Device end data flow consistency verification method and system

Abstract

The invention relates to the technical field of the Internet of Things, discloses a method and a system for verifying the consistency of big data flow based on a device side, and aims to solve the problems of low efficiency and accuracy of an existing data consistency verification mode. The scheme mainly comprises the following steps: upper computer software captures log data reported by a WiFi module of an intelligent device in real time; extracting electric control data in the data to obtain a plurality of pieces of structured equipment end local data; acquiring corresponding IoT platform cloud data and big data platform cloud data; using a preset local analysis configuration file to analyze and render the local data of the device end, and generating a first visualization result; analyzing and rendering the cloud data of the IoT platform and the cloud data of the big data platform by using a preset cloud analysis configuration file to generate a second visual result and a third visual result; and performing automatic consistency comparison, and outputting a consistency comparison result. According to the invention, the efficiency and accuracy of data consistency verification are improved, and the method is suitable for IoT equipment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of Internet of Things, and in particular to a device end data flow consistency verification method and system. BACKGROUND

[0002] With the rapid development of Internet of Things technology, intelligent devices are increasingly popular in the market, and the amount of data generated by them is also growing exponentially. These device end data is usually reported to the cloud Internet of Things platform via the WiFi module and ultimately stored in the big data platform for persistent storage and analysis. Thus, a data flow link from the device end to the cloud is formed, and ensuring the consistency, integrity and correctness of the data in this link is a key prerequisite for ensuring the accuracy and reliability of the upper layer data analysis results.

[0003] In the prior art, intelligent devices mainly include MCU and WiFi modules, and the intelligent devices package and report the electronic control data to the IoT platform, which then asynchronously stores the data to the big data platform. However, for this data link, i.e. from the device end local data, to the cloud data received by the IoT platform, to the final data stored in the big data platform, whether they remain consistent, whether there are storage exceptions or data packet loss, packet error, etc., there is currently a lack of effective automated verification tools.

[0004] Currently, the industry generally uses manual methods for data consistency verification. Test personnel need to manually capture the logs reported by the device end, and manually unpack and analyze the log data based on their deep understanding of the device reporting protocol. At the same time, the interfaces of the IoT platform and the big data platform need to be called respectively to obtain cloud data, and finally the three-end data is compared field by field by the human eye. The entire process is not only tedious and time-consuming, but also highly dependent on the experience and concentration of the test personnel, and is prone to human errors such as missing or misreading, with low verification efficiency and reliability. In addition, when the device end data protocol changes, the manual parsing and comparison rules also need to be updated simultaneously, and the maintenance cost of the entire verification process is high and agile. SUMMARY

[0005] The present application aims to solve the problem of low efficiency and accuracy of existing data consistency verification methods, and proposes a device end data flow consistency verification method and system.

[0006] The technical solution adopted by the present application to solve the above technical problems is: In a first aspect, the present application provides a device end data flow consistency verification method, comprising the following steps: Step 1: Establish serial communication between the host computer software and the WiFi module of the intelligent device to capture log data reported by the WiFi module in real time; Step 2, from the log data, screening out data packets containing complete big data fields; splitting and parsing the complete data packets to extract the packaged electric control data therein, obtaining multiple structured device end local data; Step 3, through the host computer software, calling the data query interface of the IoT platform, obtaining the IoT platform cloud data corresponding to the device end local data, and calling the data query interface of the big data platform, obtaining the big data platform cloud data corresponding to the device end local data; Step 4, using the pre-set local analysis configuration file, analyzing and rendering the device end local data obtained by analysis, to generate a first visualization result; Step 5, using the pre-set cloud analysis configuration file, respectively analyzing and rendering the IoT platform cloud data and the big data platform cloud data obtained, to generate a second visualization result and a third visualization result; Step 6, automatically comparing the consistency of the first visualization result with the second visualization result and the third visualization result, respectively, and field by field checking the consistency among the device end local data, the IoT platform cloud data and the big data platform cloud data; Step 7, outputting the consistency comparison result and displaying all inconsistent data fields in the interface of the host computer software in highlighted form.

[0007] Further, before step 1, there is also an initialization step: Enter the SN code of the intelligent device in the host computer software, and select the serial number for communication with the WiFi module of the intelligent device to establish a communication connection and start listening to log data.

[0008] Further, in step 2, the splitting and parsing of the data packets specifically includes: According to the data packet structure and separator defined in the device end electric control data protocol, a single big data packet is split into multiple independent sub-data records; Extracting the attribute fields and their corresponding attribute values in each sub-data record to form standardized data units for analysis and comparison.

[0009] Further, in step S4, the local analysis configuration file is a JavaScript file based on the device Uart electric control data protocol, which encapsulates the analysis rules and field mapping relationship of the device end data protocol; Using the local analysis configuration file to analyze the device end local data specifically includes: The parsing rule and field mapping relationship in the JavaScript file are executed to convert the device-side local data in binary or hexadecimal format into attribute key-value pairs with explicit semantics, thereby generating the first visualization result.

[0010] Further, in step S5, the cloud-side parsing configuration file is a JSON file based on device model and data attribute, and the JSON file defines the mapping relationship between cloud-side data and standard attribute fields. The cloud-side parsing configuration file is used to parse the IoT platform cloud-side data and the big data platform cloud-side data, specifically including: From the JSON structure of the IoT platform cloud-side data and the big data platform cloud-side data, the corresponding attribute fields and attribute values are extracted according to the mapping relationship in the JSON file, and are aligned with the attribute fields of the device-side local data, thereby generating the second visualization result and the third visualization result.

[0011] Further, in step 6, the automatic consistency comparison specifically includes: Each attribute field and its attribute value in the first visualization result are matched with the same attribute field in the second visualization result and the third visualization result, respectively. All attribute value mismatched fields, missing fields in the cloud-side data, and redundant fields in the device-side local data are recorded.

[0012] Further, the method further includes: In step S8, when the electric control data protocol of the intelligent device is changed or data attribute is newly added, the local parsing configuration file and / or the cloud-side parsing configuration file are updated to adapt the new data protocol and attribute for the host computer software.

[0013] In a second aspect, the application provides a device-side data flow consistency verification system for implementing the device-side data flow consistency verification method of the first aspect, and the system includes: An intelligent device includes an MCU and a WiFi module, and the WiFi module is configured to receive electric control data from the MCU and package the electric control data as log data to be reported to an IoT platform. An IoT platform is configured to receive and store log data reported by the WiFi module of the intelligent device, and provides a data query interface to obtain IoT platform cloud-side data. A big data platform is configured to receive and store data transferred from the IoT platform, and provides a data query interface to obtain big data platform cloud-side data. A host computer software includes: The communication unit is configured to capture log data reported by the WiFi module through serial communication; The local data analysis unit is configured to perform integrity check on the log data, filter out data packets containing complete big data fields, and perform splitting and analysis to extract electric control data to generate device-side local data; and perform analysis on the device-side local data by using a local analysis configuration file to obtain a first visualization result. The cloud data analysis unit is configured to call a data query interface of the IoT platform to obtain IoT platform cloud data, and call a data query interface of the big data platform to obtain big data platform cloud data; and perform analysis on the IoT platform cloud data and the big data platform cloud data by using a cloud analysis configuration file to obtain a second visualization result and a third visualization result. The data comparison unit is configured to perform consistency comparison between the first visualization result and the second visualization result and the third visualization result respectively, and verify the consistency between the device-side local data, the IoT platform cloud data, and the big data platform cloud data field by field, and display all inconsistent data fields in a highlighted form.

[0014] Further, the local analysis configuration file is a JavaScript file based on device Uart electric control data protocol, and the JavaScript file encapsulates the analysis rules and field mapping relationship of the device-side data protocol; The local analysis configuration file is used to analyze the device-side local data, specifically including: The analysis rules and field mapping relationship in the JavaScript file are executed to convert the binary or hexadecimal format of the device-side local data into attribute key-value pairs with explicit semantics, thereby generating the first visualization result.

[0015] Further, the cloud analysis configuration file is a JSON file based on device model and data attribute, and the JSON file defines the mapping relationship between cloud data and standard attribute fields; The cloud analysis configuration file is used to analyze the IoT platform cloud data and the big data platform cloud data respectively, specifically including: From the JSON structure of the IoT platform cloud data and the big data platform cloud data, the corresponding attribute fields and attribute values are extracted according to the mapping relationship in the JSON file, and are aligned with the attribute fields of the device-side local data, thereby generating the second visualization result and the third visualization result.

[0016] The device end data flow consistency verification method and system provided by the application have the beneficial effects that the device end data flow consistency verification method and system provided by the application automatically capture device end logs through upper computer software, automatically analyze data, automatically obtain cloud data and perform consistency comparison, thereby completely liberating test personnel from tedious and error-prone manual operation, and without manual intervention throughout, which not only greatly saves human resources and improves verification efficiency, but also avoids missed detection and misjudgment caused by human factors, and ensures high accuracy and reliability of data verification results. The application adopts a structure in which local JavaScript analysis configuration files are separated from cloud JSON analysis configuration files, when the electric control data protocol of the intelligent device is changed or new data attributes are added, only the corresponding configuration files need to be updated, without the need to modify the core code of the upper computer software, so that the entire verification tool can quickly respond to changes in business requirements, greatly reducing the cost and complexity of software maintenance and version iteration, and enhancing the adaptability and iterability of the system. The analysis results of the device end, the IoT platform and the big data platform are visualized and presented, and the inconsistent data fields are highlighted, so that problems in the data flow process can be intuitively and clearly exposed, which helps developers and testers quickly and accurately locate data abnormalities, and provides a clear direction for quickly repairing problems and optimizing data links. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 A structural schematic diagram of a device end data flow consistency verification system provided for an embodiment is shown in the figure. Figure 2 A flowchart of a device end data flow consistency verification method provided for an embodiment is shown in the figure. DETAILED DESCRIPTION

[0018] Since existing data consistency verification mainly relies on manual methods, the following defects exist: the process of manually capturing, analyzing and comparing data is tedious and time-consuming, seriously wasting human resources, low in efficiency and high in resource consumption; and extremely dependent on the subjective experience and concentration of test personnel, prone to human errors such as missing and misreading, and the verification result is unreliable, low in accuracy and reliability; in addition, when the device data protocol is changed, the manual analysis rules need to be relearned and adjusted, the maintenance cost is high, the iteration cost is high, and the agility is poor.

[0019] Based on this, the technical solution of this invention is proposed. In this invention, firstly, the host computer software listens in real time to the raw logs reported by the WiFi module of the smart device via a serial port. When the user selects big data on the software interface, the system records the unique identifier of the data packet, such as a timestamp or serial number. Using this identifier, the system requests data corresponding to the same device SN code within the same time period from both the IoT platform and the big data platform in parallel. This ensures that the three data samples from the device, IoT platform, and big data platform being compared are logically from the same source, laying the foundation for data comparison. Then, for local data on the device, which is usually unstructured data in binary or custom hexadecimal format, logical operations such as bit operations, endianness conversion, and formula calculations are performed using the parsing algorithm and business logic encapsulated in the local parsing configuration file. For cloud data, which is already in structured JSON format, field mapping and extraction are performed using the cloud parsing configuration file. By separating the parsing rules from the software core through the local parsing configuration file and the cloud parsing configuration file, only the external configuration file needs to be updated for changes in specific devices or protocols. Finally, the data received from the three terminals, parsed and aligned to the same standard attribute fields, is compared. A precise field-by-field matching is performed and visualized, accurately pinpointing problematic stages in the data flow process. This provides strong support for rapid troubleshooting and improved R&D quality. The above process automates the entire workflow from data acquisition and parsing to comparison, completely replacing manual labor. While significantly improving efficiency, it ensures the accuracy and reliability of the verification results.

[0020] The technical solutions in this embodiment will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0021] Figure 1 A schematic diagram of a device-side data flow consistency verification system is shown. Please refer to [link / reference]. Figure 1 The system includes: smart devices, an IoT platform, a big data platform, and host computer software.

[0022] In this context, "smart devices" refers to physical devices capable of collecting their own status or environmental data and packaging and reporting this data to a cloud platform via a WiFi network. These include MCUs and WiFi modules, such as smart home appliances, environmental monitoring devices, health devices, or other IoT devices. The MCU is responsible for collecting the device's electrical control data, such as raw signals like temperature, humidity, and operating status. The WiFi module communicates with the MCU via a UART serial port, receives the electrical control data sent by the MCU, packages the data according to a preset communication protocol, adds timestamps, device identifiers, and other information to form complete log data, and then reports it to the IoT platform via the internet.

[0023] The IoT platform serves as a data relay station, with main functions including receiving and storing log data reported by numerous smart device WiFi modules in file format, while providing RESTful APIs or similar data query interfaces, allowing host computer software to query and obtain IoT platform cloud data before mapping processing through parameters such as device SN codes and time ranges.

[0024] The big data platform is the final storage location for data. This platform can be AWS, Amazon or Alibaba Cloud servers, or a self-built server; this embodiment does not impose any restrictions. The main functions of the big data platform include: asynchronously pulling data from the IoT platform and persistently storing it, while providing data query interfaces (such as AWS S3's SelectAPI) to allow host computer software to obtain the mapped and processed cloud data from the big data platform.

[0025] The host computer software adopts a modular design, including: Communication Unit: Responsible for interacting with the hardware of smart devices. By selecting the correct serial port number and setting parameters such as baud rate, it establishes a connection with the WiFi module of the smart device with the specified SN code, and listens for and captures the log data printed by the device in real time.

[0026] Local Data Parsing Unit: Responsible for processing raw data from the device. This unit has a built-in local parsing configuration file. First, it performs integrity verification on the captured log data, filtering out data containing complete keywords; then, it splits data packets according to the protocol to extract the electronic control data; finally, it executes the parsing function of the local parsing configuration file to convert the binary or hexadecimal raw data into readable structured data, obtaining the device-side local data and generating the first visualization result.

[0027] Cloud Data Parsing Unit: Responsible for acquiring and processing cloud data. This unit has a built-in cloud parsing configuration file. It is used to call the query interfaces of the IoT platform and the big data platform in parallel to obtain the corresponding cloud data. Then, according to the field mapping rules in the cloud parsing configuration file, it extracts and standardizes the data, and generates the second and third visualization results respectively.

[0028] Data Comparison Unit: Receives three visualization results, performs precise field-by-field matching through the consistency comparison engine, automatically identifies inconsistent fields, and clearly displays the comparison report in a highlighted form on the graphical interface.

[0029] Based on the above system, Figure 2 A flowchart illustrating a method for verifying data flow consistency on the device side is shown. Please refer to [link / reference]. Figure 2 The method includes the following steps: Step 1: Establish serial communication between the host computer software and the WiFi module of the smart device, and capture the log data reported by the WiFi module in real time.

[0030] Before step 1, an initialization step is also included: Enter the SN code of the smart device into the host computer software, and select the serial port number for communicating with the WiFi module of the smart device to establish a communication connection and start listening to log data.

[0031] In practical applications, a stable data channel is established using the serial communication API provided by the operating system to directly read all information output in real time from the WiFi module's debugging serial port. Specifically, the user enters the device serial number (SN) in the host computer software interface, selects the correct serial port number, and clicks to open the serial port. The underlying software uses libraries such as SerialPort to configure the baud rate, data bits, stop bits, and parity bits to match the module, and starts a listening thread to continuously read and buffer the serial port data. Step 1 achieves lossless, real-time capture of raw data from the device, providing a reliable data source for subsequent analysis and avoiding data loss or errors that may occur due to manual interception.

[0032] Step 2: Filter out data packets containing complete big data fields from the log data; split and parse the complete data packets to extract the packaged electronic control data and obtain multiple structured local data from the device.

[0033] In this embodiment, the splitting and parsing of data packets specifically includes: Based on the data packet structure and delimiters defined in the device-side electrical control data protocol, a single large data packet is split into multiple independent sub-data records; the attribute fields and their corresponding attribute values ​​are extracted from each sub-data record to form standardized data units that can be parsed and compared.

[0034] In practical applications, pattern matching and data packet reconstruction are performed based on the rules of the device communication protocol. First, large data packets are identified by keywords. Then, the integrity of the data packets is verified according to the start character, length field, end character, and CRC checksum defined in the protocol. Finally, multiple sub-records within the packet are split according to the protocol structure. Specifically, in the cached log stream, regular expressions are used to find data segments containing keywords and data content, and their length and checksum are verified. For complete data packets, according to the protocol definition, such as each sub-record being fixed at N bytes, a single large data packet is split into multiple independent sub-data records. The sub-data records are traversed to extract the core electronic control data payload. Step 2 achieves automatic and accurate extraction of target test data from the mixed log stream and structures the complex binary packets, preparing for subsequent parsing and comparison, replacing the tedious process of manual packet unpacking and recording.

[0035] Step 3: Through the host computer software, call the data query interface of the IoT platform to obtain the IoT platform cloud data corresponding to the local data on the device, and call the data query interface of the big data platform to obtain the big data platform cloud data corresponding to the local data on the device.

[0036] In practical applications, the platform's open API is invoked using the HTTP / HTTPS protocol. Device identifier (SN code) and data identifier (timestamp) are used as query conditions to request cloud data corresponding to the local device data selected in step S2. Specifically, after the user selects a piece of local device data, the host computer software automatically extracts the timestamp and device SN of that data and initiates two asynchronous HTTP GET requests. One request is sent to the query interface of the IoT platform, and the other is sent to the query interface of the big data platform. Step 3 establishes a common-source association among the three data sources, ensuring that the compared device data, IoT platform data, and big data platform data are logically the same data in different stages, thus guaranteeing the effectiveness and scientific rigor of the comparison.

[0037] Step 4: Using a pre-set local parsing configuration file, parse and render the parsed local data on the device to generate the first visualization result.

[0038] In this embodiment, the local parsing configuration file is a JavaScript file based on the device UART electronic control data protocol. The JavaScript file encapsulates the parsing rules and field mapping relationships of the device-side data protocol. Parsing the local data on the device using a local parsing configuration file specifically includes: The parsing rules and field mapping relationships in the JavaScript file are executed to convert device-side local data in binary or hexadecimal format into attribute key-value pairs with clear semantics, thereby generating the first visualization result.

[0039] In practical applications, a JavaScript file is used as the local parsing configuration file. This is because the raw data reported by the device is typically in binary or custom hexadecimal format, requiring complex logical operations for parsing. JavaScript, as a scripting language, can easily encapsulate such complex parsing algorithms and business logic. The JavaScript file defines the parsing function, including the parsing rules and field mapping relationships of the device's data protocol. This function receives the raw electronic control data, parses it into key-value pairs through bit operations, type conversions, and calculations, renders it into a table or tree component, and generates the first visual result.

[0040] Step 5: Using the pre-set cloud parsing configuration file, parse and render the acquired cloud data from the IoT platform and the cloud data from the big data platform respectively to generate a second visualization result and a third visualization result.

[0041] In this embodiment, the cloud-based parsing configuration file is a JSON file based on the device model and data attributes. The JSON file defines the mapping relationship between cloud data and standard attribute fields. The cloud-based parsing configuration files are used to parse the cloud data of the IoT platform and the big data platform, respectively, specifically including: From the JSON structure of the IoT platform cloud data and the big data platform cloud data, the corresponding attribute fields and attribute values ​​are extracted according to the mapping relationship in the JSON file, and aligned with the attribute fields of the device's local data to generate the second visualization result and the third visualization result.

[0042] In practical applications, JSON files are used as cloud-based parsing configuration files. This is because the data obtained from IoT and big data platforms via interfaces is already in structured JSON format, requiring no complex calculations; the main task is field mapping and extraction. JSON files are ideal for statically defining this mapping relationship between cloud field names and standard attribute names. The cloud data parsing unit retrieves the corresponding field values ​​from the JSON data returned from the cloud based on this mapping relationship, unifies them to standard field names, and generates the second and third visualization results.

[0043] Steps 4 and 5 employ the most suitable configuration file types to address the complexity of device-side data and the structure of cloud-based data, ensuring efficient and accurate parsing. Furthermore, when the protocol changes, only the JavaScript or JSON file needs to be updated, eliminating the need to recompile and deploy the host computer software, significantly reducing maintenance costs and enabling rapid iteration.

[0044] Step 6: Perform an automated consistency comparison between the first visualization result and the second and third visualization results, verifying the consistency between the local data on the device, the cloud data on the IoT platform, and the cloud data on the big data platform, field by field.

[0045] In this embodiment, the automated consistency comparison specifically includes: Each attribute field and its attribute value in the first visualization result is matched with the same attribute field in the second and third visualization results; all fields with mismatched attribute values, fields missing in the cloud data, and redundant fields in the local data on the device are recorded.

[0046] Specifically, each attribute in the first visualization result is checked to see if it exists in the second and third visualization results, and whether the values ​​are strictly equal. Redundant fields that exist in the cloud but not on the device are also checked.

[0047] Step 7: Output the consistency comparison results and display all inconsistent data fields in a highlighted form on the interface of the host computer software.

[0048] Specifically, all inconsistent items (different values, missing items, redundant items) are recorded in a discrepancy report. The UI iterates through all data rows, dynamically adding highlight styles to cells with discrepancies, thus verifying the consistency of the data flow. This visual presentation accurately pinpoints problematic stages in the data flow process, providing strong support for quickly troubleshooting and improving R&D quality.

[0049] In this embodiment, the method further includes: Step S8: When the electronic control data protocol of the smart device changes or new data attributes are added, the host computer software is adapted to the new data protocol and attributes by updating the local parsing configuration file and / or the cloud parsing configuration file.

[0050] In practical applications, when device firmware upgrades or the addition of new features lead to changes in the data protocol, if the protocol change involves the packetization method, field length, or parsing algorithm of the original data on the device side, the JavaScript file is modified; if the change involves the data mapping relationship of the IoT platform or big data platform, or adds attributes, the JSON file is modified. Since the core engine does not need to be changed, the risk of introducing new defects due to code modification is avoided, ensuring the long-term stability of the basic data comparison function.

[0051] In summary, the device-side data flow consistency verification method and system provided in this embodiment achieves full automation from data acquisition and parsing to comparison, completely replacing manual labor. This significantly improves efficiency while ensuring the accuracy and reliability of the verification results. By separating the parsing rules from the software core through JavaScript and JSON configuration files, only the configuration file needs to be updated when the protocol changes, without modifying the code, demonstrating strong adaptability and rapid iteration capabilities. The visual interface highlights inconsistent fields, accurately locating problematic stages in the data flow process, providing strong support for rapid troubleshooting and improved R&D quality.

Claims

1. A method for verifying consistency of data flow at device end, characterized in that, The method comprises the following steps: Step 1, establish serial communication with the WiFi module of the intelligent device through the host computer software, and capture the log data reported by the WiFi module in real time; Step 2, from the log data, filter out data packets containing complete big data fields; split and analyze the complete data packets to extract the packaged electric control data, and obtain multiple structured device-side local data; Step 3, through the host computer software, call the data query interface of the IoT platform to obtain the IoT platform cloud data corresponding to the device-side local data, and call the data query interface of the big data platform to obtain the big data platform cloud data corresponding to the device-side local data; Step 4, use the pre-set local analysis configuration file to analyze and render the device-side local data obtained by analysis, to generate a first visualization result; Step 5, use the pre-set cloud analysis configuration file to analyze and render the IoT platform cloud data and the big data platform cloud data respectively, to generate a second visualization result and a third visualization result; Step 6, automatically compare the consistency of the first visualization result with the second visualization result and the third visualization result respectively, and verify the consistency between the device-side local data, the IoT platform cloud data and the big data platform cloud data field by field; Step 7, output the consistency comparison result, and display all inconsistent data fields in the form of highlighting on the interface of the host computer software.

2. The method of claim 1, wherein, Before step 1, there is also an initialization step: Enter the SN code of the intelligent device in the host computer software, and select the serial number for communication with the WiFi module of the intelligent device to establish a communication connection and start listening to log data.

3. The method of claim 1, wherein, In step 2, the splitting and analyzing of the data packet specifically includes: According to the data packet structure and separator defined in the device-side electric control data protocol, split a single big data packet into multiple independent sub-data records; Extract the attribute fields and their corresponding attribute values in each sub-data record to form a standardized data unit for analysis and comparison.

4. The method of claim 1, wherein, In step S4, the local analysis configuration file is a JavaScript file based on the device Uart electric control data protocol, which encapsulates the analysis rules and field mapping relationship of the device-side data protocol in the JavaScript file; Use the local analysis configuration file to analyze the device-side local data, specifically including: Execute the analysis rules and field mapping relationship in the JavaScript file to convert the binary or hexadecimal format device-side local data into attribute key-value pairs with clear semantics, thereby generating the first visualization result.

5. The method of claim 1, wherein, In step S5, the cloud analysis configuration file is a JSON file based on the device model and data attributes, which defines the mapping relationship between cloud data and standard attribute fields in the JSON file; Use the cloud analysis configuration file to analyze the IoT platform cloud data and the big data platform cloud data respectively, specifically including: From the JSON structure of the IoT platform cloud data and the big data platform cloud data, corresponding attribute fields and attribute values are extracted according to the mapping relationship in the JSON file, and are aligned with the attribute fields of the device end local data, so as to generate the second visualization result and the third visualization result.

6. The method of claim 1, wherein, In step 6, the automatic consistency comparison specifically includes: Each attribute field and its attribute value in the first visualization result are matched with the same attribute field in the second visualization result and the third visualization result respectively; All attribute value mismatched fields, missing fields in the cloud data and redundant fields in the device end local data are recorded.

7. The method of claim 1, wherein, The method further includes: In step S8, when the electric control data protocol of the intelligent device is changed or data attributes are newly added, the host computer software is adapted to the new data protocol and attributes by updating the local parsing configuration file and / or the cloud parsing configuration file.

8. A device-side data flow consistency verification system, comprising: To implement the device end data flow consistency verification method of any one of claims 1 to 7, the system includes: An intelligent device including an MCU and a WiFi module, the WiFi module being configured to receive electric control data from the MCU and report the electric control data as log data to an IoT platform after being packaged; An IoT platform configured to receive and store log data reported by the WiFi module of the intelligent device and provide a data query interface for obtaining IoT platform cloud data; A big data platform configured to receive and store data transferred from the IoT platform and provide a data query interface for obtaining big data platform cloud data; A host computer software including: A communication unit configured to capture log data reported by the WiFi module through serial communication; A local data parsing unit configured to perform integrity check on the log data, filter out data packets containing complete big data fields, and perform splitting and parsing to extract electric control data to generate device end local data; and perform parsing on the device end local data by using a local parsing configuration file to obtain a first visualization result; A cloud data parsing unit configured to call a data query interface of the IoT platform to obtain IoT platform cloud data, and call a data query interface of the big data platform to obtain big data platform cloud data; and perform parsing on the IoT platform cloud data and the big data platform cloud data by using a cloud parsing configuration file to obtain a second visualization result and a third visualization result; A data comparison unit configured to perform consistency comparison between the first visualization result and the second visualization result and the third visualization result respectively, verify the consistency between device end local data, IoT platform cloud data and big data platform cloud data field by field, and display all inconsistent data fields in a highlighted form.

9. The device-side data flow consistency verification system of claim 8, wherein, The local parsing configuration file is a JavaScript file based on device Uart electric control data protocol, and the JavaScript file encapsulates the parsing rules and field mapping relationship of the device end data protocol. The local analysis configuration file is used to analyze the device-side local data, specifically including: The analysis rules and field mapping relationship in the JavaScript file are executed to convert the binary or hexadecimal format device-side local data into attribute key-value pairs with explicit semantics, thereby generating the first visualization result.

10. The device-side data flow consistency verification system of claim 8, wherein, The cloud-side analysis configuration file is a JSON file based on device model and data attribute, and the mapping relationship between cloud-side data and standard attribute fields is defined in the JSON file. The cloud-side analysis configuration file is used to analyze the IoT platform cloud-side data and the big data platform cloud-side data, specifically including: From the JSON structure of the IoT platform cloud-side data and the big data platform cloud-side data, the corresponding attribute fields and attribute values are extracted according to the mapping relationship in the JSON file, and are aligned with the attribute fields of the device-side local data, thereby generating the second visualization result and the third visualization result.

Images