An artificial intelligence-based water meter protocol matching method and system

Abstract

The application relates to an artificial intelligence-based water meter protocol matching method and system, and relates to the field of machine learning, which comprises the following steps: collecting a water meter data stream and accumulating a receiving frequency in response to a water meter communication signal; separating the water meter data stream based on the receiving frequency to determine a water meter data packet; determining a key field based on the water meter data stream and the water meter data packet; determining a key field offset based on the key field; searching for a protocol corresponding to the key field offset in a preset known protocol feature library, and defining the protocol as a matching protocol; and controlling the water meter to parse according to the matching protocol. The application has the effects of being capable of automatically identifying and matching different types of water meter protocols and adapting to various complex use scenarios.

Description

Technical Field

[0001] This invention relates to the field of machine learning, and in particular to a water meter protocol matching method and system based on artificial intelligence. Background Technology

[0002] With the development of IoT technology, smart water meters are being used more and more widely in the water industry.

[0003] Currently, data parsing of smart water meters relies on manual operation. Specifically, when installing smart water meters, staff manually debug the protocol and parse the smart water meter data format so that the terminal can read the data from the smart water meter.

[0004] Regarding the aforementioned technologies, the communication protocols between water meters and platforms are complex and diverse, with different protocol formats for different platforms. When a platform is replaced, a new water meter protocol suitable for the new platform needs to be manually rewritten. Parsing these protocols by rewriting the code is a very labor-intensive process due to the large number of protocols, which is not only time-consuming and laborious but also prone to errors. Summary of the Invention

[0005] To address the issue of code being unable to adapt to the parsing of water meter protocols across multiple platforms, this invention provides a water meter protocol matching method and system based on artificial intelligence.

[0006] This invention provides a water meter protocol matching method and system based on artificial intelligence, employing the following technical solution:

[0007] Firstly, an artificial intelligence-based water meter protocol matching method includes:

[0008] Step 1: In response to the water meter communication signal, collect the water meter data stream and accumulate the number of receptions;

[0009] Step 2: Divide the water meter data stream based on the number of receptions to determine the water meter data packet;

[0010] Step 3: Determine key fields based on the water meter data stream and the water meter data packet;

[0011] Step 4: Determine the offset of the key field based on the key field;

[0012] Step 5: Locate the protocol corresponding to the key field offset in the preset known protocol feature library and define it as the matching protocol;

[0013] Step 6: Control the water meter to parse according to the matching protocol.

[0014] By adopting the above technical solution, it is possible to automatically match the protocols in the known protocol feature library to parse the water meter protocols of different platforms without having to write code manually, which greatly improves work efficiency.

[0015] Optionally, the verification method for the matching protocol includes:

[0016] Step 60: Define the found matching protocol as the expected protocol before verifying it;

[0017] Step 61: Parse the water meter data packet based on the predicted protocol to obtain the predicted water meter information;

[0018] Step 62: Determine the water meter device number based on the predicted water meter information;

[0019] Step 63: Collect sensor data corresponding to the water meter device number based on the water meter device number;

[0020] Step 64: When the sensor data matches the expected water meter information, output the expected protocol as the target protocol;

[0021] Step 65: When the sensor data does not match the expected water meter information, output a preset matching failure signal and define the water meter data packet as an unknown data packet.

[0022] By adopting the above technical solution, the accuracy of protocol matching is further improved. When the expected protocol does not match the sensor data, it can be confirmed that the expected protocol may not be the target protocol for parsing the current water meter data stream. This avoids the generation of false water meter information due to the incorrect protocol being matched, and defines the current water meter data packet as an unknown data packet to prevent data offset of the water meter data packet from causing the correct protocol to be unable to parse the correct water meter data, thus reducing the error rate of protocol matching.

[0023] Optionally, the method for outputting the matching failure signal includes:

[0024] Step 650: Collect the unknown data packets into a preset data packet queue and accumulate the number of unknown data packets;

[0025] Step 651: When the number of unknown data packets reaches a preset reliable number, execute the preset KMP algorithm to determine the offset of the unknown data packets;

[0026] Step 652: Determine the message length based on the unknown data packet offset;

[0027] Step 653: Determine redundant data based on the message length and the number of unknown data packets;

[0028] Step 654: When redundant data exists, remove the redundant data based on the KMP algorithm to determine the corrected water meter data report;

[0029] Step 655: Based on the expected protocol, parse the corrected water meter data packet to obtain the corrected expected water meter information and match the corresponding corrected sensor data;

[0030] Step 656: When the corrected sensor data and the corrected expected water meter information do not match, output the matching failure signal and output a preset error protocol signal at the same time;

[0031] Step 657: When the corrected sensor data matches the corrected predicted water meter information, output the predicted protocol as the target protocol and do not output the matching failure signal.

[0032] By adopting the above technical solution, the true length of the current water meter data packet can be confirmed by comparing the common prefix and suffix information of multiple unknown data packets through the KMP protocol. Based on the true length of the water meter data packet, the redundancy information of a single water meter data packet is determined. The redundancy information of the water meter data packet is deleted to restore the correct water meter data packet, and then it is parsed again through the expected protocol. If the match is successful, the target protocol is determined. If the match is unsuccessful, it means that the current protocol cannot parse the water meter data packet, and the protocol matching work continues. This solves the problem that data redundancy causes the protocol to fail to parse correctly.

[0033] Optionally, it also includes a method for handling error protocols, which includes:

[0034] Step 658: When an erroneous protocol exists, arbitrarily select a current protocol and the offset of the protocol code field corresponding to the current protocol based on the known protocol feature library;

[0035] Step 659: Determine the protocol code character based on the offset of the protocol code field and the corrected water meter data packet;

[0036] Step 660: Based on the protocol code character, find the protocol corresponding to the protocol code character in the known protocol feature library and use it as the verification protocol;

[0037] Step 661: When the verification protocol is not equal to the current protocol, reselect the current protocol and the corresponding protocol code field offset based on the known protocol feature library and accumulate the number of matches;

[0038] Step 662: When the number of matching times is greater than the preset number of known protocols, output a preset unknown protocol signal;

[0039] Step 663: When the verification protocol is equal to the current protocol, define the current protocol as the protocol to be determined;

[0040] Step 664: Based on the protocol to be determined, parse the modified water meter data packet to determine the current water meter information and match the corresponding current sensor data;

[0041] Step 665: When the current water meter information does not match the current sensor data, reselect the current protocol and the offset of the protocol code field corresponding to the current protocol, and accumulate the number of matches;

[0042] Step 666: When the current water meter information is matched with the current sensor data, the protocol to be determined is output as the target protocol.

[0043] By adopting the above technical solution, during the matching process, the system sequentially extracts a protocol from the known protocol feature library to parse the current water meter data packet. Based on the format of this protocol, it determines the protocol field in the water meter data packet, and then obtains the corresponding protocol from the numbers in the protocol field. If the protocol mapped by the protocol field is not the same as the protocol currently being parsed in the water meter data packet, it indicates that the protocol being parsed is incorrect. The system then tries other protocols in the known protocol feature library until a successful match is found. If no matching protocol is found after multiple attempts, the system outputs an unknown protocol signal, indicating that the current water meter data packet may use an unknown protocol that cannot be parsed by matching protocols in the known protocol feature library. This method of quickly determining the protocol through the protocol code reduces the workload of the matching process.

[0044] Optionally, it also includes a method for updating the known protocol signature database in response to the unknown protocol signal, the method comprising:

[0045] Step 667: Determine the variable fields based on the corrected water meter data packet and the water meter data stream;

[0046] Step 668: Based on the convolutional layers in the preset CNN model, parse the variable fields to determine the time field and water consumption field;

[0047] Step 669: Parse the time field and the water consumption field to obtain time information and water consumption information;

[0048] Step 670: Locate the sensor data based on the time information and the water consumption information;

[0049] Step 671: When the sensor data matches the time information and water consumption information, determine the device number;

[0050] Step 672: When the sensor data and the time information and water consumption information do not match, a preset unidentifiable protocol signal is issued and the protocol matching operation is stopped;

[0051] Step 673: When matching the sensor data with time information and water consumption information, determine fixed fields based on the corrected water meter data report and the water meter data stream;

[0052] Step 674: Analyze the fixed field based on the device number to obtain the device number field and surrounding fields;

[0053] Step 675: Parse the surrounding fields based on the preset water meter standard protocol format to obtain XML text;

[0054] Step 676: Send the XML text to the preset AI port;

[0055] Step 677: Upon receiving a data correctness signal, determine the generation of a self-learning protocol and self-learning protocol features based on the fixed field and the variable field, and store them in the known protocol feature library.

[0056] By adopting the above technical solution, upon receiving an unknown protocol signal, the system uses the convolutional layer in a pre-set CNN model to parse the variable fields and obtain specific time and water consumption information. Then, the system searches for data matching this information in the sensor data, further determining the device number and using the AI ​​port to decipher the water meter data packet step by step. If the sensor data does not match the time or water consumption information, the system issues an unrecognizable protocol signal and stops protocol matching, achieving a self-learning effect that automatically updates the known protocol feature library.

[0057] Optionally, it also includes a method for handling data error signals received, the method comprising:

[0058] Step 678: Upon receiving the data error signal, obtain the correct field and the error field;

[0059] Step 679: Concatenate the characters based on the error field to obtain the unknown character sequence;

[0060] Step 680: Obtain the missing key field based on the key field and the correct field;

[0061] Step 681: Based on the missing key field, parse the erroneous field to obtain the XML text and send it to the AI ​​port until the correct data signal is received.

[0062] By employing the above technical solution, upon receiving a data error signal, the signal is first parsed to distinguish between correct and erroneous fields. By concatenating the erroneous fields, an unknown character sequence is formed, which helps identify abnormal parts of the data. Subsequently, by combining information from key fields and correct fields, it is determined which key fields are missing in the erroneous fields. Based on these missing key fields, further parsing attempts are made on the erroneous fields, with the goal of determining the format of the water meter data report. The XML text obtained from this parsing process is then sent to the AI ​​port, utilizing artificial intelligence's parsing capabilities to correct errors in the data until the system receives a signal indicating correct data, thereby ensuring the accuracy and integrity of the data.

[0063] Optionally, the method for parsing the erroneous field based on the missing key field to obtain the XML text includes:

[0064] Step 6810: Based on the correct field, find the corresponding missing key field floating range in the preset standard water meter data floating range table and use it as the missing key field threshold;

[0065] Step 6811: Determine the target field based on the unknown character sequence and the missing key field threshold;

[0066] Step 6812: Based on the missing key field, parse the target field and generate the XML text to the AI ​​port until the key field no longer exists;

[0067] Step 6813: Output a preset analog data signal when the target field does not exist.

[0068] By employing the above technical solution, when parsing erroneous fields, the system first refers to a preset standard water meter data fluctuation range table to determine the possible value range of the missing key field. This range is called the missing key field threshold. This threshold helps the system narrow down the search scope and improves the efficiency of finding the correct field. Next, the system combines the unknown character sequence and the missing key field threshold to attempt to identify and locate possible target fields. Once a target field is identified, the system performs deep parsing on these target fields based on the missing key field. If the system fails to find a valid target field in this step, it will output an unrecognized protocol signal and stop the current protocol matching operation. This design ensures that the system can take efficient measures to correct errors when faced with data errors and stop working promptly when correction is impossible, thus accelerating the system's parsing speed.

[0069] Optionally, it also includes a method for simulating the corrected water meter data message in response to the simulated signal, the method comprising:

[0070] Step 6814: Generate a virtual water meter protocol and virtual water meter protocol features based on the missing key field floating range and the correct field, and store them in the known protocol feature library;

[0071] Step 6815: Output the modified water meter datagram based on the virtual water meter protocol as the target protocol and parse the datagram to obtain the verification XML text;

[0072] Step 6816: Send the verification XML text to the AI ​​port;

[0073] Step 6817: When the data error signal is received, output the unrecognizable protocol signal.

[0074] By adopting the above technical solution, this invention proposes an innovative solution when the target field is missing, i.e., the system cannot parse a valid water meter data packet format from an erroneous field: simulating a water meter data packet. Specifically, the system generates a virtual water meter protocol based on the floating range of the missing key field and the known correct fields, and stores it in a known protocol feature library. Subsequently, the system outputs this virtual water meter protocol as the target protocol and attempts to parse the water meter data packet based on this protocol to obtain a verification XML text. This XML text is then sent to the AI ​​port for further verification. If the system receives data error information during the verification process at the AI ​​port, it will output an unrecognized protocol signal, indicating that the current water meter data packet may use a protocol that cannot be recognized by the current system. This design not only enhances the system's fault tolerance but also enables the system to adopt more flexible and intelligent processing strategies when facing unknown or erroneous data.

[0075] Optionally, it also includes a processing method in response to the unidentified protocol signal, the method comprising:

[0076] Step 6818: Search the preset list of client terminals to identify unknown client terminals;

[0077] Step 6819: Send a preset water meter information loss alarm to the unknown user terminal and obtain the device number corresponding to the unknown user terminal to determine the unknown device number;

[0078] Step 6820: Perform manual cracking protocol operation based on the unknown device number.

[0079] By adopting the above technical solution, when the system receives an unidentified protocol signal, it first searches a preset list of user terminals. Known user terminals are identified through the parsable water meter data packets. Unknown user terminals are then excluded to identify unknown user terminals. The unknown device number is then determined from the unknown user terminal, and staff are assigned to record and parse the protocol of the water meter data packet offline. This step ensures that even when the water meter data packet cannot be automatically identified, the system can still guarantee the accuracy and integrity of the data through manual intervention, thus improving the system's reliability and usability.

[0080] Secondly, this invention provides a water meter protocol matching system based on artificial intelligence, employing the following technical solution:

[0081] An artificial intelligence-based water meter protocol matching system includes:

[0082] The acquisition module is used to acquire sensor data and water meter data streams;

[0083] A memory for storing a program of a control method for a water meter protocol matching method based on artificial intelligence as described in any one of claims 1 to 9;

[0084] The processor loads and executes programs from memory.

[0085] By employing the above technical solution, the processor can execute the program stored in the memory, which implements the aforementioned AI-based water meter protocol matching method. The acquisition module is responsible for acquiring sensor data and water meter data streams in real time, which are the foundation for protocol matching. The memory stores various algorithms, logic, and steps required to implement the method, including processing unknown datagrams, determining protocols, handling erroneous protocols, updating the known protocol feature library, and responding to unrecognized protocol signals. The processor loads and executes these programs, processing and analyzing the acquired data through intelligent algorithms to achieve efficient matching of water meter protocols. The entire system has a clear structure and comprehensive functions, significantly improving the accuracy and efficiency of water meter data processing.

[0086] In summary, the present invention has at least one of the following beneficial technical effects:

[0087] 1. Use the offset between key fields as features to quickly retrieve protocols from the known protocol feature library to parse water meter data packets;

[0088] 2. The length of a single water meter data packet is determined by obtaining a common prefix and suffix through the KMP algorithm, thus avoiding unnecessary matching protocol operations caused by incorrect data affecting the correct protocol.

[0089] 3. By analyzing through the AI ​​port, the unknown data packet format is gradually restored, and the corresponding protocol for parsing the unknown data packet is obtained through self-learning, thereby improving the system's adaptability. Attached Figure Description

[0090] Figure 1 This is a flowchart of an artificial intelligence-based water meter protocol matching method in an embodiment of this application;

[0091] Figure 2 This is a flowchart of the matching protocol verification method in the embodiments of this application;

[0092] Figure 3 This is a flowchart of the method for outputting a matching failure signal in an embodiment of this application;

[0093] Figure 4 This is a flowchart illustrating the method for parsing erroneous fields based on missing key fields to obtain XML text in an embodiment of this application. Detailed Implementation

[0094] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0095] This invention discloses a water meter protocol matching method and system based on artificial intelligence. (Refer to...) Figure 1 An artificial intelligence-based water meter protocol matching method includes:

[0096] Step 1: In response to the water meter communication signal, collect the water meter data stream and accumulate the number of receptions.

[0097] Water meter communication signals refer to the signals emitted by the water meter for transmitting water meter data. Water meter data streams refer to the raw data sequence containing the water meter readings received from the water meter. The number of receptions refers to the number of times the system receives the water meter data stream from the water meter. Upon receiving a water meter communication signal, the system responds to the signal, begins collecting the water meter data stream, and accumulates the number of receptions.

[0098] Step 2: Segment the water meter data stream based on the number of receptions to determine the water meter data packet.

[0099] A water meter datagram is a message that encapsulates a single water meter data transmission. The number of times the water meter transmits data is received determines the number of times the water meter transmits data. Based on the number of receptions, the water meter data stream is segmented to obtain the water meter datagram. For example, if the number of receptions is 2, the middle position of the water meter data stream is determined based on its length. This middle position is then split into two separate water meter data streams, which are then used as water meter datagrams. Since the water meter data here is transmitted via the TCP / IP protocol, and the data processed by the TCP / IP protocol at the application layer is the water meter data, the format of the water meter datagram here is a message format.

[0100] Step 3: Determine key fields based on water meter data streams and water meter data packets.

[0101] Key fields refer to specific fields in water meter data packets that are of significant importance. These fields typically contain important data that identifies water meter information, device status, or data integrity. By analyzing the water meter data stream and data packets, the system can extract key fields, such as device number, timestamp, and water consumption. These fields provide the foundation for subsequent protocol matching and data parsing. The determination of key fields relies on a preset Rabin-Karp algorithm to identify key features in the string sequence. Specifically, the system collects water meter data streams into a preset hash table (a hash table is a common data structure; specifically, it's a technique that maps data to storage locations using a specific function. Its core is to use a hash function to convert keywords into storage addresses, thereby achieving fast lookup and storage), and then applies the Rabin-Karp algorithm to the hash table to determine the key features in the string sequence.

[0102] Step 4: Determine the offset of the key field based on the key field.

[0103] The key field offset refers to the offset of a key field in a water meter data packet relative to the starting point. It is determined by calculating the distance between the key field and the first character of the water meter data packet; this distance is the difference in the number of characters between the two fields. The key field offset allows for quick location of these fields within the data packet.

[0104] Step 5: Locate the protocol corresponding to the key field offset in the preset known protocol feature library and define it as the matching protocol.

[0105] The known protocol feature library is a database that stores various known water meter protocol features. A matching protocol refers to the protocol corresponding to a key field offset. By searching this library for protocols that match the key field offset, the protocol type used by the current water meter data packet can be quickly determined. Specifically, the system constructs a feature vector based on the key field offset and matches it with protocols in the protocol feature library. For example, field A corresponds to offset 16, and field B corresponds to offset 28.

[0106] Step 6: Control the water meter to parse according to the matching protocol.

[0107] When a matching protocol is determined, it means the system has successfully identified the protocol type used by the water meter data packet and can further parse the data according to that protocol. If the matching protocol cannot be determined based on the key field offset, the system will directly use the first protocol in the known protocol feature library as the matching protocol for matching.

[0108] Reference Figure 2 The verification methods for the matching protocol include:

[0109] Step 60: Define the found matching protocol as the expected protocol before verifying it.

[0110] A projected agreement refers to an unverified matching agreement. The primary purpose of this definition is to provide a basis for further verification and validation.

[0111] Step 61: Obtain the expected water meter information by parsing the water meter data report based on the expected protocol.

[0112] The expected water meter information refers to the water meter-related data parsed from the protocol. Specifically, it includes key information such as device number, timestamp, and water consumption.

[0113] Step 62: Determine the water meter device number based on the expected water meter information.

[0114] The water meter device number is a unique identifier for each water meter device. The system determines the device number field by parsing the expected water meter information.

[0115] Step 63: Collect sensor data corresponding to the water meter device number based on the water meter device number.

[0116] Sensor data refers to the associated data installed on water meters for real-time monitoring of smart water meters. This data is typically collected by sensors installed near the water meter and transmitted to the system for processing via a communication module. The system searches for the corresponding sensor data in a database or real-time data stream based on the parsed water meter device number to ensure the accuracy and consistency of the information.

[0117] Step 64: When the sensor data matches the expected water meter information, output the expected protocol as the target protocol.

[0118] The target protocol refers to a protocol that has been verified and confirmed to be correct. When the sensor data matches the predicted water meter information, it indicates that the predicted protocol can accurately parse the water meter data packet. Therefore, the system defines the predicted protocol as the target protocol and outputs it for subsequent data processing and analysis of the water meter data packet.

[0119] Step 65: When the sensor data does not match the expected water meter information, output a preset matching failure signal and define the water meter data packet as an unknown data packet.

[0120] A "Match Failure" signal is issued by the system when it detects a discrepancy between sensor data and the expected water meter information during the verification process. This signal indicates that the current expected protocol cannot correctly parse the water meter data packet. An "Unknown Data Packet" refers to a data packet whose protocol type or format cannot be identified by the expected protocol. When sensor data and expected water meter information do not match, it means that the system cannot parse the water meter data packet using the existing known protocol feature library, and therefore marks it as an "Unknown Data Packet."

[0121] Reference Figure 3 Methods for outputting a match failure signal include:

[0122] Step 650: Collect unknown data packets to the preset data packet queue and accumulate the number of unknown data packets.

[0123] A datagram queue is a data structure that stores a certain number of unknown datagrams. According to the characteristics of the queue's data structure, unknown datagrams entered first will be processed first in subsequent operations. The number of unknown datagrams refers to the total number of unknown datagrams stored in the datagram queue.

[0124] Step 651: When the number of unknown data packets reaches a preset reliable number, execute the preset KMP algorithm to determine the offset of the unknown data packets.

[0125] The reliable quantity refers to a threshold set by the system for determining the offset of unknown datagrams. The location datagram offset refers to the positional information used to identify potentially critical fields in an unknown datagram. The KMP algorithm, short for Knuth-Morris-Pratt algorithm, is a string matching algorithm that effectively skips some characters in the event of a mismatch, thereby improving matching efficiency. By executing the KMP algorithm, the system can efficiently analyze the structure of unknown datagrams and determine the offsets of their internal fields. When the number of unknown datagrams reaches the reliable quantity, it indicates that the system has sufficient data samples for analysis, thus improving its ability to parse unknown datagrams. Therefore, at this point, the system activates the KMP algorithm to determine the offset by calculating the common prefixes and suffixes of the unknown datagrams.

[0126] Step 652: Determine the message length based on the unknown datagram offset.

[0127] The message length refers to the total length of characters in an unknown data packet, from the start field to the end field. By analyzing the offset of the unknown data packet, the system can determine the fixed beginning and fixed end of each unknown data packet. Here, the CRC field, which is usually set at the end of the water meter data packet, can be parsed from the high fixed field near the end and the high fixed field at the beginning of the unknown data packet. The high fixed field is a field that generally does not change in value, usually the device number field. The CRC field is a check code used to verify the water meter data transmission or storage and may contain errors. The CRC field is different for each water meter data packet, so it is necessary to analyze the high fixed field at the end and the high fixed field at the beginning of the unknown data packet to confirm the length of the unknown data packet.

[0128] Step 653: Determine redundant data based on message length and the number of unknown datagrams.

[0129] Redundant data refers to extra characters or fields in an unknown datagram that exceed the message length. The system can identify redundant data by comparing the actual length of each unknown datagram with the calculated message length. This redundancy may be due to transmission errors, data corruption, or protocol incompatibility. The presence of redundant data indicates that the expected protocol is not necessarily incorrect; rather, the water meter datagram may be affected by redundant information, causing field offsets and preventing the correct data from being parsed.

[0130] Step 654: When redundant data exists, remove redundant data based on the KMP algorithm to determine the correct water meter data report.

[0131] A corrected water meter data packet refers to an unknown data packet after redundancy removal. Since redundant data is typically located between two unknown data packets, the system will further remove redundant data by analyzing the packet length and the number of unknown data packets using the KMP algorithm. The presence of redundant data indicates that the offsets of various fields in the unknown water meter data packet are affected, thus requiring redundancy removal to obtain the corrected water meter data packet.

[0132] Step 655: Based on the expected protocol, parse the corrected water meter data report to obtain the corrected expected water meter information and match the corresponding corrected sensor data.

[0133] Corrected predicted water meter information refers to the water meter-related information obtained by the system through the predicted protocol after removing redundant data. Corrected sensor data refers to sensor data that matches the corrected predicted water meter information. When the system receives the corrected predicted water meter information, it indicates that the parsing errors caused by redundant data in the corrected water meter data packet have been eliminated. It then needs to verify the accuracy of the corrected predicted water meter information again to determine if the predicted protocol is the target protocol; therefore, it will match it with the corresponding corrected sensor data.

[0134] Step 656: When the corrected sensor data and the corrected expected water meter information do not match, output a matching failure signal and a preset error protocol signal.

[0135] An error protocol signal is issued by the system when it confirms a mismatch between the corrected sensor data and the corrected predicted water meter information. This signal indicates that even after redundancy data removal and correction, the prediction protocol still cannot correctly parse the corrected water meter data packet. In this case, the system will simultaneously output a mismatch failure signal and an error protocol signal to indicate a potential problem with the current prediction protocol, requiring further analysis or adjustment.

[0136] Step 657: When the corrected sensor data matches the corrected expected water meter information, the expected protocol is output as the target protocol and no matching failure signal is output.

[0137] When the corrected sensor data matches the corrected predicted water meter information, it indicates that the predicted protocol can accurately parse the corrected water meter data packet. Therefore, the system defines this predicted protocol as the target protocol and outputs it. Simultaneously, the system will not output a matching failure signal, indicating that the current protocol has been verified and can correctly parse the information in the corrected water meter data packet.

[0138] This also includes a method for handling error protocols, which includes:

[0139] Step 658: When an erroneous protocol exists, arbitrarily select a current protocol and its corresponding protocol code field offset based on the known protocol feature library.

[0140] The current protocol refers to the protocol selected according to the storage order of protocols in the known protocol feature library. The protocol code field offset refers to the offset value of a specific field in the corrected water meter data packet corresponding to the current protocol relative to the starting position. When an erroneous protocol is found, it indicates that the expected protocol cannot parse the corrected water meter data packet. This may be due to an error in the previous operation process, such as an error in determining the message length or a critical field. Therefore, the protocol needs to be re-determined.

[0141] Step 659: Determine the protocol code character based on the protocol code field offset and the corrected water meter data packet.

[0142] The protocol code character refers to the character corresponding to the offset position of the protocol code field in the corrected water meter data packet. The protocol code character in the corrected water meter data packet corresponds to the protocol code in the known protocol feature library, specifically 00 corresponds to the first protocol code, and 01 corresponds to the second protocol code.

[0143] Step 660: Based on the protocol code characters, find the protocol corresponding to the protocol code characters in the known protocol feature library and use it as the verification protocol.

[0144] The verification protocol refers to the protocol retrieved from the known protocol feature database using protocol code characters. The main purpose of the verification protocol is to further verify and confirm whether the current corrected water meter data packet can be correctly parsed. If the current protocol is not the protocol corresponding to the corrected water meter data packet, it will cause errors in the protocol code field offset positioning, resulting in incorrect protocol code character information. Therefore, further verification is necessary. The system will match the protocol code characters with records in the known protocol feature database, find the corresponding protocol, and define it as the verification protocol for subsequent parsing and verification operations.

[0145] Step 661: If the verification protocol is not equal to the current protocol, reselect the current protocol and the offset of the protocol code field corresponding to the current protocol based on the known protocol feature library and accumulate the number of matches.

[0146] The number of matches refers to the number of loops the system performs when attempting to match the verification protocol with the current protocol. If the current protocol is the target protocol for the corrected water meter data packet, then the protocol code corresponding to the currently parsed protocol code characters will definitely match itself. When the verification protocol and the current protocol are inconsistent, it means that the current protocol cannot be the target protocol, and therefore a new protocol needs to be selected. The system will select the next protocol according to the order in the known protocol feature library, update the current protocol, and then parse the offset of the protocol code field corresponding to the corrected water meter data packet based on the updated current protocol. At the same time, the number of matches is accumulated to record the number of attempts.

[0147] Step 662: When the number of matches exceeds the preset number of known protocols, output the preset unknown protocol signal.

[0148] The Unknown Protocol signal is a signal used by the system to confirm that the corrected water meter data packet cannot be parsed using protocols in the known protocol signature library. This signal indicates that after multiple attempts and verifications, the system still cannot find a protocol in the known protocol signature library that can correctly parse the corrected water meter data packet. At this point, the system will stop further matching attempts and output the Unknown Protocol signal to indicate that the corrected water meter data packet may use a completely new or unrecorded protocol type.

[0149] Step 663: When the verification protocol is equal to the current protocol, define the current protocol as the protocol to be determined.

[0150] A protocol to be determined refers to a protocol temporarily marked in the system as potentially correct for parsing correction water meter data packets. The definition of a protocol to be determined implies that it has passed initial verification but still requires further validation to ensure its accuracy. This is because there is a possibility that protocol A is not the target protocol corresponding to the correction water meter data packet, but parsing the data packet yields an incorrect protocol code field. Parsing this incorrect protocol code field then points to protocol A. In such cases, further data analysis is necessary to avoid this situation.

[0151] Step 664: Based on the protocol to be determined, parse and correct the water meter data packet to determine the current water meter information and match the corresponding current sensor data.

[0152] Current water meter information refers to the water meter-related data obtained by the system after parsing and correcting the water meter data packet using the protocol to be determined. Current sensor data refers to the sensor data that matches the smart water meter device number corresponding to the current water meter information. The system will compare the parsed current water meter information with the corresponding current sensor data to verify whether the protocol to be determined can accurately parse and correct the water meter data packet.

[0153] Step 665: When the current water meter information does not match the current sensor data, reselect the current protocol and the offset of the protocol code field corresponding to the current protocol, and accumulate the number of matching times.

[0154] If the current water meter information does not match the current sensor data, it means that the current protocol cannot correctly parse and correct the water meter data packet. The system will select the next protocol in the known protocol feature library as the current protocol, parse and correct the water meter data packet through the offset of the protocol code field of the current protocol, and execute steps 659 to 665.

[0155] Step 666: When matching the current water meter information with the current sensor data, output the protocol to be determined as the target protocol.

[0156] If the current water meter information matches the current sensor data, it means that the protocol to be determined can accurately parse and correct the water meter data packet. Therefore, the system defines the protocol to be determined as the target protocol and outputs it.

[0157] This also includes a method for updating a known protocol feature library in response to unknown protocol signals, the method comprising:

[0158] Step 667: Determine the variable fields based on the corrected water meter data report and water meter data stream.

[0159] Variable fields refer to the fields in the corrected water meter data report that change. Variable fields are determined by comparing multiple samples from the corrected water meter data report with the water meter data stream, analyzing the fields that change at different times or under different device numbers. These fields typically include dynamic information such as water consumption and timestamps.

[0160] Step 668: Based on the convolutional layers in the preset CNN model, parse the variable fields to determine the time field and water consumption field.

[0161] CNN stands for Convolutional Neural Network. The time field refers to the field recording time information in water meter data reports, typically represented in a specific format such as year, month, day, or a timestamp accurate to the second. The water consumption field refers to the field used to record user water usage data. By using a CNN model for deep analysis of variable fields, the time field and water consumption field can be effectively distinguished. Convolutional layers, through local perception and weight sharing, extract key features from the variable fields and classify them as either time fields or water consumption fields based on feature distribution patterns.

[0162] Step 669: Parse the time field and water consumption field to obtain time information and water consumption information.

[0163] Time information refers to the specific time data extracted from the time field, including detailed time records such as year, month, day, hour, minute, and second. Water consumption information refers to the specific numerical values ​​related to the user's water consumption parsed from the water consumption field, specifically in cubic meters.

[0164] Step 670: Locate sensor data based on time information and water consumption information.

[0165] Once the time and water consumption information are obtained, the system can use this information to reverse-engineer and simulate the data of the corresponding smart water meter, and then deduce the corresponding protocol format of the corrected water meter data report based on the simulated smart water meter data.

[0166] Step 671: When the sensor data, time information, and water consumption information are matched, determine the device number.

[0167] When sensor data matches time and water consumption information, it indicates that the system can determine the smart water meter device number. Because the corrected water meter data packet cannot be parsed using protocols in the known protocol signature library, its format is unknown. Therefore, the system cannot accurately identify the device number field in the corrected water meter data packet; hence, matching it with sensor data, time information, and water consumption information is necessary.

[0168] Step 672: When the sensor data and time information and water consumption information do not match, a preset unrecognizable protocol signal is issued and the protocol matching work is stopped.

[0169] An unrecognized protocol signal is a signal emitted when the system cannot recognize the format of the corrected water meter data packet. This signal indicates that the current corrected water meter data packet cannot be parsed using protocols in the known protocol feature library, and the variable fields cannot be reconstructed using a CNN model; the current parsing process has reached its limit.

[0170] Step 673: When matching sensor data with time information and water consumption information, determine fixed fields based on the corrected water meter data report and water meter data stream.

[0171] Fixed fields refer to the fields corresponding to strings that remain unchanged in the corrected water meter data packet. Fixed fields typically include information such as device number, protocol header, and checksum. These fields are stable and unique throughout the data packet, used to identify the device or ensure data integrity. When sensor data matches time and water consumption information, it indicates that the system has correctly obtained the time and water consumption fields through the CNN model. At this point, the fixed fields can be identified for further analysis. The determination of fixed fields relies on in-depth comparative analysis of the corrected water meter data packet and the water meter data stream. By comparing multiple samples from the corrected water meter data packet and the water meter data stream, it can be determined which fields remain consistent across different time points or different device numbers, thus identifying them as fixed fields.

[0172] Step 674: Analyze the fixed fields based on the device number to obtain the device number field and surrounding fields.

[0173] The device number field is used to identify the device in the corrected water meter data packet. Surrounding fields are other fixed fields adjacent to or closely related to the device number field, such as protocol headers and checksums. The existence of the device number field enables the system to distinguish different smart water meter devices, thereby achieving accurate data parsing and matching. By analyzing the device number field and its surrounding fields, more information about the structure of the corrected water meter data packet can be extracted.

[0174] Step 675: Parse the surrounding fields based on the preset water meter standard protocol format to obtain XML text.

[0175] The water meter standard protocol format refers to a universal, industry-standard water meter data transmission format. XML text refers to an Extensible Markup Language (XML) file generated by parsing surrounding fields, used to describe the structure and content of fixed fields in the modified water meter data packet. Since the modified water meter data packet format is unknown, but the system has already determined the fixed and surrounding fields, the fields near the fixed fields can be analyzed using the water meter standard protocol format. Then, the fixed and surrounding fields can be converted into XML text according to the preset water meter standard protocol format conversion. For example, the original message is as follows:

[0176] "68670101000100242206722502000099060a25".

[0177] According to the standard water meter protocol format, it needs to be parsed into the following XML content:

[0178] {"companyNo":"654024","protocolType":"03","did":"22-24-722502000099-03-5","operators":5}.

[0179] Step 676: Send the XML text to the preset AI port.

[0180] The AI ​​port is a specific interface in the system used to receive and process XML text. Its main function is to perform further intelligent analysis and processing of the received XML text. The system sends the XML text to the AI ​​port and waits for a response from the AI.

[0181] Step 677: Upon receiving a data correctness signal, determine the self-learning protocol and self-learning protocol features based on fixed fields and variable fields, and store them in the known protocol feature library.

[0182] The data correctness signal refers to the correct signal returned by the AI ​​port, specifically determined by binary representation. For example, 0 represents a correct signal, indicating that the sent XML text has been correctly parsed and conforms to the expected format. The self-learning protocol refers to the protocol by which the system, after correctly parsing surrounding fields using the water meter standard protocol format, transforms unknown surrounding fields into known fields and gradually improves the process. The characteristics of the self-learning protocol refer to its specific format, such as the offset of the device number field. Receiving a data correctness signal indicates that the system has successfully parsed and understood the structure of some fields in the corrected water meter data packet. Therefore, a self-learning protocol can be generated based on this information. The core of the self-learning protocol lies in gradually transforming unknown fields into known fields, and through continuous optimization and improvement, ultimately forming a complete set of parsing rules.

[0183] This also includes a method for handling data error signals, which includes:

[0184] Step 678: Obtain the correct and incorrect fields upon receiving the data error signal.

[0185] Data error signals refer to error signals returned by the AI ​​port, specifically determined by binary representation. For example, a 1 indicates an error signal, signifying that the sent XML text failed to be parsed correctly. Correct fields are those fields that the AI ​​port determines are accurately identified and understood during the parsing process. Error fields, on the other hand, are those fields that the AI ​​port determines failed parsing validation or have formatting issues. Receiving data error signals means the system needs to further correct and optimize the parsing results to ensure all fields are correctly understood.

[0186] Step 679: Concatenate based on the error field to obtain the unknown character sequence.

[0187] An unknown character sequence refers to a continuous set of characters composed of error fields. Unknown character sequences are generated by concatenating the error fields. The existence of unknown character sequences indicates that there are still unrecognized or ununderstood parts in the corrected water meter data report, requiring further analysis and processing.

[0188] Step 680: Obtain the missing key field based on the key field and the correct field.

[0189] Missing critical fields refer to key fields that could not be identified or parsed in the corrected water meter data report. Key fields are pre-determined; when parsing correct and incorrect fields, the key fields can be used to exclude which key fields were not confirmed as correct. These unconfirmed key fields are the missing critical fields.

[0190] Step 681: Parse the erroneous field based on the missing key field to obtain XML text and send it to the AI ​​port until a signal indicating that the data is correct is received.

[0191] When a key field is missing, it indicates that the self-learning protocol is not yet complete. In this case, it is necessary to narrow down the parsing scope by identifying the missing and erroneous fields. The missing key fields are then gradually parsed from the erroneous fields to generate accurate XML text. This process is continuously verified through the AI ​​portal until the data is completely correct, ensuring the protocol's self-learning function is gradually improved.

[0192] Reference Figure 4 Methods for parsing erroneous fields based on missing key fields to obtain XML text include:

[0193] Step 6810: Based on the correct field, find the corresponding missing key field floating range in the preset standard water meter data floating range table and use it as the missing key field threshold.

[0194] The standard water meter data fluctuation range refers to the allowed range of numerical variations for certain key fields in common water meter data reports. This range is obtained by consulting a pre-defined standard water meter data fluctuation table. The standard water meter data fluctuation table stores the range of numerical variations for key fields in smart water meters. When a missing key field is found, it indicates that the format of the water meter data report needs further parsing. The system will then search the pre-defined standard water meter data fluctuation range table for the corresponding fluctuation range of the missing key field based on the known correct field values, and use this range as the threshold for the missing key field.

[0195] Step 6811: Determine the target field based on the unknown character sequence and the threshold for missing key fields.

[0196] The target field refers to a field with potential parsing value in an unknown character sequence, filtered by a missing key field threshold. The determination of the target field is based on in-depth analysis of the unknown character sequence, combined with matching and verification results using the missing key field threshold. The system extracts fields that meet the criteria from the unknown character sequence based on the allowed fluctuation range of the missing key field and marks them as target fields for subsequent parsing.

[0197] Step 6812: Parse the target field based on the missing key field and generate XML text to the AI ​​port until there is no missing key field.

[0198] This section describes an exhaustive approach that repeatedly analyzes the data to identify the missing key fields and determine them as the correct fields.

[0199] Step 6813: Output a preset analog data signal when the target field does not exist.

[0200] A simulated data signal is a control signal issued by the system when it cannot find the target field. The output of a simulated data signal indicates that the current parsing process has encountered an obstacle, possibly due to the lack of a field in the unknown character sequence that meets the missing critical field threshold, or due to significant uncertainty in the data itself. Triggering this signal will guide the system into a preset data compensation mechanism to attempt to complete the parsing of the corrected water meter data report through other methods.

[0201] This also includes a method for simulating and correcting water meter data packets in response to the simulated signal, the method comprising:

[0202] Step 6814: Generate the virtual water meter protocol and virtual water meter protocol features based on the missing key field floating range and the correct field, and store them in the known protocol feature library.

[0203] A virtual water meter protocol is a temporary protocol used to simulate the parsing process of corrected water meter data packets. The generation of the virtual water meter protocol is based on a comprehensive analysis of the floating range of missing key fields and the correct fields. By setting reasonable floating ranges and field rules, the system can construct a virtual protocol that conforms to the expected format. The characteristics of the virtual water meter protocol refer to the specific parsing rules and field offsets contained within it. These characteristics enable the system to better understand and process similar data packets in subsequent parsing processes.

[0204] Step 6815: Output and parse the corrected water meter data packets based on the virtual water meter protocol to obtain the verification XML text.

[0205] The XML verification text refers to the XML format file generated after parsing the corrected water meter data packet using the virtual water meter protocol. It is used to verify the accuracy and applicability of the virtual protocol. Because the format of the corrected water meter data packet cannot be fully parsed, and therefore the corrected water meter data packet cannot be obtained, data simulation is used here to replace the target protocol.

[0206] Step 6816: Send the verification XML text to the AI ​​port.

[0207] Step 6817: Output an unrecognized protocol signal when a data error signal is received.

[0208] When a data error signal is received, indicating that the virtual water meter protocol has failed to successfully simulate the corrected water meter data packet, the system will output an unrecognized protocol signal. This signal indicates that the current virtual protocol cannot meet the parsing requirements, possibly because the corrected water meter data packet contains fields outside the floating range or data that does not conform to the expected format.

[0209] This also includes a processing method in response to unidentified protocol signals, the method comprising:

[0210] Step 6818: Locate the preset client list to identify unknown clients.

[0211] The client list refers to the collection of client information stored in the system, used to identify and manage different client devices and their related information. An unknown client refers to a device or data source for which no known client information was found in the current parsing process.

[0212] Step 6819: Send a preset water meter information loss alarm to the unknown user terminal and obtain the device number corresponding to the unknown user terminal to determine the unknown device number.

[0213] A water meter information loss alarm is a text-based alert triggered when the system detects that the corrected water meter data packet cannot be parsed using a protocol in the known protocol signature library or a virtual water meter protocol.

[0214] Step 6820: Perform manual protocol cracking operation based on the unknown device number.

[0215] Manual protocol decryption refers to the in-depth analysis and parsing of corrected water meter data packets corresponding to unknown device numbers using manual methods. This process relies on the professional knowledge and experience of technical personnel, combining known protocol feature libraries, data analysis tools, and relevant contextual information to gradually deduce the specific format and field meanings of the corrected water meter data packets. Manual protocol decryption typically includes the following steps: First, preliminary decoding of the corrected water meter data packets; second, inferring the function and structure of unknown fields by comparing with known protocol formats; finally, verifying the inferred results with actual data to ensure the accuracy of the parsing. If manual decryption is successful, the newly parsed protocol format is updated in the known protocol feature library for subsequent automatic parsing; if manual decryption fails, offline maintenance is performed based on the address information corresponding to the unknown device number.

[0216] Based on the same inventive concept, embodiments of the present invention provide a water meter protocol matching system based on artificial intelligence.

[0217] One of them is an artificial intelligence-based water meter protocol matching system, which includes: an acquisition module for acquiring sensor data and water meter data stream;

[0218] A memory for storing a program of a control method for a water meter protocol matching method based on artificial intelligence as described in any one of claims 1 to 9;

[0219] The processor loads and executes programs from memory.

[0220] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0221] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A water meter protocol matching method based on artificial intelligence, characterized in that, include: Step 1: In response to the water meter communication signal, collect the water meter data stream and accumulate the number of receptions; Step 2: Divide the water meter data stream based on the number of receptions to determine the water meter data packet; Step 3: Determine key fields based on the water meter data stream and the water meter data packet; Step 4: Determine the offset of the key field based on the key field; Step 5: Locate the protocol corresponding to the key field offset in the preset known protocol feature library and define it as the matching protocol; Step 6: Control the water meter to parse according to the matching protocol; The verification method of the matching protocol includes: Step 60: Define the found matching protocol as the expected protocol before verifying it; Step 61: Parse the water meter data packet based on the predicted protocol to obtain the predicted water meter information; Step 62: Determine the water meter device number based on the predicted water meter information; Step 63: Collect sensor data corresponding to the water meter device number based on the water meter device number; Step 64: When the sensor data matches the expected water meter information, output the expected protocol as the target protocol; Step 65: When the sensor data does not match the expected water meter information, output a preset matching failure signal and define the water meter data packet as an unknown data packet; The method for outputting the matching failure signal includes: Step 650: Collect the unknown data packets into a preset data packet queue and accumulate the number of unknown data packets; Step 651: When the number of unknown data packets reaches a preset reliable number, execute the preset KMP algorithm to determine the offset of the unknown data packets; Step 652: Determine the message length based on the unknown data packet offset; Step 653: Determine redundant data based on the message length and the number of unknown data packets; Step 654: When redundant data exists, remove the redundant data based on the KMP algorithm to determine the corrected water meter data report; Step 655: Based on the expected protocol, parse the corrected water meter data packet to obtain the corrected expected water meter information and match the corresponding corrected sensor data; Step 656: When the corrected sensor data and the corrected expected water meter information do not match, output the matching failure signal and output a preset error protocol signal at the same time; Step 657: When the corrected sensor data matches the corrected predicted water meter information, output the predicted protocol as the target protocol and do not output the matching failure signal; This also includes a method for handling error protocols, which includes: Step 658: When an erroneous protocol exists, arbitrarily select a current protocol and the offset of the protocol code field corresponding to the current protocol based on the known protocol feature library; Step 659: Determine the protocol code character based on the offset of the protocol code field and the corrected water meter data packet; Step 660: Based on the protocol code character, find the protocol corresponding to the protocol code character in the known protocol feature library and use it as the verification protocol; Step 661: When the verification protocol is not equal to the current protocol, reselect the current protocol and the corresponding protocol code field offset based on the known protocol feature library and accumulate the number of matches; Step 662: When the number of matching times is greater than the preset number of known protocols, output a preset unknown protocol signal; Step 663: When the verification protocol is equal to the current protocol, define the current protocol as the protocol to be determined; Step 664: Based on the protocol to be determined, parse the modified water meter data packet to determine the current water meter information and match the corresponding current sensor data; Step 665: When the current water meter information does not match the current sensor data, reselect the current protocol and the offset of the protocol code field corresponding to the current protocol, and accumulate the number of matches; Step 666: When the current water meter information is matched with the current sensor data, the protocol to be determined is output as the target protocol; It also includes a method for updating the known protocol feature library in response to the unknown protocol signal, the method comprising: Step 667: Determine the variable fields based on the corrected water meter data packet and the water meter data stream; Step 668: Based on the convolutional layers in the preset CNN model, parse the variable fields to determine the time field and water consumption field; Step 669: Parse the time field and the water consumption field to obtain time information and water consumption information; Step 670: Locate the sensor data based on the time information and the water consumption information; Step 671: When the sensor data matches the time information and water consumption information, determine the device number; Step 672: When the sensor data and the time information and water consumption information do not match, a preset unidentifiable protocol signal is issued and the protocol matching operation is stopped; Step 673: When matching the sensor data with time information and water consumption information, determine fixed fields based on the corrected water meter data report and the water meter data stream; Step 674: Analyze the fixed field based on the device number to obtain the device number field and surrounding fields; Step 675: Parse the surrounding fields based on the preset water meter standard protocol format to obtain XML text; Step 676: Send the XML text to the preset AI port; Step 677: Upon receiving a data correctness signal, determine and generate a self-learning protocol and self-learning protocol features based on the fixed field and the variable field, and store them in the known protocol feature library; This also includes a method for handling data error signals, which includes: Step 678: Upon receiving the data error signal, obtain the correct field and the error field; Step 679: Concatenate the characters based on the error field to obtain the unknown character sequence; Step 680: Obtain the missing key field based on the key field and the correct field; Step 681: Based on the missing key field, parse the erroneous field to obtain the XML text and send it to the AI ​​port until the data correct signal is received; The method for parsing the erroneous field based on the missing key field to obtain the XML text includes: Step 6810: Based on the correct field, find the corresponding missing key field floating range in the preset standard water meter data floating range table and use it as the missing key field threshold; Step 6811: Determine the target field based on the unknown character sequence and the missing key field threshold; Step 6812: Based on the missing key field, parse the target field and generate the XML text to the AI ​​port until the key field no longer exists; Step 6813: Output a preset analog data signal when the target field does not exist; The method further includes a method for simulating the corrected water meter data message in response to the simulated data signal, the method comprising: Step 6814: Generate a virtual water meter protocol and virtual water meter protocol features based on the missing key field floating range and the correct field, and store them in the known protocol feature library; Step 6815: Output the modified water meter datagram based on the virtual water meter protocol as the target protocol and parse the datagram to obtain the verification XML text; Step 6816: Send the verification XML text to the AI ​​port; Step 6817: When the data error signal is received, output the unrecognizable protocol signal.

2. The water meter protocol matching method based on artificial intelligence according to claim 1, characterized in that, It also includes a processing method in response to the unidentified protocol signal, the method comprising: Step 6818: Search the preset list of client terminals to identify unknown client terminals; Step 6819: Send a preset water meter information loss alarm to the unknown user terminal and obtain the device number corresponding to the unknown user terminal to determine the unknown device number; Step 6820: Perform manual cracking protocol operation based on the unknown device number.

3. A water meter protocol matching system based on artificial intelligence, characterized in that, include: The acquisition module is used to acquire sensor data and water meter data streams; A memory for storing a program of a control method for a water meter protocol matching method based on artificial intelligence as described in any one of claims 1 to 2; The processor loads and executes programs from memory.

Images