Low-power-consumption embedded data recording system and algorithm for on-line monitoring of SF density

Abstract

The invention provides a low-power-consumption embedded data recording system and algorithm for on-line monitoring of SF density. The system is provided with an MCU main control unit, an external independent watchdog circuit, a LoRa wireless communication module, an RS485 interface, an SF density and temperature measurement sensor, a battery power supply and voltage acquisition module and a Flash data storage and real-time clock module. Algorithms of the system comprise an intelligent data recording strategy, a dynamic sensor activation and sleep mechanism, a power failure data protection and breakpoint continuation mechanism, a multi-parameter compression coding algorithm, a dynamic sampling frequency adjustment strategy, a real-time clock synchronization and timestamp calibration mechanism, a multi-sensor fusion sensing mechanism and an ultra-low power consumption system architecture mechanism. The low-power-consumption embedded data recording system and algorithm for on-line monitoring of the SF density have the functions of multi-data change trend intelligent judgment, multi-data fusion perception judgment, ultra-low power consumption and flexible communication, and are suitable for long-term field deployment.

Description

Technical Field

[0001] The present invention relates to the technical field of power monitoring systems, and in particular to a low-power embedded data recording system and algorithm for online monitoring of SF6 density. Background Art

[0002] SF6 gas is widely used in high-voltage electrical equipment. Its density and humidity changes directly affect the safety and stability of equipment operation. Therefore, real-time data monitoring of SF6 gas is extremely important to ensure the stable operation of high-voltage electrical equipment. Currently, there are many devices and systems for monitoring SF6 gas on the market, but they have many inconveniences, such as frequent data recording without screening, which shortens the Flash storage life, high overall power consumption of the monitoring system, and is not suitable for long-term battery-powered deployment; partial neglect of the "validity" judgment of parameter changes, resulting in the storage of a large amount of redundant data, which may be lost after power failure and cannot be continued after system recovery; single communication mode, limited deployment scenarios, low data compression efficiency, and low storage utilization, which seriously affect the application and monitoring of SF6 gas in high-voltage electrical equipment. Therefore, based on the application and monitoring needs of SF6 gas in high-voltage electrical equipment, the applicant proposes a low-power embedded data recording system and algorithm for online monitoring of SF6 density. The embedded system and algorithm have intelligent judgment capabilities, low power operation, compressed storage and breakpoint persistence functions, and are suitable for long-term online monitoring of the gas status of high-voltage power equipment. Summary of the Invention

[0003] To solve the above technical problems, the present invention proposes a low-power embedded data recording system and algorithm for online monitoring of SF6 density. The system is equipped with an MCU main control unit, an external independent watchdog circuit, a LoRa wireless communication module and an RS485 interface, SF6 density and temperature measurement sensors, a battery power supply and voltage acquisition module, a Flash data storage and a real-time clock module. The system algorithm sets an intelligent data recording strategy, a dynamic sensor activation and sleep mechanism, a power-off data protection and breakpoint persistence mechanism, a multi-parameter compression encoding algorithm, a dynamic sampling frequency adjustment strategy, a real-time clock synchronization and timestamp calibration mechanism, a multi-sensor fusion perception mechanism and an ultra-low power system architecture mechanism. The low-power embedded data recording system and algorithm for online monitoring of SF6 density realizes intelligent judgment of multiple data change trends, multi-data fusion perception judgment, ultra-low power consumption and flexible communication functions.

[0004] To achieve the above object, the technical solution adopted by the present invention is:

[0005] A low-power embedded data recording system and algorithm for online monitoring of SF6 density are characterized in that: the low-power embedded data recording system for online monitoring of SF6 density is equipped with an MCU main control unit, an external independent watchdog circuit, a LoRa wireless communication module and an RS485 interface, SF6 density and temperature measurement sensors, a battery power supply and voltage acquisition module, a Flash data storage and a real-time clock module; the algorithm of the low-power embedded data recording system for online monitoring of SF6 density is equipped with an intelligent data recording strategy, a dynamic sensor activation and sleep mechanism, a power-off data protection and breakpoint persistence mechanism, a multi-parameter compression encoding algorithm, a dynamic sampling frequency adjustment strategy, a real-time clock synchronization and timestamp calibration mechanism, a multi-sensor fusion perception mechanism and an ultra-low power system architecture mechanism.

[0006] Furthermore, the algorithm of the low-power embedded data recording system for online monitoring of SF6 density is provided with an intelligent data recording strategy. The intelligent data recording strategy analyzes the sensor data change rate ΔT and Δρ through a sliding time window; records data when a valid change is detected that exceeds a preset threshold; if the changes in temperature, pressure, and density are all less than the threshold within 10 minutes, the data is not written to the Flash data storage during that period; thus, redundant data writing is reduced, storage life is extended, and power consumption is reduced; specifically:

[0007] 1) Sliding time window mechanism

[0008] The system sets a sliding window with a fixed time length, which can be 5 minutes, 10 minutes, or a user-defined period. Whenever a new round of sampled data enters the window, a window sliding is triggered and data analysis is performed.

[0009] Window length W: defines the number of groups of data collected within the period, N groups;

[0010] Window update: Use timing or sampling count sliding window to maintain real-time performance;

[0011] Analysis target: The maximum change value of each parameter in the statistical window, that is:

[0012] ΔT_max=max(T)-min(T)

[0013] ΔP_max=max(P)-min(P)

[0014] Δρmax=max(ρ)-min(ρ)

[0015] 2) Valid change determination logic

[0016] To distinguish real data changes from system noise or minor fluctuations, the system presets an "effective change threshold" for each parameter. When the change in a parameter within the window exceeds the threshold, data recording is triggered.

[0017] Judgment conditions:

[0018] If ΔT_max≥θ_T or ΔPmax≥θP or Δρ_max≥θ_ρ:

[0019] If the data has a "valid change", record the current data point; otherwise, skip the write operation and continue sampling;

[0020] Through intelligent data recording, the system automatically masks fluctuations or stable state data within a short period of time, and writes data when meaningful environmental or equipment changes are detected.

[0021] Furthermore, the algorithm of the low-power embedded data recording system for online monitoring of SF6 density is provided with a dynamic sensor activation and sleep mechanism. The dynamic sensor activation and sleep mechanism adopts a two-level monitoring architecture. The system is in a periodic low-power wake-up state and uses a low-speed ADC for coarse sampling. When it is detected that the data change trend exceeds the set threshold, it enters a high-precision measurement mode and activates all sensors and communication modules, realizing the "lightweight detection + fine capture" energy consumption control strategy.

[0022] Furthermore, the algorithm of the low-power embedded data recording system for online monitoring of SF6 density is provided with a power-off data protection and breakpoint resumption mechanism. The power-off data protection and breakpoint resumption mechanism solves the problems of data loss and write interruption in abnormal situations such as unexpected power failure or system reset, thereby ensuring data continuity and reliability; specifically:

[0023] 1) Flash partition cyclic writing mechanism:

[0024] The external Flash memory space is divided into fixed-size pages, each of which contains a write flag byte, such as 0xAA for "written" and 0xFF for "not written".

[0025] Data content; CRC checksum; when the system writes data, it writes the data first, then the CRC, and finally updates the write flag byte to ensure atomicity;

[0026] 2) Breakpoint continuation process:

[0027] After each cold start or power-off restart, the system first executes a page scan algorithm, reading the write flag bits in page order. Once the last page in the "written" state is found, the latest valid address is determined. Subsequent data writing begins on the page following that address. If a page's flag bit is abnormal or the CRC check fails, the system automatically skips the page and marks it invalid to prevent data contamination.

[0028] 3) Status determination and fault tolerance mechanism:

[0029] A three-state flag mechanism is used to improve robustness: 0xFF: empty page; 0x55: writing in progress; 0xAA: writing completed;

[0030] Write process: The page is being initialized and written. It is marked as 0x55. After writing is completed, it is marked as 0xAA. If the power is cut off during the process, the page detected as 0x55 at startup will be automatically rolled back and discarded.

[0031] 4) Functional advantages and effects:

[0032] Avoid data inconsistency, address confusion or information loss caused by accidental power failure; provide continuous writing capability and automatic recovery capability after interruption; improve Flash service life and avoid repeated writing to the same page; ensure time sequence consistency and sampling data continuity, which is conducive to subsequent data traceability analysis.

[0033] Furthermore, the algorithm of the low-power embedded data recording system for online monitoring of SF6 density is provided with a multi-parameter compression coding algorithm. The multi-parameter compression coding algorithm introduces variable-length coding compression mechanisms such as Huffman; adapts to data formats such as temperature, pressure, and density; compresses the storage data volume without sacrificing measurement accuracy, and improves storage efficiency. Specifically:

[0034] 1) Difference coding - Δ coding

[0035] To avoid repeated storage of data with small fluctuations, this algorithm first performs difference encoding on the continuously sampled raw data, recording only the difference Δ between adjacent sample values. This method is particularly suitable for physical quantities such as temperature, pressure, and density, which have small fluctuations and stable trends within a short period of time.

[0036] Original sampling sequence: X1, X2, X3, ..., Xn

[0037] Difference sequence: Δ2 = X2-X1, Δ3 = X3-X2, ..., Δn = Xn-X n-1

[0038] By only saving the initial value X1 and the subsequent difference value Δ sequence, the complete data stream can be reconstructed;

[0039] The algorithm significantly reduces the numerical range of most data, and the Δ value is usually much smaller than the original value, providing a basis for subsequent variable-length coding;

[0040] 2) Variable-length coding - Huffman coding

[0041] To further compress the storage volume of the delta value sequence, the classic Huffman variable-length coding algorithm is introduced. By counting the frequencies of different differences in the delta value sequence and constructing a Huffman tree, compression is achieved by using shorter codes for high-frequency differences and longer codes for low-frequency differences. The coding dictionary can be dynamically generated during the initial data collection phase or pre-set in the device firmware, allowing dynamic updates to adapt to environmental changes or different measurement objects. This mechanism is particularly suitable for compressing small-range integer values, and when combined with delta coding, the compression rate is significantly improved.

[0042] 3) Multi-parameter data format adaptation

[0043] The algorithm is adaptable to a variety of parameter data formats, with a flexible packaging structure and unified difference processing logic:

[0044] Temperature-T: usually with an accuracy of 0.1°C, a small difference range, and significant Δ encoding effect;

[0045] Pressure-P: expressed in kPa or MPa, changes slowly, suitable for sparse sampling + compression;

[0046] Density - ρ: usually calculated from temperature and pressure or collected by sensors, with small fluctuations and high compression potential;

[0047] For the case of multi-parameter parallel sampling, the following packet structure is used:

[0048] [Timestamp][T1][P1][ρ1]

[0049] [ΔT2][ΔP2][Δρ2]

[0050] The timestamp can be implicitly generated based on the sampling period and recorded only when an event triggers, such as wake-up or non-periodic data occurs;

[0051] 4) Storage and decoding strategy

[0052] The encoded compressed data is written to Flash or EEPROM in the form of a byte stream, and the reliability is improved with the CRC check code.

[0053] Furthermore, the algorithm of the low-power embedded data recording system for online monitoring of SF6 density is provided with a dynamic sampling frequency adjustment strategy, which analyzes the "stability" of the gas state in real time; when stable, the sampling frequency is reduced to once per hour; when abnormal fluctuations occur, the frequency is automatically increased to once per minute; and an adaptive optimal balance between power consumption and data quality is achieved; specifically:

[0054] 1) Stability analysis model:

[0055] The strategy uses the volatility of the measured gas parameters as the core criterion to dynamically evaluate whether the current system state is "stable". The monitoring parameters involved include but are not limited to temperature (T), pressure (P), and density (ρ). Within each sampling period, a comprehensive stability index (S) is derived by calculating the rate of change (Δ) or standard deviation (σ) of each parameter within a short time window. This index is used to determine whether the current state is "stable". The formula is:

[0056] S=max(ΔT,ΔP,Δρ) or S=max(σ_T,σ_P,σ_ρ)

[0057] 2) Adaptive frequency adjustment mechanism, the system sets two key sampling frequency levels:

[0058] Low frequency sampling mode: sampling frequency is 1 time / hour;

[0059] High frequency sampling mode: sampling frequency is 1 time / minute;

[0060] The system is judged based on the stability index S and the preset threshold θ_s: if S < θ_s, the system is stable and switches to low-frequency mode; if S ≥ θ_s, the system fluctuates abnormally and switches to high-frequency mode. The frequency adjustment has a hysteresis control mechanism, which means that the switching condition must be met for multiple consecutive cycles before it can be executed to avoid frequent oscillation switching.

[0061] 3) State switching logic

[0062] When the state switches from high frequency to low frequency, the switching condition is S<θ_s within N consecutive cycles, which is considered stable and the new sampling frequency is 1 time / hour; when the state switches from low frequency to high frequency, the switching condition is S≥θ_s within any cycle, which is considered abnormal fluctuation and the new sampling frequency is 1 time / minute. N can be set to 3 to 5 to ensure switching stability.

[0063] 4) Power consumption and data quality optimization effect

[0064] In a stable state, the gas in the substation is sealed normally and the system is in low-frequency mode, which greatly reduces sampling power consumption and data redundancy anomalies. In a sudden state, gas leakage occurs and the temperature rises rapidly. The system quickly enters high-frequency mode to capture the details of the change in time, ensuring the monitoring sensitivity and achieving the goal of "few measurements but precise". The data is dense at critical moments and sparse in stable conditions.

[0065] Furthermore, the algorithm of the low-power embedded data recording system for online monitoring of SF6 density is provided with a real-time clock synchronization and timestamp calibration mechanism. The real-time clock synchronization and timestamp calibration mechanism has a built-in high-stability real-time clock module with power-off retention capability to ensure continuous timing when the equipment is powered off or in sleep mode; all collected data are automatically timestamped to achieve time series recording and traceability analysis of the data; the system supports remote time synchronization through the LoRa or RS485 communication interface, and the master station or host computer can regularly issue time calibration instructions to calibrate the local RTC to ensure long-term accurate operation of the system clock; through this mechanism, historical data has good timeliness and traceability, providing a reliable basis for subsequent abnormality analysis, trend judgment and operation diagnosis.

[0066] Furthermore, the algorithm of the low-power embedded data recording system for online monitoring of SF6 density is set with a multi-sensor fusion perception mechanism, which integrates multiple sensor parameters such as density, temperature and pressure to achieve multi-dimensional fusion perception of the SF6 gas state, thereby improving the integrity and accuracy of the monitoring data; the communication system adopts a modular architecture design, supports LoRa remote wireless communication and RS485 field bus communication dual-channel mode, and adapts to the transmission requirements in different deployment scenarios; LoRa is suitable for long-distance, low-power data reporting scenarios, and RS485 is suitable for high-reliability, low-latency communication environments in industrial sites; the modular design facilitates flexible expansion and multi-scenario deployment, adapts to complex industrial environments such as substations and GIS equipment indoors and outdoors, and realizes remote and local synchronous transmission and management of monitoring data.

[0067] Furthermore, the algorithm of the low-power embedded data recording system for online monitoring of SF6 density is provided with an ultra-low-power system architecture mechanism, which adopts a multi-level low-power management strategy, and the overall current consumption can be controlled below 15μA in sleep mode; all peripheral modules adopt an on-demand wake-up and on-demand power supply control mechanism, activating relevant functional modules only when needed, and automatically powering off or entering a low-power state when idle, effectively reducing the overall energy consumption of the system and extending the battery life of the equipment; it is suitable for remote sites and long-term unmanned operation in the field; it significantly reduces operating costs and enhances system battery life.

[0068] The benefits of this application are:

[0069] 1. The low-power embedded data recording system and algorithm for online monitoring of SF6 density have the ability to intelligently judge data change trends and reduce redundant records;

[0070] 2. The low-power embedded data recording system and algorithm for online monitoring of SF6 density have the ability to continue writing after power failure, data protection and compressed storage, extending the life of Flash memory;

[0071] 3. A low-power embedded data recording system and algorithm for online monitoring of SF6 density, enabling multi-parameter fusion perception and highly reliable communication;

[0072] 4. A low-power embedded data logging system and algorithm for online monitoring of SF6 density enables adaptive adjustment of sampling frequency and power consumption, and a multi-level energy management strategy extends battery life;

[0073] 5. Low-power embedded data recording system and algorithm for online monitoring of SF6 density. Extremely low power consumption design and flexible communication make it suitable for long-term field deployment. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 Schematic diagram of the system architecture of the present invention;

[0075] Figure 2 This is a flow chart of the low power consumption control mechanism of the present invention;

[0076] Figure 3 This is a flow chart of the data collection, storage and recording mechanism of the present invention;

[0077] Figure 4 This is a flow chart of the Flash data writing mechanism of the present invention. DETAILED DESCRIPTION

[0078] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0079] like Figure 1-4 As shown in the figure, a low-power embedded data recording system and algorithm for online monitoring of SF6 density is shown. Figure 1As shown, the low-power embedded data recording system for online monitoring of SF6 density is equipped with an MCU main control unit, an external independent watchdog circuit, a LoRa wireless communication module and an RS485 interface, an SF6 density and temperature measurement sensor, a battery power supply and voltage acquisition module, a Flash data storage and a real-time clock module; the algorithm settings of the low-power embedded data recording system for online monitoring of SF6 density are intelligent data recording strategy, dynamic sensor activation and sleep mechanism, power-off data protection and breakpoint persistence mechanism, multi-parameter compression encoding algorithm, dynamic sampling frequency adjustment strategy, real-time clock synchronization and timestamp calibration mechanism, multi-sensor fusion perception mechanism and ultra-low power system architecture mechanism.

[0080] The algorithm of the low-power embedded data logging system for online SF6 density monitoring shown in the figure is equipped with an intelligent data logging strategy. The intelligent data logging strategy analyzes the sensor data change rate ΔT and Δρ through a sliding time window. When a valid change is detected that exceeds a preset threshold, data is recorded. If the change in temperature, pressure, and density is less than the threshold within 10 minutes, the data is not written to the Flash data storage during that period. This reduces redundant data writing, extends storage life, and reduces power consumption. Specifically:

[0081] 1) Sliding time window mechanism

[0082] The system sets a sliding window with a fixed time length, which can be 5 minutes, 10 minutes, or a user-defined period. Whenever a new round of sampled data enters the window, a window sliding is triggered and data analysis is performed.

[0083] Window length W: defines the number of groups of data collected within the period, N groups;

[0084] Window update: Use timing or sampling count sliding window to maintain real-time performance;

[0085] Analysis target: The maximum change value of each parameter in the statistical window, that is:

[0086] ΔT max=max(T)-min(T)

[0087] ΔP max=max(P)-min(P)

[0088] Δρmax=max(ρ)-min(ρ)

[0089] 2) Valid change determination logic

[0090] To distinguish real data changes from system noise or minor fluctuations, the system presets an "effective change threshold" for each parameter. When the change in a parameter within the window exceeds the threshold, data recording is triggered.

[0091] Judgment conditions:

[0092] If ΔT_max≥θT or ΔPmax≥θP or Δρmax≥θ_ρ:

[0093] If the data has a "valid change", record the current data point; otherwise, skip the write operation and continue sampling;

[0094] Through intelligent data recording, the system automatically masks fluctuations or stable state data within a short period of time, and writes data when meaningful environmental or equipment changes are detected.

[0095] The algorithm of the low-power embedded data recording system for online monitoring of SF6 density is set up with a dynamic sensor activation and sleep mechanism. The dynamic sensor activation and sleep mechanism adopts a two-level monitoring architecture. The system is in a periodic low-power wake-up state and uses a low-speed ADC for coarse sampling. When it is detected that the data change trend exceeds the set threshold, it enters a high-precision measurement mode and activates all sensors and communication modules, realizing the energy consumption control strategy of "lightweight detection + fine capture".

[0096] The algorithm of the low-power embedded data recording system for online monitoring of SF6 density is equipped with a power-off data protection and breakpoint persistence mechanism. The power-off data protection and breakpoint persistence mechanism solves the problems of data loss and write interruption in abnormal situations such as unexpected power failure or system reset, ensuring data continuity and reliability. Specifically:

[0097] 1) Flash partition cyclic writing mechanism:

[0098] The external Flash memory space is divided into fixed-size pages, each of which contains a write flag byte, such as 0xAA for "written" and 0xFF for "not written".

[0099] Data content; CRC checksum; when the system writes data, it writes the data first, then the CRC, and finally updates the write flag byte to ensure atomicity;

[0100] 2) Breakpoint continuation process:

[0101] After each cold start or power-off restart, the system first executes a page scan algorithm, reading the write flag bits in page order. Once the last page in the "written" state is found, the latest valid address is determined. Subsequent data writing begins on the page following that address. If a page's flag bit is abnormal or the CRC check fails, the system automatically skips the page and marks it invalid to prevent data contamination.

[0102] 3) Status determination and fault tolerance mechanism:

[0103] A three-state flag mechanism is used to improve robustness: 0xFF: empty page; 0x55: writing in progress; 0xAA: writing completed;

[0104] Write process: The page is being initialized and written. It is marked as 0x55. After writing is completed, it is marked as 0xAA. If the power is cut off during the process, the page detected as 0x55 at startup will be automatically rolled back and discarded.

[0105] 4) Functional advantages and effects:

[0106] Avoid data inconsistency, address confusion or information loss caused by accidental power failure; provide continuous writing capability and automatic recovery capability after interruption; improve Flash service life and avoid repeated writing to the same page; ensure time sequence consistency and sampling data continuity, which is conducive to subsequent data traceability analysis.

[0107] The algorithm of the low-power embedded data recording system for online monitoring of SF6 density is set up with a multi-parameter compression coding algorithm. The multi-parameter compression coding algorithm introduces variable-length coding compression mechanisms such as Huffman; adapts to data formats such as temperature, pressure, and density; compresses the storage data volume and improves storage efficiency without sacrificing measurement accuracy. Specifically:

[0108] 1) Difference coding - Δ coding

[0109] To avoid repeated storage of data with small fluctuations, this algorithm first performs difference encoding on the continuously sampled raw data, recording only the difference Δ between adjacent sample values. This method is particularly suitable for physical quantities such as temperature, pressure, and density, which have small fluctuations and stable trends within a short period of time.

[0110] Original sampling sequence: X1, X2, X3, ..., Xn

[0111] Difference sequence: Δ2 = X2-X1, Δ3 = X3-X2, ..., Δn = Xn-X n-1

[0112] By only saving the initial value X1 and the subsequent difference value Δ sequence, the complete data stream can be reconstructed;

[0113] The algorithm significantly reduces the numerical range of most data, and the Δ value is usually much smaller than the original value, providing a basis for subsequent variable-length coding;

[0114] 2) Variable-length coding - Huffman coding

[0115] To further compress the storage volume of the delta value sequence, the classic Huffman variable-length coding algorithm is introduced. By counting the frequencies of different differences in the delta value sequence and constructing a Huffman tree, compression is achieved by using shorter codes for high-frequency differences and longer codes for low-frequency differences. The coding dictionary can be dynamically generated during the initial data collection phase or pre-set in the device firmware, allowing dynamic updates to adapt to environmental changes or different measurement objects. This mechanism is particularly suitable for compressing small-range integer values, and when combined with delta coding, the compression rate is significantly improved.

[0116] 3) Multi-parameter data format adaptation

[0117] The algorithm is adaptable to a variety of parameter data formats, with a flexible packaging structure and unified difference processing logic:

[0118] Temperature-T: usually with an accuracy of 0.1°C, a small difference range, and significant Δ encoding effect;

[0119] Pressure-P: expressed in kPa or MPa, changes slowly, suitable for sparse sampling + compression;

[0120] Density - ρ: usually calculated from temperature and pressure or collected by sensors, with small fluctuations and high compression potential;

[0121] For the case of multi-parameter parallel sampling, the following packet structure is used:

[0122] [Timestamp][T1][P1][ρ1]

[0123] [ΔT2][ΔP2][Δρ2]

[0124] The timestamp can be implicitly generated based on the sampling period and recorded only when an event triggers, such as wake-up or non-periodic data occurs;

[0125] 4) Storage and decoding strategy

[0126] The encoded compressed data is written to Flash or EEPROM in the form of a byte stream, and the reliability is improved with the CRC check code.

[0127] The algorithm of the low-power embedded data logging system for online SF6 density monitoring is equipped with a dynamic sampling frequency adjustment strategy. The dynamic sampling frequency adjustment strategy analyzes the "stability" of the gas state in real time. When stable, the sampling frequency is reduced to once per hour. When abnormal fluctuations occur, the frequency is automatically increased to once per minute. This achieves an adaptive optimal balance between power consumption and data quality. Specifically,

[0128] 1) Stability analysis model:

[0129] The strategy uses the volatility of the measured gas parameters as the core criterion to dynamically evaluate whether the current system state is "stable". The monitoring parameters involved include but are not limited to temperature (T), pressure (P), and density (ρ). Within each sampling period, a comprehensive stability index (S) is derived by calculating the rate of change (Δ) or standard deviation (σ) of each parameter within a short time window. This index is used to determine whether the current state is "stable". The formula is:

[0130] S=max(ΔT,ΔP,Δρ) or S=max(σ_T,σ_P,σ_ρ)

[0131] 2) Adaptive frequency adjustment mechanism, the system sets two key sampling frequency levels:

[0132] Low frequency sampling mode: sampling frequency is 1 time / hour;

[0133] High frequency sampling mode: sampling frequency is 1 time / minute;

[0134] The system is judged based on the stability index S and the preset threshold θ_s: if S < θ_s, the system is stable and switches to low-frequency mode; if S ≥ θ_s, the system fluctuates abnormally and switches to high-frequency mode. The frequency adjustment has a hysteresis control mechanism, which means that the switching condition must be met for multiple consecutive cycles before it can be executed to avoid frequent oscillation switching.

[0135] 3) State switching logic

[0136] When the state switches from high frequency to low frequency, the switching condition is S<θ_s within N consecutive cycles, which is considered stable and the new sampling frequency is 1 time / hour; when the state switches from low frequency to high frequency, the switching condition is S≥θ_s within any cycle, which is considered abnormal fluctuation and the new sampling frequency is 1 time / minute. N can be set to 3 to 5 to ensure switching stability.

[0137] 4) Power consumption and data quality optimization effect

[0138] In a stable state, the gas in the substation is sealed normally and the system is in low-frequency mode, which greatly reduces sampling power consumption and data redundancy anomalies. In a sudden state, gas leakage occurs and the temperature rises rapidly. The system quickly enters high-frequency mode to capture the details of the change in time, ensuring the monitoring sensitivity and achieving the goal of "few measurements but precise". The data is dense at critical moments and sparse in stable conditions.

[0139] The algorithm of the low-power embedded data recording system for online monitoring of SF6 density shown in the figure is equipped with a real-time clock synchronization and timestamp calibration mechanism. The real-time clock synchronization and timestamp calibration mechanism shown in the figure has a built-in high-stability real-time clock module with power-off retention capability to ensure continuous timing when the equipment is powered off or in sleep mode; all collected data are automatically timestamped to achieve time series recording and traceability analysis of the data; the system supports remote time synchronization through the LoRa or RS485 communication interface, and the master station or host computer can regularly issue time calibration instructions to calibrate the local RTC to ensure long-term accurate operation of the system clock; through this mechanism, historical data has good timeliness and traceability, providing a reliable basis for subsequent abnormality analysis, trend judgment and operation diagnosis.

[0140] The algorithm setting of the low-power embedded data recording system for online monitoring of SF6 density shown in the figure has a multi-sensor fusion perception mechanism. The multi-sensor fusion perception mechanism shown in the figure integrates multiple sensor parameters such as density, temperature and pressure to achieve multi-dimensional fusion perception of the SF6 gas state, thereby improving the integrity and accuracy of the monitoring data. The communication system adopts a modular architecture design, supports LoRa long-distance wireless communication and RS485 field bus communication dual-channel mode, and adapts to the transmission requirements in different deployment scenarios. LoRa is suitable for long-distance, low-power data reporting scenarios, and RS485 is suitable for high-reliability, low-latency communication environments in industrial sites. The modular design facilitates flexible expansion and multi-scenario deployment, adapting to complex industrial environments such as substations and GIS equipment indoors and outdoors, and realizing remote and local synchronous transmission and management of monitoring data.

[0141] Figure 2 As shown, the algorithm setting of the low-power embedded data recording system for online monitoring of SF6 density is equipped with an ultra-low power system architecture mechanism. The ultra-low power system architecture mechanism adopts a multi-level low-power management strategy, and the overall current consumption can be controlled below 15μA in sleep mode; all peripheral modules adopt an on-demand wake-up and on-demand power supply control mechanism, and only activate related functional modules when needed. When idle, it automatically powers off or enters a low-power state, effectively reducing the overall energy consumption of the system and extending the battery life of the equipment; it is suitable for remote sites and long-term unmanned operation in the field; it significantly reduces operating costs and enhances system battery life.

[0142] The above description is merely a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any modification or equivalent variation based on the technical essence of the present invention shall still fall within the scope of protection claimed by the present invention.

Claims

1. Low-power embedded data logging system for online monitoring of SF6 density, characterized by: The low-power embedded data recording system for online monitoring of SF6 density is equipped with an MCU main control unit, an external independent watchdog circuit, a LoRa wireless communication module and an RS485 interface, an SF6 density and temperature measurement sensor, a battery power supply and voltage acquisition module, a Flash data storage and a real-time clock module; the algorithm of the low-power embedded data recording system for online monitoring of SF6 density is equipped with an intelligent data recording strategy, a dynamic sensor activation and sleep mechanism, a power-off data protection and breakpoint persistence mechanism, a multi-parameter compression encoding algorithm, a dynamic sampling frequency adjustment strategy, a real-time clock synchronization and timestamp calibration mechanism, a multi-sensor fusion perception mechanism and an ultra-low power system architecture mechanism.

2. The algorithm of the low-power embedded data recording system for online monitoring of SF6 density according to claim 1 is characterized in that: The algorithm of the low-power embedded data recording system for online monitoring of SF6 density is provided with an intelligent data recording strategy. The intelligent data recording strategy analyzes the sensor data change rate ΔT and Δρ through a sliding time window; records data when a valid change is detected that exceeds a preset threshold; if the changes in temperature, pressure, and density are all less than the threshold within 10 minutes, the data is not written to the Flash data storage during that period; thus, redundant data writing is reduced, storage life is extended, and power consumption is reduced. Specifically, 1) Sliding time window mechanism The system sets a sliding window with a fixed time length, which can be 5 minutes, 10 minutes, or a user-defined period. Whenever a new round of sampled data enters the window, a window sliding is triggered and data analysis is performed. Window length W: defines the number of groups of data collected within the period, N groups; Window update: Use timing or sampling count sliding window to maintain real-time performance; Analysis target: The maximum change value of each parameter in the statistical window, that is: ΔT max=max(T)-min(T) ΔP max=max(P)-min(P) Δρmax=max(ρ)-min(ρ) 2) Valid change determination logic To distinguish real data changes from system noise or minor fluctuations, the system presets an "effective change threshold" for each parameter. When the change in a parameter within the window exceeds the threshold, data recording is triggered. Judgment conditions: If ΔT_max≥θT or ΔP max≥θP or Δρmax≥θ_ρ: If the data has a "valid change", record the current data point; otherwise, skip the write operation and continue sampling; Through intelligent data recording, the system automatically masks fluctuations or stable state data within a short period of time, and writes data when meaningful environmental or equipment changes are detected.

3. The algorithm of the low-power embedded data recording system for online monitoring of SF6 density according to claim 1 is characterized in that: The algorithm of the low-power embedded data recording system for online monitoring of SF6 density is equipped with a dynamic sensor activation and sleep mechanism. The dynamic sensor activation and sleep mechanism adopts a two-level monitoring architecture. The system is in a periodic low-power wake-up state and uses a low-speed ADC for coarse sampling. When the data change trend exceeds the set threshold, it enters a high-precision measurement mode and activates all sensors and communication modules. Implement the energy consumption control strategy of "lightweight detection + fine capture".

4. The algorithm of the low-power embedded data recording system for online monitoring of SF6 density according to claim 1 is characterized in that: The algorithm of the low-power embedded data recording system for online monitoring of SF6 density is equipped with a power-off data protection and breakpoint persistence mechanism. The power-off data protection and breakpoint persistence mechanism solves the problems of data loss and write interruption in abnormal situations such as unexpected power outages or system resets, ensuring data continuity and reliability. Specifically: 1) Flash partition cyclic writing mechanism: The external Flash memory space is divided into fixed-size pages, each of which contains a write flag byte, such as 0xAA for "written" and 0xFF for "not written". Data content; CRC checksum: When the system writes data, it writes the data first, then the CRC, and finally updates the write flag byte to ensure atomicity. 2) Breakpoint continuation process: After each cold start or power-off restart, the system first executes a page scan algorithm, reading the write flag bits sequentially in page order. Once the last page in the "written" state is found, the latest valid address is determined. Subsequent data writing begins on the page following that address. If a page's flag bit is abnormal or fails the CRC check, the system automatically skips the page and marks it invalid, ensuring data is not contaminated. 3) Status determination and fault tolerance mechanism: A three-state flag mechanism is used to improve robustness: 0xFF: empty page; 0x55: writing in progress; 0xAA: writing completed; Writing process: Page initialization is being written, marked as 0x55; After writing is completed, it is marked as 0xAA; if the power is cut off during the process, the page detected as 0x55 at startup will be automatically rolled back and discarded; 4) Functional advantages and effects: Avoid data inconsistency, address confusion or information loss caused by accidental power failure; provide continuous writing capability and automatic recovery capability after interruption; improve Flash service life and avoid repeated writing to the same page; ensure time sequence consistency and sampling data continuity, which is conducive to subsequent data traceability analysis.

5. The algorithm of the low-power embedded data recording system for online monitoring of SF6 density according to claim 1 is characterized in that: The algorithm of the low-power embedded data recording system for online monitoring of SF6 density is set up with a multi-parameter compression coding algorithm. The multi-parameter compression coding algorithm introduces a variable-length coding compression mechanism such as Huffman; adapts to data formats such as temperature, pressure, and density; compresses the storage data volume without sacrificing measurement accuracy, and improves storage efficiency. Specifically: 1) Difference coding - Δ coding To avoid repeated storage of data with small fluctuations, this algorithm first performs difference encoding on the continuously sampled raw data, recording only the difference Δ between adjacent sample values. This method is particularly suitable for physical quantities such as temperature, pressure, and density, which have small fluctuations and stable trends within a short period of time. Original sampling sequence: X1, X2, X3, ..., Xn Difference sequence: Δ2 = X2-X1, Δ3 = X3-X2, ..., Δn = Xn-X n-1 By only saving the initial value X1 and the subsequent difference value Δ sequence, the complete data stream can be reconstructed; The algorithm significantly reduces the numerical range of most data, and the Δ value is usually much smaller than the original value, providing a basis for subsequent variable-length coding; 2) Variable-length coding - Huffman coding To further compress the storage volume of the delta value sequence, the classic Huffman variable-length coding algorithm is introduced. By counting the frequencies of different differences in the delta value sequence and constructing a Huffman tree, compression is achieved by using shorter codes for high-frequency differences and longer codes for low-frequency differences. The coding dictionary can be dynamically generated during the initial data collection phase or pre-set in the device firmware, allowing dynamic updates to adapt to environmental changes or different measurement objects. This mechanism is particularly suitable for compressing small-range integer values, and when combined with delta coding, the compression rate is significantly improved. 3) Multi-parameter data format adaptation The algorithm is adaptable to a variety of parameter data formats, with a flexible packaging structure and unified difference processing logic: Temperature-T: usually with an accuracy of 0.1°C, a small difference range, and significant Δ encoding effect; Pressure-P: expressed in kPa or MPa, changes slowly, suitable for sparse sampling + compression; Density - ρ: usually calculated from temperature and pressure or collected by sensors, with small fluctuations and high compression potential; For the case of multi-parameter parallel sampling, the following packet structure is used: [Timestamp][T1][P1][ρ1] [ΔT2][ΔP2][Δρ2] The timestamp can be implicitly generated based on the sampling period and recorded only when an event triggers, such as wake-up or non-periodic data occurs; 4) Storage and decoding strategy The encoded compressed data is written to Flash or EEPROM in the form of a byte stream, and the reliability is improved with the CRC check code.

6. The algorithm of the low-power embedded data recording system for online monitoring of SF6 density according to claim 1 is characterized in that: The algorithm of the low-power embedded data recording system for online monitoring of SF6 density is equipped with a dynamic sampling frequency adjustment strategy. The dynamic sampling frequency adjustment strategy analyzes the "stability" of the gas state in real time. When stable, the sampling frequency is reduced to once per hour. When abnormal fluctuations occur, the frequency is automatically increased to once per minute. This achieves an adaptive optimal balance between power consumption and data quality. Specifically, 1) Stability analysis model: The strategy uses the volatility of the measured gas parameters as the core criterion to dynamically evaluate whether the current system state is "stable". The monitoring parameters involved include but are not limited to temperature (T), pressure (P), and density (ρ). Within each sampling period, a comprehensive stability index (S) is derived by calculating the rate of change (Δ) or standard deviation (σ) of each parameter within a short time window. This index is used to determine whether the current state is "stable". The formula is: S=max(ΔT,ΔP,Δρ) or S=max(σ_T,σ_P,σ_ρ) 2) Adaptive frequency adjustment mechanism, the system sets two key sampling frequency levels: Low frequency sampling mode: sampling frequency is 1 time / hour; High frequency sampling mode: sampling frequency is 1 time / minute; The system is judged based on the stability index S and the preset threshold θ_s: if S < θ_s, the system is stable and switches to low-frequency mode; if S ≥ θ_s, the system fluctuates abnormally and switches to high-frequency mode. The frequency adjustment has a hysteresis control mechanism, which means that the switching condition must be met for multiple consecutive cycles before it can be executed to avoid frequent oscillation switching. 3) State switching logic When the state switches from high frequency to low frequency, the switching condition is S<θ_s within N consecutive cycles, which is considered stable and the new sampling frequency is 1 time / hour; when the state switches from low frequency to high frequency, the switching condition is S≥θ_s within any cycle, which is considered abnormal fluctuation and the new sampling frequency is 1 time / minute. N can be set to 3 to 5 to ensure switching stability. 4) Power consumption and data quality optimization effect In a stable state, the gas in the substation is sealed normally and the system is in low-frequency mode, which greatly reduces sampling power consumption and data redundancy anomalies. In a sudden change state, gas leakage occurs and the temperature rises rapidly. The system quickly enters high-frequency mode to capture the details of the changes in time, ensuring the monitoring sensitivity and achieving the goal of "few measurements but precise". The data is dense at critical moments and sparse in stable conditions.

7. The algorithm of the low-power embedded data recording system for online monitoring of SF6 density according to claim 1 is characterized in that: The algorithm of the low-power embedded data recording system for online monitoring of SF6 density is provided with a real-time clock synchronization and timestamp calibration mechanism. The real-time clock synchronization and timestamp calibration mechanism has a built-in high-stability real-time clock module with power-off retention capability to ensure continuous timing during device power-off or sleep periods; all collected data are automatically timestamped to achieve time series recording and traceability analysis of the data; the system supports remote time synchronization through the LoRa or RS485 communication interface, and the master station or host computer can regularly issue time calibration instructions to calibrate the local RTC to ensure the long-term accurate operation of the system clock; through this mechanism, historical data has good timeliness and traceability, providing a reliable basis for subsequent abnormality analysis, trend judgment and operation diagnosis.

8. The algorithm of the low-power embedded data recording system for online monitoring of SF6 density according to claim 1 is characterized in that: The algorithm of the low-power embedded data recording system for online monitoring of SF6 density is set up with a multi-sensor fusion perception mechanism. The multi-sensor fusion perception mechanism integrates multiple sensor parameters such as density, temperature and pressure to achieve multi-dimensional fusion perception of SF6 gas status, thereby improving the integrity and accuracy of monitoring data; The communication system adopts a modular architecture design and supports dual-channel communication of LoRa long-distance wireless communication and RS485 fieldbus communication to adapt to the transmission requirements in different deployment scenarios. LoRa is suitable for long-distance, low-power data reporting scenarios, while RS485 is suitable for high-reliability, low-latency communication environments in industrial sites. The modular design facilitates flexible expansion and multi-scenario deployment, adapting to complex industrial environments such as substations and GIS equipment indoors and outdoors, and realizing remote and local synchronous transmission and management of monitoring data.

9. The low-power embedded data recording system and algorithm for online monitoring of SF6 density according to claim 1, characterized in that: The algorithm of the low-power embedded data logging system for online monitoring of SF6 density is equipped with an ultra-low-power system architecture mechanism. The ultra-low-power system architecture mechanism adopts a multi-level low-power management strategy, and the overall current consumption can be controlled below 15μA in sleep mode. All peripheral modules adopt an on-demand wake-up and on-demand power supply control mechanism, activating relevant functional modules only when needed and automatically powering off or entering a low-power state when idle, effectively reducing the overall energy consumption of the system and extending the battery life of the device. Suitable for remote sites and long-term unmanned operation in the wild; Significantly reduce operating costs and enhance system endurance.

Images