Relay protection intelligent automatic detection system and method based on voice recognition

Abstract

The invention discloses an intelligent automatic relay protection detection system and method based on voice recognition. The system comprises a hardware equipment layer, a communication interface layer, an automatic test layer and a voice control layer, the hardware equipment layer comprises a relay protection tester; the communication interface layer comprises a relay protection tester interface program and a protocol engine program, the relay protection tester interface program supports control and acquisition of a relay protection instrument, and the protocol engine program realizes communication with a tested device; the automatic test layer supports one-button automatic test of the relay protection device and can load the intelligently generated test template; a voice dictation module is embedded in the voice control layer, and the voice control layer converts a natural language into a text output instruction defined by an automatic test system interface; the purpose of applying and popularizing the relay protection detection technology which is simple in operation and logic is achieved.

Description

Technical Field

[0001] This invention relates to the field of relay protection, specifically to an intelligent automated detection system and method for relay protection based on voice recognition. Background Technology

[0002] Relay protection devices are a crucial component of power systems. As the first line of defense in the three-tiered power system defense system, relay protection plays a vital role in ensuring the safe operation of the power system. The stable and reliable operation of relay protection devices is a necessary condition for the stable operation of the power system. Currently, the testing technologies and methods for relay protection devices are becoming increasingly mature, and automated testing of relay protection devices has been achieved. With the development of artificial intelligence technology, it has been widely applied in power line inspection. However, its application and promotion have not yet been realized in the complex and logically intricate relay protection testing technology. The modeling specifications in this method are based on the 61850 modeling specification. Summary of the Invention

[0003] The purpose of this invention is to solve the problem that existing relay protection detection technologies, which are complex in operation and logic, have not yet been applied and promoted.

[0004] To address the aforementioned issues, this invention provides a voice recognition-based intelligent automated detection system for relay protection, comprising: a hardware device layer, a communication interface layer, an automatic testing layer, and a voice control layer. Hardware equipment layer: Relay protection tester; Communication interface layer: Relay protection tester interface program and protocol engine program. The relay protection tester interface program supports the control and acquisition of relay protection instruments, supports the simulation of various protection faults, and supports various protection functions, namely overcurrent protection, distance protection, and differential protection. The protocol engine program realizes communication with the device under test and supports the reading and writing of information such as setting value, control word, pressure plate, and trip. Automatic testing layer: Supports one-click automatic testing of relay protection devices, and can load intelligently generated test templates, which are test sets of various protection functions and sampling functions of relay protection devices; Voice control layer: Embedded voice dictation module, which converts natural language into text output commands defined by the automatic testing system interface; A smart automated detection method for relay protection based on voice recognition, comprising: S1: Develop a standard equipment data model file according to the modeling specifications, and use a hash algorithm to parse the relay protection device model from the SCD file to generate an actual device model file that is consistent with the standard equipment data model file in XML format; S2: The model comparison module calls the data model lexicon and uses a fuzzy matching algorithm to obtain a device mapping file formed by mapping the standard equipment data model file and the actual device model file. The specific implementation method of the model comparison module is as follows: The fuzzy matching algorithm centralizes the data descriptions in the standard equipment data model file and the actual device model file into phrases. It then performs fuzzy matching across the entire dataset based on the matching degree of these phrases, setting a matching degree threshold of 100% to ensure the accuracy of the match. The matching degree calculation formula is as follows: In the formula, The fuzzy matching degree is represented by P, the number of successfully matched word groups is represented by Xstd, the number of data segments in the standard device data model file is represented by Xrel, and the maximum value between n and m is represented by Max{n,m}. After the standard equipment data model file is matched with the actual device model file, a device mapping file is formed; When the automatic relay protection testing system performs automated testing on devices of the same voltage level and type, it uses the device mapping file for testing.

[0005] In the preferred method, the SCD file is a configuration file within the substation, provided by the production site.

[0006] In the preferred approach, phrase formation associates standard terms and idiomatic terms into phrases to create a phrase library.

[0007] In the preferred embodiment, the device mapping file describes the data association relationship, i.e., the correspondence, between the standard equipment data model file and the actual device model file. For example, the voltage of phase A is matched with the voltage of phase A. The beneficial effects of this invention are as follows: by acquiring the operator's voice, intelligently adjusting the signal intensity, and integrating it with a one-click automated testing program, a fully intelligent relay protection testing technology that eliminates manual operation is formed, reducing human intervention, further realizing the automation level of relay protection testing, and further improving the efficiency of substation operation and maintenance. Attached Figure Description

[0008] Figure 1 This is a schematic diagram of the workflow of the present invention. Detailed Implementation

[0009] Example 1: A relay protection intelligent automated detection system based on voice recognition includes: a hardware device layer, a communication interface layer, an automatic testing layer, and a voice control layer; Hardware equipment layer: Relay protection tester; Communication interface layer: Relay protection tester interface program and protocol engine program. The relay protection tester interface program supports the control and acquisition of relay protection instruments, supports the simulation of various protection faults, and supports various protection functions, namely overcurrent protection, distance protection, and differential protection. The protocol engine program realizes communication with the device under test and supports the reading and writing of information such as setting value, control word, pressure plate, and trip. Automatic testing layer: Supports one-click automatic testing of relay protection devices, and can load intelligently generated test templates, which are test sets of various protection functions and sampling functions of relay protection devices; Voice control layer: Embedded voice dictation module, which converts natural language into text output commands defined by the automatic testing system interface; The system is easy to use and can recognize the operator's voice. It can perform automatic tests based on voice prompts and voice recognition, and automatically generate standardized or customized test reports.

[0010] This invention also proposes a method for automatic matching of communication models of relay protection devices based on intelligent recognition technology. This method uses fuzzy matching + synonym matching technology to complete the automatic matching function of data models of relay protection devices of the same voltage and type, and finally realizes the generalization of automatic test templates for relay protection devices of the same voltage and type. The implementation method is as follows: A smart automated detection method for relay protection based on voice recognition, comprising: S1: Develop a standard equipment data model file according to the modeling specifications, and use a hash algorithm to parse the relay protection device model from the SCD file to generate an actual device model file that is consistent with the standard equipment data model file in XML format; S2: The model comparison module calls the data model lexicon and uses a fuzzy matching algorithm to obtain a device mapping file formed by mapping the standard equipment data model file and the actual device model file. The specific implementation method of the model comparison module is as follows: The standard equipment data model file and the actual device model file are matched one by one according to the data identifiers of the logical devices and datasets; The fuzzy matching algorithm groups the data descriptions in the standard equipment data model file and the actual device model file into phrases. Phonics formation involves associating standard and common terms with phrases to build a phrase library. Fuzzy matching of the entire dataset is then performed based on the matching degree of the phrases, with a matching degree threshold set to 100% to ensure the correctness of the match. The matching degree calculation formula is as follows: In the formula, The fuzzy matching degree is represented by P, the number of successfully matched word groups is represented by Xstd, the number of data segments in the standard device data model file is represented by Xrel, and the maximum value between n and m is represented by Max{n,m}. After the standard equipment data model file and the actual device model file are matched, a device mapping file is formed. The device mapping file describes the data relationship between the two, that is, the correspondence relationship, such as voltage phase A and phase A voltage, forming a match. When the automatic relay protection testing system performs automated testing on devices of the same voltage level and type, it uses model mapping files for testing, ensuring the universality of the test.

[0011] S3: The synonym matching algorithm primarily targets data that fails to match using fuzzy matching. This means the data doesn't share any identical characters in its textual description, but their actual meanings are consistent. To address this issue, a synonym matching function was added. Phrases with common meanings are bound and associated, allowing the model comparison module to determine if these phrases are mutually compatible. Examples include: "synchronous voltage" and "Ux," "exit" and "not deployed," etc. The synonym matching uses a self-built synonym database. For instance, "synchronous voltage" is a standard term, while "protection" might actually be a more common, more conventional term like "Ux." Therefore, when two corresponding terms are encountered, a match is performed.

[0012] The model comparison module employs a matching algorithm that primarily uses fuzzy matching, supplemented by synonym matching. During automatic testing of relay protection devices, the actual equipment data model file references the device mapping file, ultimately achieving the standardization of test templates for relay protection devices of the same voltage level and type.

[0013] The model comparison module is the core of this method, and its algorithm consists of fuzzy matching and synonym matching. The lexicon used for matching is the data model lexicon. This lexicon defines the descriptions of data in the device data models from major domestic relay protection device manufacturers through word segmentation.

[0014] The SCD file is a configuration file within the substation, provided by the production site.

[0015] Example 2: Voice testing process: The entire system is deployed on a test computer. The operator speaks: "Xiao Bo, Xiao Bo," which pops up a voice chat window. The voice feedback is: "Hello, welcome to the Relay Protection Device Voice Automatic Testing System. How can I help you?" The operator speaks: "I want to test a relay protection device." Testing keywords: test, relay protection. Voice feedback: "Please state the model number of the device you want to test, such as: PCS931, CSC103, etc." Testing keywords: device model. The operator states the device model number, and the system searches for a template containing the model number in the template name. The system then searches for the device data model containing the model number under the path Library / Voice Test Item Model / . Voice feedback: "Please state the protocol type of the device you want to test, such as: MMS, CMS." Testing keywords: MMS, CMS. The operator speaks: "MMS or CMS." Voice feedback: "Please state the IP address of the device you want to test, such as: 192.168.1.21." Testing keywords: device IP address. Operator's voice prompt: 192.168.1.121. The system calls the automatic testing system and creates a new test. After creation, the voice feedback is: Hello, do you want to start the test? Operator's voice prompt: Yes. When the system detects keywords such as "okay" or "yes" in the operator's speech, the test begins. The first step of the test is wiring prompts. Voice feedback: Hello, please complete the wiring according to the prompts. The operator completes the wiring according to the prompts. The voice system waits for the operator to say "I have completed the wiring" and continue to the next step. The system continues. During the test, the operator can say: Stop test, Continue test to control the start and stop of the overall test. After the test is completed, the voice feedback is: Test completed, would you like to close the test? Closing the test means closing the entire system. If the operator says: Close test, the voice prompt says: Okay. Operator's voice prompt: View report. This opens the Word report that was just tested. To close the intelligent voice assistant, the operator says: Close assistant. Voice feedback: Okay, thank you for using it, we look forward to seeing you again.

[0016] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of this invention is defined by the appended claims and their equivalents.

Claims

1. A relay protection intelligent automated detection system based on voice recognition, characterized in that, include: Hardware device layer, communication interface layer, automatic testing layer, voice control layer; Hardware equipment layer: Relay protection tester; Communication interface layer: Relay protection tester interface program and protocol engine program. The relay protection tester interface program supports the control and acquisition of relay protection instruments, supports the simulation of various protection faults, and supports various protection functions, namely overcurrent protection, distance protection, and differential protection. The protocol engine program realizes communication with the device under test and supports the reading and writing of information such as setting value, control word, pressure plate, and trip. Automatic testing layer: Supports one-click automatic testing of relay protection devices, and can load intelligently generated test templates, which are test sets of various protection functions and sampling functions of relay protection devices; Voice control layer: Embedded voice dictation module, which converts natural language into text output commands defined by the interface of the automated testing system.

2. A relay protection intelligent automated detection method based on voice recognition, characterized in that, include: S1: Develop a standard equipment data model file according to the modeling specifications, and use a hash algorithm to parse the relay protection device model from the SCD file to generate an actual device model file that is consistent with the standard equipment data model file in XML format; S2: The model comparison module calls the data model lexicon and uses a fuzzy matching algorithm to obtain a device mapping file formed by mapping the standard equipment data model file and the actual device model file. The specific implementation method of the model comparison module is as follows: The fuzzy matching algorithm centralizes the data descriptions in the standard equipment data model file and the actual device model file into phrases. It then performs fuzzy matching across the entire dataset based on the matching degree of these phrases, setting a matching degree threshold of 100% to ensure the accuracy of the match. The matching degree calculation formula is as follows: In the formula, The fuzzy matching degree is represented by P, the number of successfully matched word groups is represented by Xstd, the number of data segments in the standard device data model file is represented by Xrel, and the maximum value between n and m is represented by Max{n,m}. After the standard equipment data model file is matched with the actual device model file, a device mapping file is formed; When the automatic relay protection testing system performs automated testing on devices of the same voltage level and type, it uses the device mapping file for testing.

3. The intelligent automated detection method for relay protection based on voice recognition according to claim 2, characterized in that, Phonicization involves associating standard and idiomatic terms with phrases to create a phrase library.

4. The intelligent automated detection method for relay protection based on voice recognition according to claim 2, characterized in that, The device mapping file describes the data association relationship, or correspondence, between the standard equipment data model file and the actual device model file. For example, voltage phase A is matched with phase A voltage.

Images