CN109491666A - A method of the Internet of Things fidonetFido self-programming based on artificial intelligence - Google Patents
A method of the Internet of Things fidonetFido self-programming based on artificial intelligence Download PDFInfo
- Publication number
- CN109491666A CN109491666A CN201811357593.1A CN201811357593A CN109491666A CN 109491666 A CN109491666 A CN 109491666A CN 201811357593 A CN201811357593 A CN 201811357593A CN 109491666 A CN109491666 A CN 109491666A
- Authority
- CN
- China
- Prior art keywords
- interface
- artificial intelligence
- internet
- self
- semantic
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/40—Transformation of program code
- G06F8/41—Compilation
- G06F8/42—Syntactic analysis
- G06F8/425—Lexical analysis
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/30—Creation or generation of source code
- G06F8/33—Intelligent editors
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/40—Transformation of program code
- G06F8/41—Compilation
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/70—Software maintenance or management
- G06F8/75—Structural analysis for program understanding
Landscapes
- Engineering & Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer And Data Communications (AREA)
Abstract
The present invention provides a kind of method of Internet of Things fidonetFido self-programming based on artificial intelligence, and automation generation, the automatic test of device protocol code are realized using AI technology, and finally establishes communication link.The present invention utilizes artificial intelligence technology, and equipment application layer protocol interface is trained and is analyzed, and is based on standardization program framework, generates core protocol code, quickly establishes equipment connection, realizes the plug and play of equipment interconnection.
Description
Technical field
The present invention relates to a kind of internet learning method, specifically a kind of Internet of Things fidonetFido based on artificial intelligence is certainly
The method that I programs, belongs to the internet of things field in discrete processing manufacturing industry.
Background technique
In discrete processing manufacturing industry, the communications protocol of different vendor's process equipment produced is all different, for each
Kind reason, communications protocol are difficult unification.This realizes that interconnecting for intelligent plant brings obstacle for manufacturing enterprise.Traditional is mutual
Connection mode generallys use the mode of customized development, and the development cycle is long, and framework is also difficult to unification, so that deployment time is long, safeguards
It is at high cost.
Summary of the invention
The present invention is directed to technical problem set forth above, proposes a kind of Internet of Things fidonetFido self-programming based on artificial intelligence
Method, realize the automation generation of device protocol code, automatic test using AI technology, and finally establish communication link,
Realize the plug and play of equipment interconnection.
The technical solution that the present invention solves the above technical problem is: provide a kind of Internet of Things fidonetFido based on artificial intelligence from
The method that I programs, comprising the following steps:
(1) the physical connection of embedded industrial computer and equipment, embedded industrial computer abbreviation AILINK are established；
(2) user is block search protocol type, attempts identification device protocol version by connection, establishes initial connection, and
Obtain equipment associated static information；
(3) agreement matching, the matched device driver of search institute are carried out in the application program interface program library of equipment；Such as
Fruit fails to find matched device driver, and user is prompted to need to add respective drive into interface library；
(4) morphological analysis is carried out to driver, generated interface function " syntax tree "；
(5) by the convolution algorithm based on fitting of a polynomial or based on the clustering algorithm of division to the interface function " language of generation
Function name, input, output variable carry out clustering in method tree ", generate interface function " dendrogram ", and will be currently to " grammer
" leaf " in tree " carries out the binding of semantic and cluster probability, generates " semantic tree "；
(6) standard interface function code copy is automatically generated based on " semantic tree "；Groundwork is to seal " nonstandard interface "
Dress is standardized interface function, so as to the calling of " agreement generator " in embedded industrial computer；
(7) the recursive optimization of interface function code copy is carried out by " decline of probability gradient " algorithm optimization strategy；
(8), if the maximum probability protocol interface function of recursive optimization is confirmed, the semanteme of successful match is brought into step
(5) and (6), and " semantic tree " is updated；After " semantic tree " updates, the interface function code copy of maximum probability at this time is chosen, is executed
Step is (7)；
(9), according to having interface routine template, the designated position in the code copy insertion template of successful match passes through
Compile toolchain is run, compiler will compile result and store the designated position into embedded industrial computer, and showing should
Program is to the matched completeness of protocol interface；
(10) overall test automatically generates protocol adaptation performance, so far completes being autonomously generated for device protocol.
Of the invention further limits technical solution, the side of the Internet of Things fidonetFido self-programming above-mentioned based on artificial intelligence
Method, step (2) user is module marks equipment brand, for accelerating recognition speed.
The method of Internet of Things fidonetFido self-programming above-mentioned based on artificial intelligence, the recursive optimization of the step (7) exist
The interface function that maximum probability is chosen in semantic tree carries out communication test, if the characteristic value and default semantic matches that 1. obtain,
The code copy is identified that confirmation algorithm is based on training sample, mostlys come from the working condition under constraint condition and feedback spy
Value indicative；2. if the characteristic value and preset semantic mismatch, the failure of determination step (6) semantics recognition that obtain call artificial be situated between
Enter, artificial treatment is carried out to the case, and training sample is added.
The method of Internet of Things fidonetFido self-programming above-mentioned based on artificial intelligence, the training sample are related in equipment
Whether the feature extraction of parameter, the semanteme for confirming data acquired are correct.
The method of Internet of Things fidonetFido self-programming above-mentioned based on artificial intelligence, the embedded industrial computer are soft
The hardware carrier of part operation, and support RJ45 and RS232 connecting interface.
Further, the method for the Internet of Things fidonetFido self-programming above-mentioned based on artificial intelligence, in the step (2)
" leaf " of " syntax tree " carries out calculating similarity calculation using convolution algorithm and the frequency calculates.
The beneficial effects of the present invention are: the present invention utilizes artificial intelligence technology, equipment application layer protocol interface is instructed
White silk and analysis generate core protocol code based on standardization program framework, quickly establish equipment connection, and realization equipment interconnection is
Plug-and-play.
Detailed description of the invention
Fig. 1 is overall data flow graph of the present invention.
Fig. 2 is single machine topological diagram of the invention.
Fig. 3 is that the present embodiment generates syntax tree exemplary diagram.
Fig. 4 is the present embodiment Semantic Clustering analysis examples figure.
Fig. 5 is the present embodiment recursive optimization flow diagram.
Fig. 6 is the autonomous switching flow schematic diagram of the semantic-based learning model of the present embodiment.
Fig. 7 is the neural network structure schematic diagram of the present embodiment.
Specific embodiment
Embodiment 1
The method of the present embodiment provides a kind of Internet of Things fidonetFido self-programming based on artificial intelligence, comprising the following steps:
(1) the physical connection of embedded industrial computer AILINK and equipment is established；
(2) user is that EM equipment module marks brand, is marked for accelerating recognition speed；User is block search protocol type,
Identification device protocol version is attempted by connection, establishes initial connection, and obtain equipment associated static information；
(3) agreement matching, the matched device driver of search institute are carried out in the application program interface program library of equipment；Such as
Fruit fails to find matched device driver, and user is prompted to need to add respective drive into interface library；
(4) morphological analysis is carried out to driver, generated interface function " syntax tree "；
(5) by the convolution algorithm based on fitting of a polynomial or based on the clustering algorithm of division to the interface function " language of generation
Function name, input, output variable carry out clustering in method tree ", generate interface function " dendrogram ", and will be currently to " grammer
" leaf " in tree " carries out the binding of semantic and cluster probability, generates " semantic tree "；
(6) standard interface function code copy is automatically generated based on " semantic tree "；Groundwork is to seal " nonstandard interface "
Dress is standardized interface function, so as to the calling of " agreement generator " in embedded industrial computer；
(7) the recursive optimization of interface function code copy is carried out by " decline of probability gradient " algorithm optimization strategy；Recurrence is excellent
The interface function for changing the selection maximum probability in semantic tree carries out communication test, if the characteristic value 1. obtained and default semanteme
Match, then the code copy is identified；2. determination step (6) is semantic to be known if the characteristic value and preset semantic mismatch that obtain
Do not fail, call manpower intervention, artificial treatment is carried out to the case, and training sample is added；
(8), if the maximum probability protocol interface function of recursive optimization is confirmed, the semanteme of successful match is brought into step
(5) and (6), and " semantic tree " is updated；After " semantic tree " updates, the interface function code copy of maximum probability at this time is chosen, is executed
Step is (7)；
(9), according to having interface routine template, the designated position in the code copy insertion template of successful match passes through
Compile toolchain is run, compiler will compile result and store the designated position into embedded industrial computer, and showing should
Program is to the matched completeness of protocol interface；
(10) overall test automatically generates protocol adaptation performance, so far completes being autonomously generated for device protocol.
The present embodiment is when implementing, as shown in Figure 1, third party's api interface library provides in figure for device manufacturer, it is different
Device manufacturer's this document is different, and used agreement is different.Adapter die plate is logical based on embedded industrial computer AILINK
One section of software code for interrogating the standard card cage of software after program is produced and compiled, is as directed to the device driver of the equipment,
After being loaded into frame, data exchange can be carried out with equipment.Training sample above-mentioned is the feature to relevant parameter in equipment
It extracts, whether the semanteme for confirming data acquired is correct.As shown in Fig. 2, the agreement in figure is according to its class of distinct device system
Type is different, range OPC UA/DA, Modbus TCP/RTU, TCP/IP and ProfitBus for being covered etc..In Fig. 1
Middle user completes device systems type mark first, and software systems can search for automatically corresponding agreement under the initial state, and
Determine its type.AILINK, that is, software operation hardware carrier, supports two kinds of physical connection interfaces of RJ45 and RS232 in figure.
Syntax tree produced by the present embodiment generates abstract syntax tree by morphology and syntax analyzer, as shown in figure 3, this technology is
Common technology in technique of compiling needs to set different regular expressions according to different analysis targets, the target of this paper be by
Function and non-parametric segmentation come out.
" leaf " in syntax tree in the present embodiment carries out calculating similarity calculation and frequency meter using " convolution " algorithm
It calculates；
Wherein f (x) is the fitting function after the digitlization to grammatical phrases.
Ascii (phase)={ xi, i=1...n (2)
xiFor the ascii value of i-th of character in phrase.
It willVector does the fitting of 3 order polynomials, sees formula (3):
Wherein wiFor multinomial coefficient, M is order, and after fitting, we obtain f (x).
G (τ) is convolution kernel, for labeled semantic terms template, the same formula of function process (2) and formula (3).By with
Upper process, we can be obtained the semantic classification of each phrase in syntax tree, and record the frequency of hit.
The main completion code of recursive optimization process of the present embodiment it is self-confirming, as shown in figure 5, wherein Max (s) refers to
The highest function of similarity in semantic tree is tested in selection automatically, and the core of this process is confirmation algorithm, and
Training sample database relevant to confirmation.Confirmation algorithm basic principle is: the function after calling rule obtains return value, will return
It returns value and is sent into three-layer neural network, carry out confidence level confirmation.As shown in fig. 6, being that semantic-based learning model independently switches stream
Journey selects corresponding neural network weight configuration file and activation primitive by semantic anticipation in training pattern library.Load
Afterwards, return value is sent into BP network and is confirmed.If it fails, the model of semantic similarity suboptimum is then selected to carry out load instruction
Practice.Return value is the one group of vector changed over time, the input vector as ANN.In three layers of BP network of use of the present embodiment
Model is carried out as shown in fig. 7, being divided into the different models such as revolving speed, warning, state, coordinate, cycle period by the mechanism in Fig. 6
Automatic setting.The feature vector of return is sent into ANN by output layer, if the output category obtained meets with semantic classification, at this
Function is confirmed.
The technical solution of the embodiment is realized independently to be generated from device protocol guidance, protocol semantics analysis, code, to certainly
Dynamicization tests overall process.Class validation is carried out to data by artificial neural network technology, enhances the reliability of entire mechanism.
In addition to the implementation, the present invention can also have other embodiments.It is all to use equivalent substitution or equivalent transformation shape
At technical solution, fall within the scope of protection required by the present invention.
Claims (6)
1. a kind of method of the Internet of Things fidonetFido self-programming based on artificial intelligence, it is characterised in that the following steps are included:
(1) the physical connection of embedded industrial computer and equipment is established；
(2) user's mark device type attempts identification device protocol type and version by connection, establishes initial connection, and obtain
Equipment associated static information；
(3) agreement matching, the matched device driver of search institute are carried out in the application program interface program library of equipment；If not
Matched device driver can be found, user is prompted to need to add respective drive into interface library；
(4) morphological analysis is carried out to driver, generated interface function " syntax tree "；
(5) by the convolution algorithm based on fitting of a polynomial to function name, input, output in the interface function " syntax tree " of generation
Variable carries out clustering, generates interface function " dendrogram ", and will currently to " leaf " in " syntax tree " carry out it is semantic and
Cluster probability binding, generates " semantic tree "；
(6) standard interface function code copy is automatically generated based on " semantic tree "；Groundwork is to be encapsulated as " nonstandard interface "
Standardized interface function, so as to the calling of " agreement generator " in embedded industrial computer；
(7) the recursive optimization of interface function code copy is carried out by " decline of probability gradient " algorithm optimization strategy；
If (8) the maximum probability protocol interface function of recursive optimization is confirmed, by the semanteme of successful match bring into step (5) and
(6), and " semantic tree " is updated；After " semantic tree " updates, the interface function code copy of maximum probability at this time is chosen, executes step
⑺；
(9), according to having interface routine template, by the designated position in the code copy insertion template of successful match, operation is passed through
Compile toolchain, compiler will compile result and store the designated position into embedded industrial computer, and show the program
To the matched completeness of protocol interface；
(10) overall test automatically generates protocol adaptation performance, so far completes being autonomously generated for device protocol.
2. the method for the Internet of Things fidonetFido self-programming based on artificial intelligence as described in claim 1, it is characterised in that: described
Step (2) user is module marks equipment brand.
3. the method for the Internet of Things fidonetFido self-programming based on artificial intelligence as described in claim 1, it is characterised in that: described
The interface function that the recursive optimization of step (7) chooses maximum probability in semantic tree carries out communication test, if the feature 1. obtained
Value and default semantic matches, then the code copy is identified；2. if the characteristic value and preset semantic mismatch that obtain, determine
The failure of step (6) semantics recognition, calls manpower intervention, carries out artificial treatment to the case, and training sample is added.
4. the method for the Internet of Things fidonetFido self-programming based on artificial intelligence as claimed in claim 3, it is characterised in that: described
Training sample is the feature extraction of relevant parameter in equipment, and whether the semanteme for confirming data acquired is correct.
5. the method for the Internet of Things fidonetFido self-programming based on artificial intelligence as described in claim 1, it is characterised in that: described
Embedded industrial computer is the hardware carrier of software operation, and supports RJ45 and RS232 connecting interface.
6. the method for the Internet of Things fidonetFido self-programming based on artificial intelligence as described in claim 1, it is characterised in that: described
" leaf " of " syntax tree " carries out calculating similarity calculation using convolution algorithm in step (2) and the frequency calculates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811357593.1A CN109491666B (en) | 2018-11-15 | 2018-11-15 | Internet of things protocol self-programming method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811357593.1A CN109491666B (en) | 2018-11-15 | 2018-11-15 | Internet of things protocol self-programming method based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109491666A true CN109491666A (en) | 2019-03-19 |
| CN109491666B CN109491666B (en) | 2021-04-27 |
Family
ID=65694903
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811357593.1A Active CN109491666B (en) | 2018-11-15 | 2018-11-15 | Internet of things protocol self-programming method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109491666B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111835724A (en) * | 2020-06-11 | 2020-10-27 | 广州天源信息科技股份有限公司 | A protocol conflict matching method, system, storage medium and computer device |
| CN117688319A (en) * | 2023-11-10 | 2024-03-12 | 山东恒云信息科技有限公司 | A method to use AI to analyze database structure |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20020161895A1 (en) * | 2001-04-30 | 2002-10-31 | Microsoft Corporation | Systems and methods for unified remote control access |
| US6721736B1 (en) * | 2000-11-15 | 2004-04-13 | Hewlett-Packard Development Company, L.P. | Methods, computer system, and computer program product for configuring a meta search engine |
| US7024660B2 (en) * | 1998-02-17 | 2006-04-04 | National Instruments Corporation | Debugging a program intended to execute on a reconfigurable device using a test feed-through configuration |
| CN101496374A (en) * | 2006-07-24 | 2009-07-29 | 爱克发医疗保健公司 | Method for setting up data link between medical appliances and computer system |
| CN106569947A (en) * | 2016-11-01 | 2017-04-19 | 北京大学（天津滨海）新代信息技术研究院 | Method and tool for extracting API (Application Program Interface) use example from unit test code |
-
2018
- 2018-11-15 CN CN201811357593.1A patent/CN109491666B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7024660B2 (en) * | 1998-02-17 | 2006-04-04 | National Instruments Corporation | Debugging a program intended to execute on a reconfigurable device using a test feed-through configuration |
| US6721736B1 (en) * | 2000-11-15 | 2004-04-13 | Hewlett-Packard Development Company, L.P. | Methods, computer system, and computer program product for configuring a meta search engine |
| US20020161895A1 (en) * | 2001-04-30 | 2002-10-31 | Microsoft Corporation | Systems and methods for unified remote control access |
| CN101496374A (en) * | 2006-07-24 | 2009-07-29 | 爱克发医疗保健公司 | Method for setting up data link between medical appliances and computer system |
| CN106569947A (en) * | 2016-11-01 | 2017-04-19 | 北京大学（天津滨海）新代信息技术研究院 | Method and tool for extracting API (Application Program Interface) use example from unit test code |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111835724A (en) * | 2020-06-11 | 2020-10-27 | 广州天源信息科技股份有限公司 | A protocol conflict matching method, system, storage medium and computer device |
| CN111835724B (en) * | 2020-06-11 | 2022-08-26 | 广州天源信息科技股份有限公司 | Protocol conflict matching method, system, storage medium and computer equipment |
| CN117688319A (en) * | 2023-11-10 | 2024-03-12 | 山东恒云信息科技有限公司 | A method to use AI to analyze database structure |
| CN117688319B (en) * | 2023-11-10 | 2024-05-07 | 山东恒云信息科技有限公司 | Method for analyzing database structure by using AI |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109491666B (en) | 2021-04-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US12434380B2 (en) | Robot systems, methods, control modules, and computer program products that leverage large language models | |
| US20250181840A1 (en) | Conversation Graph Navigation With Language Model | |
| CN112416337B (en) | Software architecture development system for aerospace embedded system | |
| CN112416369B (en) | Intelligent deployment method oriented to heterogeneous mixed environment | |
| CN109799990A (en) | Source code annotates automatic generation method and system | |
| CN107665216A (en) | A kind of data bank access method and middleware | |
| CN110019753B (en) | Method and device for outputting rhetorical questions for user questions | |
| CN114547611A (en) | Intelligent contract Pompe fraudster detection method and system based on multi-modal characteristics | |
| CN112019403A (en) | Cross-platform automatic mining method and system for message protocol state machine of Internet of things | |
| CN118555196A (en) | A method and system for automatically issuing configurations based on business intent | |
| CN113568900A (en) | Big data cleaning method based on artificial intelligence and cloud server | |
| CN110928535B (en) | Derived variable deployment method, device, equipment and readable storage medium | |
| CN109491666A (en) | A method of the Internet of Things fidonetFido self-programming based on artificial intelligence | |
| CN113971032B (en) | Full-process automatic deployment method and system of machine learning model for code generation | |
| CN119621074A (en) | A code conversion method, device, computer equipment and storage medium | |
| CN119988185A (en) | A method and analysis system for analyzing software code similarity | |
| CN112329440A (en) | A relationship extraction method and device based on two-stage screening and classification | |
| CN119090048A (en) | Logistics status prediction method based on spatiotemporal characteristics | |
| Bedeoui et al. | Assembly plans generation of complex machines based on the stability concept | |
| CN112995339B (en) | Automatic adaptive sensor data analysis method based on dynamic byte code technology | |
| CN116204877A (en) | Code vulnerability detection method based on graph comparison learning | |
| Sanchez et al. | Bigraphical modelling of architectural patterns | |
| CN116149611A (en) | A Software Development Method Based on Visual Component Tree | |
| WO2023138078A1 (en) | Method and apparatus for parsing programming language, and non-volatile storage medium | |
| CN117668850B (en) | Intelligent contract vulnerability detection and positioning method based on multitask learning |
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 |