Remote monitoring and parameter configuration system for speed reducer

Abstract

The invention relates to the technical field of data transmission, and discloses a speed reducer remote monitoring and parameter configuration system, which comprises the following modules: a calculation module used for establishing QUIC protocol connection between a speed reducer terminal and a cloud server to realize communication; the speed reducer terminal collects vibration, temperature and load data, extracts a main frequency amplitude and calculates a comprehensive index of the health state of the equipment; the scheduling module is used for determining the priority of the current data for the cloud server to perform flow scheduling; the reporting module is used for calculating the size of a data aggregation window by the speed reducer terminal in combination with the smooth round-trip time and priority of the QUIC connection; a plurality of time sequence telemetering floating point values collected in the window size are converted into fixed-length binary sequences, and after the fixed-length binary sequences are spliced to form aggregated data blocks, bit compression coding is carried out, and the aggregated data blocks are reported; and the feedback module is used for realizing asynchronous feedback of an instruction execution state after the speed reducer terminal executes the instruction. According to the invention, the real-time performance of data transmission and the reliability of connection are improved.

Description

Technical Field

[0001] This invention relates to the field of data transmission technology, and specifically to a remote monitoring and parameter configuration system for a speed reducer. Background Technology

[0002] Most existing remote monitoring systems for speed reducers are built on the traditional TCP / IP protocol stack, with MQTT or HTTP protocols often used for data exchange at the upper layer. However, in industrial environments with unstable wireless signals or network congestion, the inherent head-of-line blocking problem of the TCP protocol and the delays caused by multiple handshakes during connection establishment can significantly affect the real-time performance and reliability of data reporting. Furthermore, the large volume of time-series data generated by industrial sensors means that using text formats such as JSON for encapsulation would greatly increase the size of data packets, wasting limited network bandwidth and placing a heavy burden on cloud servers for parsing and storage.

[0003] To differentiate the urgency of data, existing solutions typically add a priority field to the application-layer payload. However, when congestion occurs at the network layer, network devices cannot recognize these application-layer markers, causing critical alarm data to be delayed or discarded along with regular periodic data, thus losing the ability to be transmitted with priority.

[0004] In remote control, the traditional request-response model requires the terminal to return an acknowledgment immediately after executing the command. This synchronous or semi-synchronous interaction method is prone to uncertainty in status due to response timeouts or packet loss in weak network environments, and it also incurs additional communication overhead.

[0005] Meanwhile, terminal-side health status assessment models are often relatively simple and cannot comprehensively reflect the actual health status of devices under varying loads and operating conditions. There is a lack of a mechanism that can integrate multi-source sensor information at the edge and intelligently associate the assessment results with communication strategies. Summary of the Invention

[0006] This invention provides a remote monitoring and parameter configuration system for speed reducers to solve the problem of poor real-time performance and reliability of data transmission in industrial environments with unstable wireless signals or network congestion, as described in the prior art.

[0007] The remote monitoring and parameter configuration system for speed reducers of the present invention includes the following modules: The calculation module is used to establish a QUIC protocol connection between the reducer terminal and the cloud server to achieve communication, including monitoring data reporting and configuration parameter distribution. The reducer terminal periodically collects vibration, temperature, and load data, performs a Fast Fourier Transform on the vibration data to extract the main frequency amplitude, and calculates the comprehensive health status index of the equipment by combining the time gradient of the temperature data with the current load torque. The scheduling module is used to compare the comprehensive health status index of the equipment with preset multi-level health thresholds to determine the priority of the current data, and encodes the priority into preset bits of the flow ID of the QUIC unidirectional flow carrying the monitoring data for traffic scheduling by the cloud server. The reporting module is used by the reducer terminal to combine QUIC data with the network data transmission mechanism. The smooth round-trip time and priority of the C connection are used to calculate the data aggregation window size. Within the data reporting cycle, multiple time-series telemetry floating-point values ​​collected within the window size are converted into fixed-length binary sequences according to their respective range and accuracy constraints, spliced ​​into aggregated data blocks, bit-compressed, encoded, and reported. The feedback module is used to ensure that when a parameter configuration command with an embedded command association ID is received through the QUIC bidirectional stream, the reducer terminal, after executing the command, carries the same command association ID in the data frame of the subsequently reported unidirectional telemetry data stream, and fills in the status code mapped by the comprehensive index of the device health status generated immediately after execution in the preset status field of the data frame, thereby realizing asynchronous feedback of the command execution status.

[0008] Preferably, the step of extracting the dominant frequency amplitude by performing a fast Fourier transform on the vibration data includes: A 1024-point Fast Fourier Transform was performed on the vibration acceleration signal with a period of 100ms to extract the maximum amplitude value A in the 0-500Hz frequency band as the main frequency amplitude value.

[0009] Preferably, the time gradient of the temperature data is calculated in the following way: The temperature signal with a period of 1 second is read and the difference between the current temperature and the temperature 10 seconds ago is calculated to obtain the temperature gradient G. The temperature gradient G is used as the time gradient of the temperature data.

[0010] Preferably, the comprehensive health status index of the computing device includes: Read the current load torque L and calculate the comprehensive health status index H of the equipment using the following formula; ; in, The preset maximum safety threshold for amplitude, The preset maximum safe threshold for the temperature gradient. This is the preset maximum safe threshold for load torque.

[0011] Preferably, the step of determining the priority of the current data by comparing the comprehensive health status index of the equipment with preset multi-level health thresholds includes: The multi-level health thresholds for the comprehensive health status index H of the equipment are set to 0.9, 0.7, and 0.5. when At that time, the device health status level is healthy, and the priority is 3; when When the device's health status level is good, the priority is 2; when When the device's health status level is "warning" and its priority is 1, then... At that time, the equipment health status level is dangerous, and the priority is 0.

[0012] Preferably, the step of encoding the priority into the preset bits of the stream ID of the QUIC unidirectional stream carrying the monitoring data includes: Write the 2-bit priority value into the 5th and 6th bits of the stream ID of the QUIC unidirectional stream.

[0013] Preferably, the size of the computational data aggregation window includes: Obtain the smooth round-trip time (SRTT) in milliseconds from the QUIC connection session; set the base window size. And the priority weighting factor F, when the priority P is 0, 1, 2, 3, F takes the values ​​of 0.2, 0.5, 1.0, 2.0 respectively; Calculate the data aggregation window size W using the following formula; ,in, It is the preset minimum window size.

[0014] Preferably, the step of converting multiple time-series telemetry floating-point values ​​collected within the window size into fixed-length binary sequences according to their respective range and accuracy constraints, and then splicing them together to form an aggregated data block, includes: Convert floating-point temperature values ​​ranging from -50.0℃ to 150.0℃ with an accuracy of 0.1℃ to integers between 0 and 2000 using the following formula: (floating-point temperature value) 50.0) 10, and encode the converted integer into an 11-bit binary temperature sequence; The floating-point value of the vibration dominant frequency amplitude, ranging from 0 to 10g with an accuracy of 0.01g, is converted to an integer between 0 and 1000 using the following formula: Amplitude 100, and encode the converted integer into a 10-bit binary vibration sequence; Convert the floating-point value of load torque in the range of 0 to 1000 Nm with an accuracy of 1 Nm to an integer between 0 and 1000, and encode the converted integer into a 10-bit binary load sequence; The 32-bit timestamp, 11-bit binary temperature sequence, 10-bit binary vibration sequence, and 10-bit binary load sequence within a data acquisition cycle are concatenated into a 63-bit data frame, and all data frames within the data aggregation window are concatenated sequentially to form an aggregated data block.

[0015] Preferably, the bit compression encoding includes: The first 63-bit data frame within the aggregated data block is left unprocessed; starting from the second data frame, the data frame is XORed with the previous data frame; the number of leading zeros Z in the XOR result is calculated and represented by 6 bits; Z and the remaining bit stream starting from the first non-zero bit are concatenated into a compressed data frame, thereby compressing the parts of the data frames that change little.

[0016] Preferably, the asynchronous feedback of the instruction execution status includes: Upon receiving the instruction associated with the ID After configuring and executing the parameters, the comprehensive health status index of the equipment is collected and calculated. In the header of the next reported one-way telemetry data stream, set a 32-bit instruction association ID field and an 8-bit status field; Fill in the instruction association ID field, and... The data is converted into a status code according to a preset mapping rule and filled into the status field, and then sent to the cloud server. The mapping rule is: if The transition status code is This indicates that the device is in good health after the command is executed; if The transition status code is This indicates that the device's health status is abnormal after the command is executed.

[0017] The beneficial effects of this invention are as follows: This invention optimizes communication links in industrial weak network environments using the QUIC protocol, improving the real-time performance of data transmission and the reliability of connections. By integrating multi-dimensional sensor data (vibration, temperature, load) from the reducer terminal and performing edge computing to generate a comprehensive equipment health status index, a comprehensive assessment of the equipment's operating status is achieved. This invention prioritizes the assessed data and encodes it into the QUIC stream ID, enabling the cloud server to quickly identify and prioritize emergency alarm data at the transport layer without parsing application-layer payloads, ensuring zero-latency response to critical information. Furthermore, the data aggregation and bit compression encoding scheme reduces the volume of time-series telemetry data, saving network bandwidth resources. For remote parameter configuration, the asynchronous feedback mechanism embeds the command execution status into regular telemetry data for reporting, enhancing the reliability of remote control under unstable network conditions. Detailed Implementation

[0018] The embodiments of the present invention are described in detail below. The embodiments described below are exemplary and intended to explain the present invention, and should not be construed as limiting the present invention.

[0019] The remote monitoring and parameter configuration system for speed reducers provided in this embodiment of the invention includes the following modules: a calculation module, a scheduling module, a reporting module, and a feedback module.

[0020] The calculation module is used to establish a QUIC protocol connection between the reducer terminal and the cloud server to achieve communication, including monitoring data reporting and configuration parameter distribution; the reducer terminal periodically collects vibration, temperature and load data, performs fast Fourier transform on the vibration data to extract the main frequency amplitude, and calculates the comprehensive health status index of the equipment by combining the time gradient of the temperature data and the current load torque.

[0021] The speed reducer terminal, acting as a QUIC client, proactively initiates a connection request to the cloud server. Both parties complete a 1-RTT handshake process, including TLS 1.3 encryption negotiation, via UDP-based transmission, establishing a long-term encrypted connection. On this connection, monitoring data reporting is initiated by the speed reducer terminal through a client-initiated unidirectional stream, with a new stream started for each reporting cycle or event trigger. Configuration parameter distribution is handled by the cloud server through a server-initiated bidirectional stream. The server sends commands via this stream, and the terminal sends a simple ACK confirmation, rather than a complete execution result, in the return direction of the same stream.

[0022] The terminal's built-in microcontroller acquires vibration acceleration signals at a frequency of 1000Hz, and collects temperature and load torque values ​​once per second. After collecting 1024 vibration data points, a Hanning window function is applied for preprocessing, followed by a Fast Fourier Transform. The frequency corresponding to the maximum amplitude is searched in the resulting spectrum; this amplitude is the dominant frequency amplitude A. Simultaneously, the current temperature is calculated. Temperature from the previous second The difference between the two values ​​is used to obtain the time gradient G of the temperature data. The comprehensive health status index H of the equipment is calculated using the following formula: ,in , , H is a preset weighting coefficient, L is the current load torque, and a higher H value indicates a greater health risk to the equipment.

[0023] The scheduling module is used to compare the comprehensive health status index of the device with the preset multi-level health thresholds, determine the priority of the current data, and encode the priority into the preset bits of the flow ID of the QUIC unidirectional flow carrying the monitoring data, so that the cloud server can perform traffic scheduling.

[0024] Preset two-level health thresholds and ,and < If the calculated comprehensive health status index H > If H is between 1 and 1, then the data priority is defined as the highest level, 1; and If H < 2, then the priority is medium level 2; if H < 2, then the priority is medium level 2. If the priority is 1, then it is a normal level 3. The QUIC stream ID is a 62-bit integer, with the two highest bits reserved as priority bits. When creating a new unidirectional data stream, if the priority is 1, the two highest bits of the stream ID are set to binary 11; if the priority is 2, they are set to 10; if the priority is 3, they are set to 01. The cloud server's access gateway or load balancer can check these two bits of the stream ID and schedule high-priority streams to the priority processing queue without parsing the data packet content.

[0025] The reporting module is used to calculate the data aggregation window size by combining the smooth round-trip time and priority of the reducer terminal with the QUIC connection. Within the data reporting cycle, multiple time-series telemetry floating-point values ​​collected within the window size are converted into fixed-length binary sequences according to their respective range and accuracy constraints, spliced ​​into aggregated data blocks, bit-compressed and encoded, and then reported.

[0026] The speed reducer terminal obtains the smooth round-trip time from the QUIC connection instance. The data aggregation window size W is calculated using the following formula: W = (Base size / Priority coefficient) High-priority data has a larger priority coefficient, resulting in a smaller window size and more timely data reporting. For example, a temperature value with a range of 0 to 200℃ and an accuracy of 0.1℃ has 2001 discrete states, requiring 11 bits for encoding. The terminal converts all temperature and vibration amplitude data within an aggregation window into their respective fixed-length binary sequences using this method, concatenating these sequences end-to-end to form a large binary data block. This data block is then processed using the Gorilla compression algorithm, which calculates the differences between adjacent values ​​and performs variable-length encoding to compress the data volume before transmitting it via a unidirectional stream.

[0027] The feedback module is used to receive a parameter configuration instruction with an embedded instruction association ID via the QUIC bidirectional stream. After executing the instruction, the reducer terminal carries the same instruction association ID in the data frame of the subsequently reported unidirectional telemetry data stream, and fills the preset status field of the data frame with a status code mapped by the comprehensive health status index of the device generated immediately after execution, thereby realizing asynchronous feedback of the instruction execution status.

[0028] When the cloud server issues a command, such as adjusting the vibration sampling frequency, the command message contains a unique command association ID, such as a 32-bit UUID. Upon receiving this command on the bidirectional stream initiated by the server, the reducer terminal immediately adjusts its hardware configuration. After adjustment, the terminal immediately recalculates the overall health status index of the device. Assuming the newly calculated index H value is 65, it is converted into a status code according to a preset mapping relationship, for example... Mapped to status codes This indicates successful execution and a good status. Mapped to This indicates successful execution, but the device has entered a warning state. In the next regular data reporting cycle, when the terminal constructs the reporting data frame, it will add two fields to the header: one for the received instruction association ID and the other for the status code. Along with the compressed telemetry data, it is reported via a one-way stream. The server can obtain the execution result of the command and its immediate impact on the device status by parsing the data frame.

[0029] In an optional embodiment, the step of performing a fast Fourier transform on the vibration data to extract the dominant frequency amplitude includes: A 1024-point Fast Fourier Transform was performed on the vibration acceleration signal with a period of 100ms to extract the maximum amplitude value A in the 0-500Hz frequency band as the main frequency amplitude value.

[0030] The time gradient of the temperature data is calculated in the following way: The temperature signal with a period of 1 second is read and the difference between the current temperature and the temperature 10 seconds ago is calculated to obtain the temperature gradient G. The temperature gradient G is used as the time gradient of the temperature data.

[0031] The comprehensive health status index of the computing device includes: Read the current load torque L and calculate the comprehensive health status index H of the equipment using the following formula; ; in, The preset maximum safety threshold for amplitude, The preset maximum safe threshold for the temperature gradient. This is the preset maximum safe threshold for load torque.

[0032] For example, a Fast Fourier Transform (FFT) is performed on the continuously acquired vibration signal to analyze the spectrum from 0 to 500 Hz, and the maximum amplitude A is determined to be 3.5g. Simultaneously, the current temperature is read as 75.5℃ and compared with 75.0℃ recorded 10 seconds prior, calculating the temperature gradient G as 0.5℃. Furthermore, the current load torque L is read as 600 Nm. Assume a preset maximum safe amplitude threshold. 10g, maximum safe threshold for temperature gradient The maximum safe load threshold is 5℃. The value is 1000 Nm. Substituting these values ​​into the formula for calculating the comprehensive health status index, we obtain a health index H of 0.7, which represents the current overall operating status of the equipment.

[0033] In an optional embodiment, determining the priority of the current data by comparing the comprehensive health status index of the device with preset multi-level health thresholds includes: The multi-level health thresholds for the comprehensive health status index H of the equipment are set to 0.9, 0.7, and 0.5. At that time, the device health status level is healthy, and the priority is 3; when When the device's health status level is good, the priority is 2; when When the device's health status level is "warning" and its priority is 1, then... At that time, the equipment health status level is dangerous, and the priority is 0.

[0034] The step of encoding the priority into the preset bits of the stream ID of the QUIC unidirectional stream carrying the monitoring data includes: writing a 2-bit priority value into the 5th and 6th bits of the stream ID of the QUIC unidirectional stream.

[0035] For example, if the calculated H value is 0.82, this value is compared sequentially with the internally set multi-level health thresholds. If the value is not greater than or equal to 0.9, it does not belong to the health level. However, it meets the condition of being greater than or equal to 0.7 and less than 0.9, so the device status is judged as good, and a priority P of 2, i.e., medium priority, is assigned accordingly.

[0036] The decimal priority value 2 is converted into a 2-bit binary number 10. Before sending data via the QUIC protocol, a new unidirectional data stream is created. This is done by positioning specific bits in the stream ID, specifically bits 5 and 6. The binary number 10 is written to these two bits, while the remaining bits of the stream ID are generated by the protocol stack according to other rules. For example, bits 5 and 6 of an original stream ID might be 0 and 1; after this operation, two bits in the new stream ID will be updated to 10, thus embedding the data priority information into the metadata of the network transmission.

[0037] In an optional embodiment, the calculated data aggregation window size includes: Obtain the smooth round-trip time (SRTT) in milliseconds from the QUIC connection session; set the base window size. And the priority weighting factor F, when the priority P is 0, 1, 2, 3, F takes the values ​​of 0.2, 0.5, 1.0, 2.0 respectively; Calculate the data aggregation window size W using the following formula; ,in, It is the preset minimum window size.

[0038] The Smooth Round-Trip Time (SRTT) is queried and obtained from currently active QUIC network connections, assuming an SRTT value of 50ms under current network conditions. Simultaneously, the data priority P, determined based on device health status, is obtained. For example, if the current priority P is 1, it indicates high-priority data. According to a preset mapping relationship, the weight factor F corresponding to priority 1 is found to be 0.5. Then, a fixed base window size is used. The value is 200ms, and the obtained SRTT value, weight factor F, and base window size will be used. Substitute the values ​​into the formula for calculation. The calculated data aggregation window size W is 150ms. The device will collect all sensor data within 150ms, aggregate them together, and send them in a single transmission, thus achieving linkage between data transmission frequency and network conditions and data importance.

[0039] In an optional embodiment, the step of converting multiple time-series telemetry floating-point values ​​collected within the window size into fixed-length binary sequences according to their respective range and accuracy constraints, and then concatenating them to form an aggregated data block, includes: Convert floating-point temperature values ​​ranging from -50.0℃ to 150.0℃ with an accuracy of 0.1℃ to integers between 0 and 2000 using the following formula: (floating-point temperature value) 50.0) 10, and encode the converted integer into an 11-bit binary temperature sequence; The floating-point value of the vibration dominant frequency amplitude, ranging from 0 to 10g with an accuracy of 0.01g, is converted to an integer between 0 and 1000 using the following formula: Amplitude 100, and encode the converted integer into a 10-bit binary vibration sequence; Convert the floating-point value of load torque in the range of 0 to 1000 Nm with an accuracy of 1 Nm to an integer between 0 and 1000, and encode the converted integer into a 10-bit binary load sequence; The 32-bit timestamp, 11-bit binary temperature sequence, 10-bit binary vibration sequence, and 10-bit binary load sequence within a data acquisition cycle are concatenated into a 63-bit data frame, and all data frames within the data aggregation window are concatenated sequentially to form an aggregated data block.

[0040] Specifically, at a certain moment, the data collected included a temperature of 75.2℃, a vibration amplitude of 2.58g, a load of 512Nm, and a 32-bit Unix timestamp. Based on the above conversion operations, the data yielded an 11-bit binary number 10011100100, a 10-bit binary number 0100000010, and a 10-bit binary number 1000000000. These four binary sequences were then concatenated in a predetermined order: the 32-bit timestamp, the 11-bit temperature sequence, the 10-bit vibration sequence, and the 10-bit load sequence. This formed a single-point data frame with a total length of 63 bits. This process was repeated for each acquisition cycle within the calculated data aggregation window size. For example, if five sets of data were acquired within the window size, five 63-bit data frames would be generated. These five data frames were then concatenated end-to-end to form a continuous aggregated data block with a total length of 315 bits.

[0041] In an optional embodiment, the bit compression encoding includes: The first 63-bit data frame within the aggregated data block is left unprocessed; starting from the second data frame, the data frame is XORed with the previous data frame; the number of leading zeros Z in the XOR result is calculated and represented by 6 bits; Z and the remaining bit stream starting from the first non-zero bit are concatenated into a compressed data frame, thereby compressing the parts of the data frames that change little.

[0042] Specifically, the first 63-bit single-point data frame in the data block will remain unchanged and serve as the basis for compression. When processing the second data frame, a bitwise XOR operation will be performed between the second data frame and the first data frame. Since continuously acquired data typically varies very little, the result of the XOR operation is likely to be a binary sequence with a large number of leading zeros. For example, the first 52 bits of the XOR result may be 0, followed by 11 bits of non-zero data.

[0043] The XOR result contains 52 leading zeros (Z). Convert the decimal number 52 to a 6-bit binary number: 110100. Extract all remaining bits from the XOR result, starting from the first non-zero bit (11 bits of non-zero data). Concatenate the 6-bit binary number (110100) representing the number of leading zeros with the subsequent 11 bits of data to form a compressed data frame of 17 bits. This compressed data frame replaces the original 63-bit data frame, thus reducing the amount of data that needs to be transmitted.

[0044] In an optional embodiment, the asynchronous feedback of the instruction execution status includes: Upon receiving the instruction associated with the ID After configuring and executing the parameters, the comprehensive health status index of the equipment is collected and calculated. In the header of the next reported one-way telemetry data stream, set a 32-bit instruction association ID field and an 8-bit status field; Fill in the instruction association ID field, and... The data is converted into a status code according to a preset mapping rule and filled into the status field, and then sent to the cloud server. The mapping rule is: if The transition status code is This indicates that the device is in good health after the command is executed; if The transition status code is This indicates that the device's health status is abnormal after the command is executed.

[0045] For example, when a device receives a control command from the cloud, the command carries a unique 32-bit command association ID, such as... The value is AABBCCDD in hexadecimal. The device executes this instruction, for example, to adjust the internal sampling frequency. After execution, the device triggers a temporary health assessment, calculating a new overall device health index. Assume the calculated value is 0.85. According to the preset mapping rules, since... Therefore, the mapped status code is hexadecimal 01, representing that the device is in good health after the command execution. When the device is preparing to send the next cycle of regular telemetry data, it adds a 40-bit header to the beginning of the data block to be sent. In the header, the first 32 bits are filled with the command ID value AABBCCDD, and the last 8 bits are filled with the status code 01. The data packet containing the command feedback information is sent out through the standard QUIC unidirectional data stream. When the cloud parses the regular telemetry data, it can asynchronously obtain the execution results of previously issued commands by checking the header, without requiring the device to establish an additional connection or send a special response packet.

[0046] The implementation principle of the remote monitoring and parameter configuration system for speed reducers in this invention is as follows: This invention utilizes the characteristics of the QUIC protocol to optimize communication links in industrial weak network environments, improving the real-time performance of data transmission and the reliability of connections. This invention integrates multi-dimensional sensor data such as vibration, temperature, and load at the speed reducer terminal, and performs calculations at the edge to generate a comprehensive equipment health status index, thereby achieving a comprehensive assessment of the equipment's operating status. This invention prioritizes the assessed data and encodes it into the QUIC stream ID, allowing the cloud server to quickly identify and prioritize emergency alarm data at the transport layer without parsing application layer payloads, ensuring timely response to critical information. The data aggregation and bit compression encoding scheme adopted by this invention effectively reduces the volume of time-series telemetry data, significantly saving network bandwidth resources. For remote parameter configuration, the asynchronous feedback mechanism designed in this invention cleverly embeds the execution status of commands into regular telemetry data and reports it together, greatly enhancing the reliability of remote control under unstable network conditions.

[0047] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention.

Claims

1. A remote monitoring and parameter configuration system for a speed reducer, characterized in that, The system includes the following modules: a calculation module, used to establish a QUIC protocol connection between the reducer terminal and the cloud server to achieve communication, including monitoring data reporting and configuration parameter distribution; the reducer terminal periodically collects vibration, temperature, and load data, performs a fast Fourier transform on the vibration data to extract the main frequency amplitude, and calculates the comprehensive health status index of the equipment by combining the time gradient of the temperature data with the current load torque; a scheduling module, used to compare the comprehensive health status index of the equipment with preset multi-level health thresholds to determine the priority of the current data, and encodes the priority into preset bits of the flow ID of the QUIC unidirectional flow carrying the monitoring data for traffic scheduling by the cloud server; and a reporting module, used by the reducer terminal to calculate the data aggregation window size by combining the smooth round-trip time and priority of the QUIC connection; within the data reporting period, multiple time-series telemetry floating-point values ​​collected within the window size are converted into fixed-length binary sequences according to their respective range and accuracy constraints, spliced ​​into aggregated data blocks, bit-compressed and encoded, and then reported. The feedback module is used to receive a parameter configuration instruction with an embedded instruction association ID via the QUIC bidirectional stream. After executing the instruction, the reducer terminal carries the same instruction association ID in the data frame of the subsequently reported unidirectional telemetry data stream, and fills the preset status field of the data frame with a status code mapped by the comprehensive health status index of the device generated immediately after execution, thereby realizing asynchronous feedback of the instruction execution status.

2. The remote monitoring and parameter configuration system for the speed reducer according to claim 1, characterized in that, The step of extracting the dominant frequency amplitude by performing a fast Fourier transform on the vibration data includes: A 1024-point Fast Fourier Transform was performed on the vibration acceleration signal with a period of 100ms to extract the maximum amplitude value A in the 0-500Hz frequency band as the main frequency amplitude value.

3. The remote monitoring and parameter configuration system for the speed reducer according to claim 2, characterized in that, The time gradient of the temperature data is calculated in the following way: The temperature signal with a period of 1 second is read and the difference between the current temperature and the temperature 10 seconds ago is calculated to obtain the temperature gradient G. The temperature gradient G is used as the time gradient of the temperature data.

4. The remote monitoring and parameter configuration system for the speed reducer according to claim 3, characterized in that, The comprehensive health status index of the computing device includes: Read the current load torque L and calculate the comprehensive health status index H of the equipment using the following formula; ； in, The preset maximum safety threshold for amplitude, The preset maximum safe threshold for the temperature gradient. This is the preset maximum safe threshold for load torque.

5. The remote monitoring and parameter configuration system for the speed reducer according to claim 1, characterized in that, The step of determining the priority of current data by comparing the comprehensive health status index of the equipment with preset multi-level health thresholds includes: The multi-level health thresholds for the comprehensive health status index H of the equipment are set to 0.9, 0.7, and 0.

5. when At that time, the device health status level is healthy, and the priority is 3; when When the device's health status level is good, the priority is 2; when When the device's health status level is "warning" and its priority is 1, then... At that time, the equipment health status level is dangerous, and the priority is 0.

6. The remote monitoring and parameter configuration system for the speed reducer according to claim 5, characterized in that, The step of encoding priority into the preset bits of the stream ID of the QUIC unidirectional stream carrying monitoring data includes: Write the 2-bit priority value into the 5th and 6th bits of the stream ID of the QUIC unidirectional stream.

7. The remote monitoring and parameter configuration system for the speed reducer according to claim 6, characterized in that, The calculated data aggregation window size includes: Obtain the Smooth Round Trip Time (SRTT) in milliseconds from the QUIC connection session; set the base window size. And the priority weighting factor F, when the priority P is 0, 1, 2, 3, F takes the values ​​of 0.2, 0.5, 1.0, 2.0 respectively; Calculate the data aggregation window size W using the following formula; ,in, It is the preset minimum window size.

8. The remote monitoring and parameter configuration system for the speed reducer according to claim 1, characterized in that, The process of converting multiple time-series telemetry floating-point values ​​collected within the window size into fixed-length binary sequences according to their respective range and accuracy constraints, and then concatenating them to form an aggregated data block includes: Convert floating-point temperature values ​​ranging from -50.0℃ to 150.0℃ with an accuracy of 0.1℃ to integers between 0 and 2000 using the following formula: (floating-point temperature value) 50.0) 10, and encode the converted integer into an 11-bit binary temperature sequence; The floating-point value of the vibration dominant frequency amplitude, ranging from 0 to 10g with an accuracy of 0.01g, is converted to an integer between 0 and 1000 using the following formula: Amplitude 100, and encode the converted integer into a 10-bit binary vibration sequence; Convert the floating-point value of load torque in the range of 0 to 1000 Nm with an accuracy of 1 Nm to an integer between 0 and 1000, and encode the converted integer into a 10-bit binary load sequence; The 32-bit timestamp, 11-bit binary temperature sequence, 10-bit binary vibration sequence, and 10-bit binary load sequence within a data acquisition cycle are concatenated into a 63-bit data frame, and all data frames within the data aggregation window are concatenated sequentially to form an aggregated data block.

9. The remote monitoring and parameter configuration system for the speed reducer according to claim 8, characterized in that, The bit compression encoding includes: The first 63-bit data frame within the aggregated data block is left unprocessed; starting from the second data frame, the data frame is XORed with the previous data frame; the number of leading zeros Z in the XOR result is calculated and represented by 6 bits; Z and the remaining bit stream starting from the first non-zero bit are concatenated into a compressed data frame, thereby compressing the parts of the data frames that change little.

10. The remote monitoring and parameter configuration system for the speed reducer according to claim 1, characterized in that, The asynchronous feedback of instruction execution status includes: Upon receiving the instruction associated with ID After configuring and executing the parameters, the comprehensive health status index of the equipment is collected and calculated. In the header of the next reported one-way telemetry data stream, set a 32-bit instruction association ID field and an 8-bit status field; Fill in the instruction association ID field, and... The data is converted into a status code according to a preset mapping rule and filled into the status field, and then sent to the cloud server. The mapping rule is: if The transition status code is This indicates that the device is in good health after the command is executed; if The transition status code is This indicates that the device's health status is abnormal after the command is executed.

Images