation scheme of the circuit optimization design process currently executed by the training optimization design module 13 is in question. It should be noted that, the standard of the reward and punishment signal output by the calculation and analysis module 14 is not limited to the above technical solution, and in other embodiments, the experience of a circuit designer may also be used as one of the standards for determining the magnitude of the reward and punishment signal during reinforcement learning, for example: in the process of optimizing and debugging the circuit, when it is determined that the parameter of a certain circuit element in the circuit model should be adjusted downward based on the experience of the circuit designer, and the optimization design process performed by the training optimization design module 13 also adjusts the parameter of the circuit element downward, the calculation and analysis module 14 outputs a reward signal, otherwise outputs a penalty signal.
The training optimization design module 13 may record historical data of the influence of the adjustment of each circuit element on the circuit model design index and real-time data of the influence of the adjustment of each circuit element on the circuit model design index during the simulation operation of the circuit optimization model according to the positive feedback signal or the negative feedback signal output by the calculation and analysis module 14 in each circuit optimization design process. Meanwhile, the training optimization design module 13 may further determine, according to the positive feedback signal or the negative feedback signal output by the calculation and analysis module 14, that the probability that the calculation and analysis module 14 outputs the positive feedback signal is continuously increased as an optimization design reference, and determine the operation of the next circuit optimization adjustment process executed for the circuit model.
For example, the training optimization design module 13 analyzes the variation rule between the simulation operation result (experimental data) and the circuit model design index (theoretical data) according to the output positive feedback signal or negative feedback signal, so as to modify the design parameters of the related circuit elements, to execute the circuit optimal design processing operation, after the circuit model receives the circuit optimal design processing, the simulation operation result (simulation result) will also change correspondingly, the calculation and analysis module 14 can generate a reinforcement signal (reward or punishment) based on the result of the simulation result and feed back the reinforcement signal to the training optimization design module 13, the training optimization design module 13 will select the next circuit optimization design processing operation to be executed again according to the feedback reward and punishment signal and the operation environment, the principle of selecting the next circuit optimization operation to be performed is to increase the probability of the reward signal given by the calculation and analysis module 14. If a circuit optimal design processing operation performed by the training optimal design module 13 results in a positive reward, the tendency to perform this circuit optimal design processing operation later on is strengthened; otherwise, the trend of performing this circuit optimization design processing operation later will be reduced. By means of the above technical means, reinforcement learning is performed for the optimal design function of the training optimal design module 13. Therefore, by the repeated interaction between the training optimization design module 13 and the calculation analysis module 14, the training optimization design module 13 can obtain the mapping characteristics of the passive circuit design based on deep reinforcement learning, and learn the optimal decision of the relevant decision of the circuit design by using the reinforcement learning technology, so as to accelerate the efficiency of building different types of passive circuit design models by adopting transfer learning.
Fig. 3 is a schematic diagram of a basic architecture of a passive circuit optimization design system according to another embodiment of the present application. As shown in the figure, the passive circuit optimization design system 20 provided by the present application mainly includes a receiving unit 21, a setting unit 22, and an artificial intelligence 23.
The receiving unit 21 is used for receiving a circuit model to be optimized, the setting unit 22 is used for providing circuit designer with setting circuit model design indexes, the artificial intelligence body 23 is generated by training of the artificial intelligence body training system 10 shown in fig. 1 or fig. 2, and has a passive circuit optimization design function, and the artificial intelligence body can perform at least one time of optimization design processing on the circuit model to be optimized received by the receiving unit 21 according to the circuit model design indexes set by the setting unit 22 so as to automatically generate a circuit optimization model with simulation operation results conforming to the circuit model design indexes.
Fig. 4 is a schematic diagram of a basic flow of a passive circuit optimization design method according to another embodiment of the present application. As shown in the drawings, the method for optimally designing a passive circuit according to the present application executes an optimal design of a passive circuit by an artificial intelligence, and mainly includes the following processing steps:
step S41, the circuit model to be optimized is input to the artificial intelligence, and the circuit model design index is input, followed by step S42. In the embodiment of the present application, the input circuit model design criteria include dielectric material type parameters, circuit element type and dimension parameters, conductor type and dimension parameters, input interface parameters, and output interface parameters.
Step S42, the artificial intelligence is enabled to execute a circuit optimization design process for the circuit model according to the circuit model design index to generate a circuit optimization model, and then step S43 is executed. In an embodiment of the present application, the artificial intelligence body further stores initialization parameters of circuit elements used for forming the circuit model, and the artificial intelligence body performs at least one circuit optimization design process on the circuit model according to the stored initialization parameters of the circuit elements and the circuit model design indexes.
Step S43 is to simulate and operate the circuit optimization model generated by the artificial intelligence to generate a simulation operation result, analyze whether the simulation operation result of the circuit optimization model meets the circuit model design index, and output the analysis result (please refer to details in fig. 5), and then execute step S44.
And step S44, enabling the artificial intelligence body to receive the analysis result, and when the analysis result is that the simulation operation result does not accord with the circuit model design index, repeating the executed circuit optimization design processing to generate a new circuit optimization model, and ending the execution of the circuit optimization design processing until the analysis result is that the simulation operation result accords with the circuit model design index.
Fig. 5 is a flowchart showing an embodiment of the method for optimizing a design of a passive circuit of fig. 4.
As shown, step S51 is first performed to simulate and run the circuit optimization model generated by the artificial intelligence to generate a simulation run result, and step S52 is then performed.
And step S52, judging whether the simulation operation result of the generated circuit optimization model accords with the circuit model design index, and when the judgment result accords with the circuit model design index, outputting the generated circuit optimization model by the artificial intelligence body, and finishing the optimization design processing aiming at the circuit model to be optimized. If the determination result is not in agreement, the process proceeds to step S53.
In step S53, the artificial intelligence determines whether the simulation operation result of the currently generated circuit optimization model is closer to the circuit model design index than the simulation operation result of the circuit optimization model generated in the previous time, if the determination result is no, step S641 is performed, and if the determination result is yes, step S642 is performed.
In step S541, if the determination result is no, it indicates that the currently generated circuit optimization model of the artificial intelligence is unsuccessful, or the currently executed circuit optimization design process is not ideal, a negative feedback signal (e.g., a penalty signal) is output, and then step S55 is executed.
In step S542, if the determination result is yes, it represents that the currently generated circuit optimization model of the artificial intelligence has achieved the optimization effect, that is, the currently executed circuit optimization design process is successful, a forward feedback signal (e.g., a reward signal) is output, and then step S55 is executed.
Step S55, the artificial intelligence determines the operation scheme of the next circuit optimization design process to be executed for the circuit model according to the received positive feedback signal or negative feedback signal, so that the probability of the output positive feedback signal is continuously increased as the optimization design reference, executes the next circuit optimization design process for the circuit model to generate a new circuit optimization model, and returns to step S51.
In an embodiment of the present application, step S55 further includes enabling the artificial intelligence to record historical data of the impact of the adjustment of each circuit element on the circuit model design index during each circuit optimization design process and real-time data of the impact of the adjustment of each circuit element on the circuit model design index during the simulation run of the circuit optimization model. For example, the artificial intelligence analyzes the variation between the simulation operation result (experimental data) and the circuit model design index (theoretical data) according to the output positive feedback signal or negative feedback signal, so as to modify the design parameters of the related circuit components, i.e. to execute the circuit optimization design processing operation, and the principle of selecting the next circuit optimization processing operation to be executed is to increase the probability of the given reward signal. If a circuit optimization design process performed by the artificial intelligence results in a positive reward, then the tendency to perform this circuit optimization design process at a later time is enhanced; otherwise, the trend of performing this circuit optimization design processing operation later will be reduced.
In summary, the artificial intelligence training system provided by the application trains the passive circuit optimization design function of artificial intelligence by digitizing the design experience of the passive circuit designer, and collecting the historical data of the influence of the adjustment of each component on the design index in the circuit optimization process and the real-time data of the influence of the adjustment of each component on the design index in the online simulation of the passive circuit.
Meanwhile, the application also discloses a passive circuit optimization design system and a passive circuit optimization design method, which adopt artificial intelligence and reinforcement learning technology to realize automatic optimization design of the passive circuit, have the advantages of high design efficiency, high reusability and low design cost, can reduce the risk and uncertainty of the design and implementation process, and make the circuit easy to customize.
The above-described embodiments of the apparatus are merely illustrative, wherein the modules described as separate parts may or may not be physically separate, and the parts displayed as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions and/or portions thereof that contribute to the prior art may be embodied in the form of a software product that can be stored on a computer-readable storage medium including any mechanism for storing or transmitting information in a form readable by a computer (e.g., a computer). For example, a machine-readable medium includes Read Only Memory (ROM), Random Access Memory (RAM), magnetic disk storage media, optical storage media, flash memory storage media, electrical, optical, acoustical or other form of propagated signals (e.g., carrier waves, infrared signals, digital signals, etc.), and others, and the computer software product includes instructions for causing a computing device (which may be a personal computer, server, or network device, etc.) to perform the methods described in the various embodiments or portions of the embodiments.
Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the embodiments of the present application, and are not limited thereto; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus (device), or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (devices) and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
Claims (9)
1. An artificial intelligence body training system, applied to an artificial intelligence body, for training passive circuit optimal design performance of the artificial intelligence body, comprising:
a receiving module for receiving the circuit model;
the setting module is used for providing design indexes of the set circuit model;
a training optimization design module, configured to perform at least one circuit optimization design process on the circuit model according to the circuit model design index and initialization parameters of circuit elements constituting the circuit model, and generate a circuit optimization model; and
the calculation analysis module is used for calculating and analyzing whether a simulation operation result of the circuit optimization model meets the circuit model design index or not and outputting an analysis result, wherein the calculation analysis module is also used for simulating and operating the circuit optimization model generated by the circuit optimization design processing executed by the training optimization design module each time so as to generate at least one simulation operation result, judging whether the simulation operation result of the circuit optimization model generated by the training optimization design module at present is closer to the circuit model design index than the simulation operation result of the circuit optimization model generated by the training optimization design module at the previous time or not, if the analysis result is positive, outputting a positive feedback signal, otherwise, outputting a negative feedback signal; and wherein the one or more of the one,
the training optimization design module further comprises a step of determining the next operation of circuit optimization adjustment processing executed for the circuit model according to the positive feedback signal or the negative feedback signal output by the calculation analysis module so that the probability of the positive feedback signal output by the calculation analysis module is continuously increased as an optimization design reference, and the circuit optimization design processing is finished when the calculation analysis module analyzes that the simulation operation result meets the circuit model design reference.
2. The artificial intelligence training system of claim 1 wherein the circuit model design criteria include dielectric material type parameters, circuit element type and size parameters, conductor type and size parameters, input interface parameters, and output interface parameters.
3. The artificial intelligence training system of claim 2 further comprising a database for storing initialization parameters for the circuit elements comprising the circuit model.
4. The artificial intelligence training system of claim 1, wherein the training optimization design module further includes a history data recording an influence of the adjustment of each circuit component on the circuit model design index in each circuit optimization design process and a real-time data recording an influence of the adjustment of each circuit component on the circuit model design index in simulation operation according to the positive feedback signal or the negative feedback signal output by the calculation analysis module.
5. A passive circuit optimization design system implemented based on artificial intelligence, comprising:
a receiving unit for receiving a circuit model to be optimized;
the setting unit is used for providing a design index of a set circuit model; and
an artificial agent generated by training of the artificial agent training system according to any of claims 1 to 4; wherein the content of the first and second substances,
and the artificial intelligence body carries out at least one time of optimization design treatment on the received circuit model to be optimized according to the set circuit model design index so as to generate a circuit optimization model meeting the circuit model design index.
6. A method for optimally designing a passive circuit, the method comprising the steps of:
inputting a circuit model to be optimized and inputting a circuit model design index into the artificial intelligence body;
enabling the artificial intelligence body to execute at least one time of circuit optimization design processing aiming at the circuit model according to the circuit model design index and the initialization parameter of the circuit element forming the circuit model so as to generate a circuit optimization model;
simulating and operating the artificial intelligence body to execute the circuit optimization model generated by the circuit optimization design processing each time so as to generate a simulation operation result;
enabling the artificial intelligence body to judge whether the simulation operation result of the circuit optimization model generated at present is closer to the circuit model design index than the simulation operation result of the circuit optimization model generated at the previous time, if the simulation operation result is yes, outputting a positive feedback signal, otherwise, outputting a negative feedback signal; and
and enabling the artificial intelligence body to continuously increase the probability of the output positive feedback signal as an optimization design reference according to the output positive feedback signal or negative feedback signal, judging an operation scheme of next circuit optimization design processing to be executed aiming at the circuit model, executing the next circuit optimization design processing aiming at the circuit model to generate a new circuit optimization model, and ending the execution of the circuit optimization design processing when the simulation operation result meets the circuit model design reference.
7. The method of claim 6, wherein the circuit model design criteria include dielectric material type parameters, circuit element type and dimension parameters, conductor type and dimension parameters, input interface parameters, and output interface parameters.
8. A passive circuit optimization design method according to claim 6, characterized in that the artificial intelligence is further adapted to store initialization parameters of circuit elements constituting the circuit model.
9. The method of claim 6, further comprising: and enabling the artificial intelligence body to record historical data of the influence of the adjustment of each circuit element on the circuit model design index in each circuit optimization design processing process and real-time data of the influence of the adjustment of each circuit element on the circuit model design index in the simulation operation of the circuit optimization model.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810350393.7A CN108595815B (en) | 2018-04-18 | 2018-04-18 | Artificial intelligence body training system and passive circuit optimization design system and method |
| PCT/CN2018/085640 WO2019200626A1 (en) | 2018-04-18 | 2018-05-04 | Artificially intelligent body training system, passive circuit optimization design system, and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810350393.7A CN108595815B (en) | 2018-04-18 | 2018-04-18 | Artificial intelligence body training system and passive circuit optimization design system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108595815A CN108595815A (en) | 2018-09-28 |
| CN108595815B true CN108595815B (en) | 2021-05-28 |
Family
ID=63611194
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810350393.7A Active CN108595815B (en) | 2018-04-18 | 2018-04-18 | Artificial intelligence body training system and passive circuit optimization design system and method |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN108595815B (en) |
| WO (1) | WO2019200626A1 (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109447245A (en) * | 2018-10-29 | 2019-03-08 | 石家庄创天电子科技有限公司 | Equivalent model generation method and modeling method neural network based |
| CN111368497A (en) * | 2020-03-13 | 2020-07-03 | 浪潮商用机器有限公司 | Circuit board schematic diagram design method and device, electronic equipment and storage medium |
| WO2023284088A1 (en) | 2021-07-12 | 2023-01-19 | 苏州贝克微电子股份有限公司 | Circuit design method based on artificial intelligence, and implementation system |
| CN113553794B (en) * | 2021-07-12 | 2023-04-07 | 苏州贝克微电子股份有限公司 | Artificial intelligence implementation system and method for circuit design |
| CN115048885B (en) * | 2022-08-12 | 2022-11-15 | 阿里巴巴（中国）有限公司 | Circuit design parameter adjusting method and device, electronic equipment and storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101714176A (en) * | 2009-12-21 | 2010-05-26 | 宁波大学 | Method for optimizing integrated circuit of analog operational amplifier |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6876961B1 (en) * | 1999-08-27 | 2005-04-05 | Cisco Technology, Inc. | Electronic system modeling using actual and approximated system properties |
| CN102024067B (en) * | 2009-09-09 | 2012-08-22 | 中国科学院微电子研究所 | Method for technology transplant of analog circuit |
| CN103279824A (en) * | 2013-05-22 | 2013-09-04 | 国家电网公司 | Modeling method for relay protection setting calculation system |
| CN106022523B (en) * | 2016-05-23 | 2017-06-30 | 广东工业大学 | A kind of automatic production line Optimization Design based on integrated emulation |
| CN107679266A (en) * | 2017-08-22 | 2018-02-09 | 珠海泓芯科技有限公司 | The emulation mode and simulator of flash memory circuit |
| CN107679282A (en) * | 2017-09-05 | 2018-02-09 | 珠海博雅科技有限公司 | The emulation mode and simulator of charge pump |
-
2018
- 2018-04-18 CN CN201810350393.7A patent/CN108595815B/en active Active
- 2018-05-04 WO PCT/CN2018/085640 patent/WO2019200626A1/en active Application Filing
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101714176A (en) * | 2009-12-21 | 2010-05-26 | 宁波大学 | Method for optimizing integrated circuit of analog operational amplifier |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2019200626A1 (en) | 2019-10-24 |
| CN108595815A (en) | 2018-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108595815B (en) | Artificial intelligence body training system and passive circuit optimization design system and method | |
| CN110147456B (en) | Image classification method and device, readable storage medium and terminal equipment | |
| CN110221975B (en) | Method and device for creating interface case automation test script | |
| US11455523B2 (en) | Risk evaluation method, computer-readable recording medium, and information processing apparatus | |
| CN109344969B (en) | Neural network system, training method thereof, and computer-readable medium | |
| CN116127899B (en) | Chip design system, method, electronic device, and storage medium | |
| US20190220924A1 (en) | Method and device for determining key variable in model | |
| CN109255148B (en) | Mechanical product design method and system | |
| CN114547917A (en) | Simulation prediction method, device, equipment and storage medium | |
| KR20220117123A (en) | System and method for economic virtuous cycle simulation based on ai twin | |
| CN109800887B (en) | Generation method and device of prediction process model, storage medium and electronic equipment | |
| CN109743200B (en) | Resource feature-based cloud computing platform computing task cost prediction method and system | |
| Ortega-Zamorano et al. | FPGA implementation of neurocomputational models: comparison between standard back-propagation and C-Mantec constructive algorithm | |
| US11966851B2 (en) | Construction of a machine learning model | |
| CN112288032B (en) | Method and device for quantitative model training based on generation of confrontation network | |
| CN117010272A (en) | Decision generation method, device, computer equipment and medium based on reinforcement learning | |
| CN116166967A (en) | Data processing method, equipment and storage medium based on meta learning and residual error network | |
| CN112633516B (en) | Performance prediction and machine learning compiling optimization method and device | |
| CN114969148A (en) | System access amount prediction method, medium and equipment based on deep learning | |
| US8954307B1 (en) | Chained programming language preprocessors for circuit simulation | |
| CN108595816B (en) | Electronic product modeling system and method based on artificial intelligence | |
| CN107766238B (en) | Software testing method, device, equipment and computer readable storage medium | |
| CN114692888A (en) | System parameter processing method, device, equipment and storage medium | |
| CN114757244A (en) | Model training method, device, storage medium and equipment | |
| CN114254764B (en) | Feedback-based machine learning model searching method, system, equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |