Energy-aware adaptive correction method for wireless sensor node duty cycle

Abstract

The application discloses an energy-aware wireless sensor node working cycle adaptive correction method, and belongs to the technical field of wireless sensor node energy management. The method acquires energy state parameters, vibration state parameters and communication state parameters, calculates the actual accumulated energy of the current cycle and predicts the accumulated energy of the next cycle; corrects the energy consumption of wireless communication tasks according to the communication state parameters to generate an overall energy consumption threshold; generates a dynamic safety margin and a dynamic wake-up energy threshold according to the energy prediction deviation; determines the node working mode of the next cycle according to the predicted accumulated energy of the next cycle and each energy threshold, and adjusts the adaptive correction cycle length of the next cycle according to the task execution feedback. The application can improve the node energy utilization efficiency, task execution reliability and long-term operation stability.

Description

Technical Field

[0001] This application relates to the field of energy management technology for wireless sensor nodes, and in particular to an adaptive correction method for the duty cycle of wireless sensor nodes based on energy sensing. Background Technology

[0002] With the development of industrial equipment condition monitoring technology, long-term, continuous monitoring of the operational status of key components in equipment such as high-temperature fans, motors, pumps, and compressors is required for temperature and vibration monitoring. Taking the bearing of a high-temperature fan in a steel plant as an example, this component operates under high temperature, high load, and strong vibration environments. Its temperature and vibration status effectively reflect the bearing's lubrication condition, wear level, imbalance, and early failure trends. Therefore, continuous collection and uploading of relevant monitoring data is necessary to achieve equipment fault early warning and operational status assessment. However, such industrial sites often face challenges such as difficult wiring, limited power supply, and high battery replacement and maintenance costs. Traditional wired monitoring methods or battery-powered methods are insufficient to meet the requirements for long-term stable operation. Therefore, wireless sensor nodes are needed for long-term, stable data collection and transmission.

[0003] Existing wireless sensor nodes typically perform monitoring tasks using battery power or ambient energy harvesting. Ambient energy harvesting-based wireless sensor nodes can acquire energy through piezoelectric energy harvesting, electromagnetic energy harvesting, or radio frequency energy harvesting, storing the energy in supercapacitors or small-capacity batteries. When the energy storage unit reaches a certain voltage or meets preset wake-up conditions, the wireless sensor node initiates temperature measurement, vibration measurement, data processing, and wireless communication tasks. To reduce communication energy consumption, some wireless sensor nodes also employ backscatter communication, modulating and reflecting the radio frequency signal emitted by the reader by changing their own impedance state, thereby achieving low-power data upload.

[0004] For example, the Chinese invention patent with announcement number CN105511591B discloses a DVFS adjustment algorithm based on dual-threshold power consumption adaptation, which includes: acquiring real-time equipment power consumption and utilization information, and adjusting the operating state according to preset thresholds, thereby reducing energy consumption and improving energy utilization efficiency. This type of solution mainly implements power consumption control and energy-saving management based on changes in equipment load, and can improve energy consumption levels to a certain extent.

[0005] The above-mentioned technology has at least the following technical problems:

[0006] In temperature and vibration monitoring scenarios, wireless sensor nodes control their operating status using fixed-period wake-up or simple voltage threshold triggering. Changes in vibration conditions can cause fluctuations in the efficiency of piezoelectric-electromagnetic composite energy harvesting, while complex electromagnetic environments can affect communication link quality and communication energy consumption requirements. Consequently, the actual energy available to the node and the actual energy consumption for the task exhibit dynamic fluctuations. Existing solutions still execute monitoring and communication tasks according to preset working cycles or fixed energy consumption models. This can easily lead to problems such as invalid wake-ups, communication failures, repeated retransmissions, and rapid depletion of stored energy when energy supply is insufficient or the communication environment deteriorates. Consequently, these issues reduce the energy utilization efficiency and long-term stable operation capability of the wireless sensor nodes. Summary of the Invention

[0007] To address the technical problems of existing wireless sensor nodes, which, when using fixed-period wake-up or simple voltage threshold triggering, cannot adapt to fluctuations in the vibration energy input of high-temperature fan bearings in steel plants and changes in the radio frequency communication environment, easily leading to invalid wake-up, communication failure, energy waste, and inaccurate duty cycle adjustment, this application provides an energy-sensing-based adaptive correction method for the duty cycle of wireless sensor nodes. The technical solution is as follows:

[0008] On the one hand, an adaptive correction method for the duty cycle of a wireless sensor node based on energy sensing is provided. This method is implemented by the wireless sensor node and includes: after entering the current adaptive correction cycle, the wireless sensor node acquires energy state parameters, vibration state parameters, and communication state parameters. Among them, the energy state parameters include energy storage unit voltage, charging current, and historical accumulated energy; the vibration state parameters include vibration frequency and vibration amplitude; and the communication state parameters include radio frequency power density, echo signal-to-noise ratio, historical communication success rate, historical retransmission count, and target communication distance.

[0009] Based on the task type of the wireless sensor node, the node's operation tasks are divided into temperature measurement tasks, vibration measurement tasks, data processing tasks, wireless communication tasks, and power management tasks, with temperature measurement tasks, vibration measurement tasks, data processing tasks, wireless communication tasks, and power management tasks serving as the basic tasks.

[0010] Based on the historical energy consumption corresponding to the temperature measurement task, vibration measurement task, data processing task, and power management task, the energy requirements for the temperature measurement task, vibration measurement task, data processing task, and power management task are determined. The actual accumulated energy for the current cycle is determined based on the energy storage unit voltage, charging current, and historical accumulated energy. The predicted accumulated energy for the next adaptive correction cycle is determined based on the vibration state parameters and historical accumulated energy.

[0011] The communication energy consumption corresponding to the wireless communication task is corrected based on the communication status parameters to obtain the corrected communication energy consumption. The corrected communication energy consumption is then summarized with the energy requirements of the temperature measurement task, vibration measurement task, data processing task, and power management task to generate the overall energy consumption threshold.

[0012] The dynamic safety margin is determined based on the prediction deviation between the historical adaptive correction cycle predicted cumulative energy and the current cycle actual cumulative energy. The dynamic safety margin provides safety compensation for the overall energy consumption threshold and generates the dynamic wake-up energy threshold.

[0013] Based on the predicted cumulative energy, total energy consumption threshold, and dynamic wake-up energy threshold of the current cycle, determine the mode of the wireless sensor node to perform a complete task, sampling buffer task, core monitoring task, or deep sleep task. Based on the actual energy consumption, task completion rate, and communication success rate generated during task execution, adjust the length of the adaptive correction cycle for the next adaptive correction cycle.

[0014] The beneficial effects of the technical solutions provided in this application include at least the following:

[0015] 1. This application obtains the voltage, charging current, vibration frequency, vibration amplitude, and historical records of actual new energy added in the energy storage unit. Based on the vibration state parameters and the historical records of actual new energy added in the energy storage unit, the predicted new energy for the next adaptive correction cycle is determined. This enables the wireless sensing node to predict the subsequent energy supply capacity in advance based on changes in vibration energy input, avoiding reliance on fixed cycles or single voltage thresholds to control node wake-up and improving the node's adaptability to vibration energy fluctuation scenarios.

[0016] 2. This application dynamically corrects the communication energy consumption corresponding to the wireless communication task by acquiring radio frequency power density, echo signal-to-noise ratio, historical communication success rate, historical retransmission count and target communication distance. The corrected communication energy consumption is used to generate the overall energy consumption threshold, so that the overall energy consumption threshold can be adjusted with changes in communication environment and communication distance. This reduces the risk of communication failure, repeated retransmission and communication energy consumption estimation distortion in complex electromagnetic environments, and improves the reliability of communication task execution.

[0017] 3. Based on the currently available stored energy, dynamic wake-up energy threshold, overall energy consumption threshold, and minimum monitoring energy threshold, this application determines the execution mode of the wireless sensor node to perform a complete task, a sampling buffer task, a core monitoring task, or a deep sleep task. This enables the node to select the corresponding task level under different energy supply conditions, avoiding invalid wake-ups and task interruptions caused by forcibly executing a complete task when energy is insufficient, and improving the monitoring data acquisition and uploading capabilities when energy is sufficient.

[0018] 4. This application calculates the absolute value of the prediction deviation between the predicted increase in energy in the current cycle and the actual increase in energy in the current cycle. It then combines the absolute values ​​of the prediction deviations corresponding to several consecutive historical adaptive correction cycles and takes the average value as the dynamic safety margin for the current cycle. Based on the actual energy consumption, task completion rate, and communication success rate generated during task execution, the adaptive correction cycle length of the next adaptive correction cycle is adjusted. This enables the wireless sensor node to continuously correct its working cycle based on energy prediction errors and task execution feedback, thereby reducing energy waste and improving long-term operational stability and the accuracy of adaptive adjustment of the working cycle. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 A flowchart illustrating the overall architecture implementation method provided in this application embodiment;

[0021] Figure 2 A flowchart illustrating the parameter acquisition and state perception implementation method provided in this application embodiment;

[0022] Figure 3 This is a schematic diagram illustrating the implementation process of energy prediction and total energy consumption threshold generation provided in the embodiments of this application;

[0023] Figure 4 A flowchart illustrating the implementation method of dynamic security margin and task mode decision-making provided in the embodiments of this application;

[0024] Figure 5 A schematic diagram illustrating the implementation process of feedback correction and periodic adaptive update in the embodiments of this application is provided. Detailed Implementation

[0025] like Figure 1The flowchart shown in this application embodiment illustrates the overall architecture implementation method. Embodiment 1 provides an adaptive correction method for the working cycle of a wireless sensor node based on energy sensing. The wireless sensor node provided in this embodiment is deployed at the industrial equipment condition monitoring site to continuously monitor the equipment's operating status over a long period under conditions of no continuous external power supply or limited external power supply. Preferably, the wireless sensor node in this embodiment is installed in key parts of industrial equipment with continuous vibration characteristics, such as high-temperature fan bearing housings, motor bearing housings, pump bearing housings, compressor bearing housings, etc., in steel plants. It collects the vibration energy generated during equipment operation as the node's main power source and performs real-time sensing and dynamic analysis of the equipment's operating status. This allows the node to automatically adjust its working cycle based on the current energy supply capacity, achieving long-term stable operation.

[0026] The wireless sensing node in this embodiment mainly includes a vibration energy acquisition unit, an energy storage unit, a temperature acquisition unit, a vibration acquisition unit, a communication monitoring unit, an energy management unit, and a duty cycle control unit. These units are connected via an internal control bus and a data bus, forming a complete system for energy acquisition, status sensing, energy prediction, and duty cycle control.

[0027] The vibration energy harvesting unit is used to acquire the mechanical vibration energy generated during the operation of industrial equipment and convert it into electrical energy output. Specifically, the vibration energy harvesting unit can adopt a piezoelectric energy harvesting structure, an electromagnetic energy harvesting structure, or a piezoelectric-electromagnetic composite energy harvesting structure. The electrical energy output by the vibration energy harvesting unit is processed by a rectifier circuit and a voltage regulator circuit before being input into the energy storage unit for storage.

[0028] The energy storage unit stores the ambient energy acquired by the vibration energy acquisition unit and serves as an energy buffer for the wireless sensing node. In this embodiment, the energy storage unit can be implemented using a supercapacitor, a small lithium battery, or a combination of both. The energy storage unit smooths out energy fluctuations generated during vibration energy acquisition and provides a stable operating voltage to the temperature acquisition unit, vibration acquisition unit, communication monitoring unit, energy management unit, and duty cycle control unit.

[0029] The temperature acquisition unit is used to collect operating temperature information of key components of the equipment. In this embodiment, the temperature acquisition unit is installed near the bearing housing of the equipment and can use a digital temperature sensor or a thermocouple sensor to detect the temperature. The temperature acquisition unit completes the temperature data acquisition according to the sampling instructions issued by the duty cycle control unit and sends the acquisition results to the energy management unit for caching and processing. The acquired temperature parameters can reflect the changes in the equipment's operating load status, friction status, and lubrication status, and are an important basis for assessing the equipment's health status.

[0030] The vibration acquisition unit is used to collect vibration status information during equipment operation. Specifically, this includes vibration frequency Fv, vibration amplitude Av, and vibration waveform data Sv. The vibration acquisition unit can use MEMS (Micro-Electro-Mechanical Systems) accelerometers or piezoelectric vibration sensors for vibration detection. The acquired vibration data is used for two purposes: monitoring equipment operating status and providing early warning of faults, and predicting energy harvesting capacity for the next adaptive correction cycle.

[0031] The communication monitoring unit is used to monitor the current wireless communication environment status in real time. Specifically, the communication monitoring unit is responsible for acquiring communication status parameters such as radio frequency power density (Pr), echo signal-to-noise ratio (SNR), historical communication success rate (Rs), historical retransmission count (Nr), and target communication distance (Dt). Among them, radio frequency power density (Pr) reflects the wireless signal coverage strength in the current space; echo signal-to-noise ratio (SNR) reflects the communication link quality; historical communication success rate (Rs) reflects the reliability of historical data transmission; historical retransmission count (Nr) reflects communication failures; and target communication distance (Dt) reflects the link transmission distance.

[0032] The energy management unit uses the aforementioned communication status parameters to analyze changes in the current communication environment and calculates the actual energy consumption requirements of the communication task, providing a basis for generating the overall energy consumption threshold in the future.

[0033] The energy management unit is connected to the vibration energy acquisition unit, energy storage unit, temperature acquisition unit, vibration acquisition unit, and communication monitoring unit. Its main functions include energy state parameter acquisition, state data caching, historical data management, energy prediction and analysis, communication energy consumption correction, overall energy consumption threshold generation, and dynamic safety margin generation. Specifically, the energy management unit first acquires the energy storage unit voltage Vcap, charging current Icap, and historical accumulated energy Eh to form an energy state parameter set; then it acquires the vibration frequency Fv and vibration amplitude Av to form a vibration state parameter set; and further acquires the radio frequency power density Pr, echo signal-to-noise ratio SNR, historical communication success rate Rs, historical retransmission count Nr, and target communication distance Dt to form a communication state parameter set. Based on the current vibration state parameters and historical accumulated energy, the energy management unit predicts the predicted accumulated energy Epredict for the next adaptive correction cycle; calculates the corrected communication energy consumption Ecom based on the communication state parameters; further generates the overall energy consumption threshold Etotal based on the energy requirements of the temperature measurement task Et, vibration measurement task Ev, data processing task Ep, power management task Em, and the final corrected communication energy consumption Efinal; at the same time, it generates a dynamic safety margin Safety based on the prediction deviation, and finally forms the dynamic wake-up energy threshold Ewake.

[0034] The duty cycle control unit is used to determine whether the wireless sensor node enters a full task, sampling buffer task, core monitoring task, or deep sleep state based on the current available stored energy and various energy thresholds output by the energy management unit, and to adjust the adaptive correction cycle length of the next adaptive correction cycle based on task execution feedback.

[0035] At the same time, the work cycle control unit is also responsible for collecting feedback information such as the actual energy consumption, task completion rate and communication success rate of the current cycle, and sending the feedback results to the energy management unit as the basis for updating the parameters of the next adaptive correction cycle.

[0036] like Figure 2 The diagram shown is a flowchart of the parameter acquisition and state perception implementation method provided in this application. Figure 2 It can be seen that after the wireless sensing node enters the current adaptive correction cycle, it first activates the parameter acquisition and status awareness function to obtain various key parameters required for the node's operating status in real time, providing a data foundation for subsequent energy prediction, task energy consumption calculation, and adaptive correction of the work cycle. First, the energy management unit monitors the node's energy storage status in real time, collecting the energy storage unit voltage sampling value, charging current sampling value, and historical accumulated energy Eh within the current adaptive correction cycle. The energy storage unit voltage sampling value and charging current sampling value are then averaged to obtain the average energy storage unit voltage Vcap and average charging current Icap corresponding to the current adaptive correction cycle. The average energy storage unit voltage Vcap reflects the available voltage level of the node's energy storage unit within the current adaptive correction cycle, and the average charging current Icap reflects the average energy input capability after environmental vibration energy conversion within the current adaptive correction cycle. The historical accumulated energy Eh records the accumulated energy changes over multiple previous adaptive correction cycles and is used for trend analysis and energy prediction. The actual accumulated energy Eactual in the current cycle can be expressed as:

[0037] Eactual = Eh + Vcap × Icap × Tc

[0038] Where Tc represents the current adaptive correction cycle length. By continuously acquiring and recording energy storage state parameters, the energy management unit can form a complete energy state curve, providing a quantitative basis for predicting the energy available in the next cycle.

[0039] Subsequently, the vibration acquisition unit initiates real-time monitoring of the equipment's operating vibration status, including vibration frequency Fv and vibration amplitude Av. Vibration frequency Fv reflects the operating status of the rotating parts of the equipment, while vibration amplitude Av reflects changes in the equipment's vibration intensity.

[0040] The vibration acquisition unit caches the acquired vibration frequency Fv and vibration amplitude Av in the node's local memory and periodically sends them to the energy management unit. The vibration frequency Fv and vibration amplitude Av are used by the energy management unit to calculate the vibration energy input intensity Ie and the predicted cumulative energy Epredict for the next adaptive correction cycle in the subsequent energy prediction and overall energy consumption assessment stages.

[0041] Simultaneously, the communication monitoring unit initiates state awareness of the wireless communication environment, including parameters such as RF power density Pr, echo signal-to-noise ratio (SNR), historical communication success rate (Rs), historical retransmission count (Nr), and target communication distance (Dt). Among these, RF power density Pr reflects the environmental signal strength, echo signal-to-noise ratio (SNR) assesses link quality, historical communication success rate (Rs) and retransmission count (Nr) measure communication stability, and target communication distance (Dt) is used to correct the communication energy consumption model.

[0042] The energy management unit dynamically corrects the energy consumption of wireless communication tasks using communication status parameters to obtain the final corrected communication energy consumption Efinal. The final corrected communication energy consumption Efinal is then combined with the energy requirements of temperature measurement task Et, vibration measurement task Ev, data processing task Ep, and power management task Em to generate the overall energy consumption threshold Etotal, providing a decision-making basis for the work cycle control unit.

[0043] In terms of data processing and caching, the energy management unit establishes a multi-dimensional historical data cache to record time-series data of energy storage status, vibration status, and communication status. Energy status parameters, vibration status parameters, and communication status parameters are stored in timestamp order to facilitate subsequent trend analysis and anomaly detection. The data caching mechanism supports circular caching, ensuring that historical data occupies limited storage space during long-term operation, while allowing energy prediction and safety margin calculation to access data from the most recent N periods. The parameter update mechanism includes periodic refresh and event-triggered refresh: periodic refresh collects and updates parameters at regular intervals according to the current adaptive correction period length Tc, while event-triggered refresh immediately collects and updates relevant parameters when vibration amplitude exceeds a threshold or communication link anomalies occur, ensuring responsiveness to sudden changes in status.

[0044] Through the above steps, the wireless sensing node completes a comprehensive perception of the energy state, vibration state, and communication state in each adaptive correction cycle, and caches and updates the collected data in a timely manner, providing a complete and reliable parameter basis for subsequent steps. This enables the node to dynamically adjust its working cycle, predict energy supply, generate task energy consumption thresholds, and perform safety margin adjustments, thereby achieving adaptive closed-loop control of the entire node.

[0045] like Figure 3The diagram shown is a schematic representation of the implementation process for energy prediction and overall energy consumption threshold generation provided in this application. Figure 3 It can be seen that after collecting energy state parameters, vibration state parameters, and communication state parameters, the energy management unit enters the energy prediction and overall energy consumption assessment stage. This embodiment establishes a task energy consumption model, an energy accumulation prediction model, and a modified communication energy consumption model to comprehensively analyze the energy supply capacity and task execution requirements of the wireless sensor node in the future period, thereby generating an overall energy consumption threshold and providing a basis for subsequent dynamic safety margin calculations and operating mode decisions.

[0046] Specifically, the energy management unit first establishes a task energy consumption database based on the node's historical operation records. The task energy consumption database records the actual energy consumption of different tasks within multiple historical adaptive correction cycles, including the actual energy consumption of temperature measurement tasks, vibration measurement tasks, data processing tasks, and power management tasks.

[0047] The energy management unit uses a sliding window statistical method to extract the average energy consumption of temperature measurement, vibration measurement, data processing, and power management tasks over the most recent N adaptive correction cycles. Abnormal data is then removed, resulting in the energy requirements Et for temperature measurement, Ev for vibration measurement, Ep for data processing, and Em for power management. These energy requirements, along with the final corrected communication energy consumption Efinal, are used to generate the overall energy consumption threshold Etotal.

[0048] The energy management unit then calculates the actual accumulated energy for the current cycle.

[0049] After determining the energy requirements Et for temperature measurement, Ev for vibration measurement, Ep for data processing, and Em for power management, the energy management unit further calculates the actual accumulated energy Eactual for the current cycle. Specifically, it obtains the energy storage unit voltage Vcap, charging current Icap, and historical accumulated energy Eh from the energy storage unit and constructs an energy state vector:

[0050] Ve = {Vcap, Icap, Eh}

[0051] Among them, the energy storage unit voltage reflects the current energy storage level, the charging current reflects the current environmental energy input capacity, and the historical accumulated energy reflects the energy growth trend over multiple historical cycles. Subsequently, the energy management unit calculates the new energy input Ein for the current cycle based on the current cycle length Tc.

[0052] Ein = Vcap × Icap × Tc,

[0053] Furthermore, the actual cumulative energy for the current period (Eactual) is generated based on the new energy input and historical cumulative energy.

[0054] Eactual = Eh + Ein

[0055] To avoid distortion of energy assessment results caused by short-term vibration fluctuations, the energy management unit performs mean filtering on voltage and current data corresponding to multiple consecutive sampling times and uses an exponential smoothing algorithm to generate the energy change curve for the current period. By analyzing the slope of the energy change curve, the energy growth rate of the current node can be obtained, thereby improving the stability of subsequent prediction results.

[0056] After obtaining the actual accumulated energy (Eactual) for the current period, the energy management unit further generates the predicted accumulated energy (Epredict) for the next adaptive correction period. Since wireless sensor nodes primarily rely on device vibration to obtain environmental energy, the device vibration state can reflect the energy supply capacity within the next adaptive correction period.

[0057] The energy management unit uses a sliding window method to perform mean filtering on the vibration frequency and amplitude collected within the current adaptive correction period, obtaining the average vibration frequency Fv and average vibration amplitude Av for the current adaptive correction period. Based on the average vibration frequency Fv and average vibration amplitude Av, the vibration energy input intensity Ie is calculated to estimate the relative vibration energy input level that the node can obtain within the current adaptive correction period, where Ie is a calibration model or relative intensity index, rather than strict physical energy.

[0058] Ie = α × Fv × Av,

[0059] Wherein, α is the vibration energy conversion coefficient, which is determined based on the pre-established calibration relationship between vibration frequency, vibration amplitude and actual new energy. The calibration relationship includes calibration coefficient, calibration formula or calibration lookup table.

[0060] The energy management unit extracts the actual new energy sequence He for the most recent N adaptive correction cycles from the historical accumulated energy Eh, and calculates the average value of each actual new energy in the actual new energy sequence He, using the average value as the historical predicted new energy Kh.

[0061] Based on the current vibration energy input intensity Ie and the next adaptive correction period length Tc_next, calculate the predicted new vibration energy Evib:

[0062] Evib = Ie × Tc_next,

[0063] Based on the historical predicted incremental energy Kh and the vibration predicted incremental energy Evib, calculate the predicted incremental energy Eadd_pred for the next adaptive correction cycle:

[0064] Eadd_pred=λ1×Kh+λ2×Evib,

[0065] Based on the actual accumulated energy Eactual and the predicted new energy Eadd_pred for the current period, generate the predicted accumulated energy Epredict for the next adaptive correction period:

[0066] Epredict=Eactual+Eadd_pred,

[0067] Wherein, λ1 and λ2 are prediction weight coefficients, satisfying λ1 + λ2 = 1; λ1 is used to characterize the contribution of historical energy change trends to the predicted new energy, and λ2 is used to characterize the contribution of the current vibration state to the predicted new energy.

[0068] In a preferred embodiment, λ1 is set to 0.6 and λ2 is set to 0.4. When the equipment is operating relatively stably, the value of λ1 is increased; when the equipment vibration changes significantly, the value of λ2 is increased, thereby combining historical energy change trends and the current vibration state to predict the energy supply for the next adaptive correction cycle.

[0069] After the next adaptive correction cycle ends, the energy management unit obtains the actual cumulative energy Eactual corresponding to that adaptive correction cycle, and compares the predicted cumulative energy Epredict generated in the previous adaptive correction cycle with the actual cumulative energy Eactual to obtain the prediction deviation ΔE.

[0070] ΔE = Epredict - Eactual

[0071] The energy management unit acquires the prediction deviations for N consecutive historical adaptive correction cycles, where N and the N used for historical data statistics mentioned earlier represent the same preset statistical cycle number. The prediction deviation is the difference between the predicted cumulative energy Epredict and the actual cumulative energy Eactual for the corresponding adaptive correction cycle. The absolute value of each prediction deviation is calculated, and the dynamic safety margin is determined based on the absolute value of each prediction deviation. A larger absolute value of the prediction deviation results in a larger dynamic safety margin; a smaller absolute value of the prediction deviation results in a smaller dynamic safety margin.

[0072] After completing the predicted cumulative energy generation, the communication energy consumption correction stage begins. Since wireless sensor nodes are affected by factors such as changes in the wireless environment, communication distance, and link stability during actual operation, the actual energy consumption of the communication task will dynamically change. Therefore, this embodiment utilizes communication status parameters to dynamically correct the energy consumption of the wireless communication task.

[0073] Specifically, the communication monitoring unit acquires the radio frequency power density Pr, the echo signal-to-noise ratio (SNR), and the target communication distance Dt, and sends these parameters to the energy management unit. The energy management unit first normalizes the radio frequency power density Pr and the echo SNR to eliminate the influence of different parameter dimensions on the calculation results. The normalized parameters are denoted as Pr' and SNR', respectively. Then, based on preset weights ω1 and ω2, the normalized radio frequency power density Pr' and echo SNR' are weighted and calculated to obtain the current communication link quality parameter Qlink.

[0074] Qlink=ω1×Pr'+ω2×SNR',

[0075] Wherein, ω1 and ω2 represent the corresponding weight coefficients. The weight coefficients ω1 and ω2 are preset values, with a value range of [0,1], and satisfy ω1+ω2=1. In the preferred embodiment, ω1=0.6 and ω2=0.4 can be taken, so that the radio frequency power density and signal quality account for 60% and 40% respectively in the link evaluation.

[0076] The energy management unit queries a preset communication energy consumption correction table based on the current communication link quality parameter Qlink and the target communication distance Dt to determine the communication distance correction coefficient Kd. The preset communication energy consumption correction table is established based on the calibration relationship between different communication link quality parameters, different target communication distances, and actual communication energy consumption.

[0077] The lower the current communication link quality parameter Qlink, or the larger the target communication distance Dt, the higher the energy required for the wireless sensor node to complete a wireless communication task, and the larger the corresponding communication distance correction coefficient Kd. The higher the current communication link quality parameter Qlink, or the smaller the target communication distance Dt, the smaller the corresponding communication distance correction coefficient Kd.

[0078] The energy management unit corrects the basic communication energy consumption Ec based on the communication distance correction factor Kd, resulting in the corrected communication energy consumption Ecom:

[0079] Ecom = Ec × Kd,

[0080] Where Ec represents the basic communication energy consumption required for a wireless sensor node to complete a wireless communication task under preset communication link quality and preset communication distance; Kd represents the communication distance correction coefficient, which refers to the comprehensive correction degree of the basic communication energy consumption by the current communication link quality and target communication distance; Ecom represents the corrected communication energy consumption, which refers to the energy expected to be required for a wireless sensor node to complete a wireless communication task under the current communication link quality and target communication distance.

[0081] To further improve the accuracy of communication energy consumption assessment results, the energy management unit obtains the historical communication success rate Rs. The historical communication success rate Rs is determined based on the communication results of the most recent M wireless communication tasks:

[0082] Rs = Ns / M,

[0083] Where Ns represents the number of times data upload was successfully completed in the most recent M wireless communication tasks, M represents the total number of wireless communication tasks used to calculate the historical communication success rate, and the historical communication success rate Rs ranges from 0 to 1.

[0084] A higher historical communication success rate (Rs) indicates a higher stability in data upload across the current wireless communication link, and a lower risk of additional energy consumption due to communication failures or repeated transmissions. Conversely, a lower historical communication success rate (Rs) indicates a higher risk of additional energy consumption due to communication failures or repeated transmissions. Therefore, the energy management unit uses 1 - Rs to characterize the additional energy consumption risk corresponding to the wireless communication task, and compensates for the corrected communication energy consumption (Ecom) based on this risk, obtaining the final corrected communication energy consumption result (Efinal).

[0085] Efinal=Ecom×[1+μ×(1-Rs)],

[0086] Wherein, μ is the communication energy consumption compensation coefficient, used to limit the maximum energy consumption compensation ratio corresponding to the risk of communication failure or repeated transmission. The communication energy consumption compensation coefficient μ is determined according to the proportion of additional communication energy consumption caused by communication failure or repeated transmission in historical wireless communication tasks to the corrected communication energy consumption Ecom. The value of μ ranges from 0 to 0.5. In the preferred embodiment, μ is taken as 0.1.

[0087] When the historical communication success rate Rs equals 1, it means that the most recent M wireless communication tasks have been successfully completed. No additional energy consumption compensation is performed, and the final corrected communication energy consumption result Efinal equals the corrected communication energy consumption Ecom. When the historical communication success rate Rs decreases, the risk of communication failure or repeated transmission increases, and the final corrected communication energy consumption result Efinal increases accordingly.

[0088] Finally, the energy management unit uses the final corrected communication energy consumption result Efinal as the energy consumption requirement for the wireless communication task corresponding to the current adaptive correction cycle, and combines it with the energy requirements Et for the temperature measurement task, Ev for the vibration measurement task, Ep for the data processing task, and Em for the power management task to generate the overall energy consumption threshold Etotal.

[0089] Etotal=Et+Ev+Ep+Em+Efinal,

[0090] The above methods enable dynamic adjustment of wireless communication task energy consumption requirements based on the current communication environment and communication reliability, making the overall energy consumption threshold more accurately reflect the energy consumption level during actual node operation.

[0091] The total energy consumption threshold Etotal represents the total energy required for the wireless sensor node to perform temperature measurement, vibration measurement, data processing, wireless communication, and power management tasks. It is used for subsequent dynamic wake-up energy threshold generation and operating mode determination. In summary, this embodiment achieves a complete processing flow from state parameter acquisition results to total energy consumption threshold generation by constructing a task energy consumption model, an energy prediction model, and a modified communication energy consumption model. Quantitative analysis of energy supply capacity and energy demand levels provides a reliable data foundation for subsequent dynamic safety margin calculations, dynamic wake-up energy threshold generation, and adaptive adjustment of operating modes for the wireless sensor node, thereby improving energy utilization efficiency and task execution stability during long-term node operation.

[0092] like Figure 4 The diagram shown is a flowchart illustrating the implementation method of dynamic security margin and task mode decision-making provided in this application. Figure 4 As can be seen, after the overall energy consumption threshold is generated, the wireless sensor node enters the energy risk assessment and working mode decision-making stage. This embodiment establishes a dynamic safety margin mechanism by analyzing the deviation between the predicted accumulated energy and the actual accumulated energy, and generates a dynamic wake-up energy threshold based on this, thereby realizing the adaptive adjustment of the working mode of the wireless sensor node, thus avoiding problems such as frequent node wake-ups, task execution failures, or excessive energy consumption caused by energy prediction errors.

[0093] When the absolute value of the prediction deviation is small, it indicates that the prediction model can accurately reflect the node's future energy acquisition capability; when the absolute value of the prediction deviation is large, it indicates that the current environmental state has changed or that there is a significant difference between the historical trend and the current operating state. In actual operation, the prediction deviation of a single cycle may be affected by factors such as instantaneous vibration fluctuations, changes in the communication environment, or load changes. Therefore, this embodiment further establishes a continuous cycle deviation analysis mechanism. Specifically, the energy management unit reads the prediction deviation sequence HΔE corresponding to N consecutive adaptive correction cycles:

[0094] HΔE={ΔE1,ΔE2,ΔE3,…,ΔEN},

[0095] The energy management unit calculates the dynamic safety margin (Safety) based on the absolute value of the prediction deviation ΔE corresponding to N consecutive adaptive correction cycles.

[0096] Safety=(│ΔE1│+│ΔE2│+…+│ΔEN│) / N,

[0097] Where N represents the number of consecutive adaptive correction cycles, and ΔE1, ΔE2, ..., ΔEN represent the prediction deviations corresponding to N consecutive adaptive correction cycles.

[0098] The dynamic safety margin is used to compensate for energy prediction errors generated over multiple consecutive adaptive correction cycles. When the absolute value of the prediction deviation increases over multiple consecutive adaptive correction cycles, the dynamic safety margin increases accordingly; when the absolute value of the prediction deviation decreases over multiple consecutive adaptive correction cycles, the dynamic safety margin decreases accordingly, thereby adaptively adjusting the safety compensation amount corresponding to the overall energy consumption threshold according to the changes in energy prediction errors.

[0099] After generating the dynamic safety margin (Safety), the energy management unit generates the dynamic wake-up energy threshold (Ewake) based on the total energy consumption threshold (Etotal) used for determining the task mode in the next adaptive correction cycle and the dynamic safety margin (Safety): Ewake = Etotal + Safety, where Etotal represents the total energy required for the wireless sensor node to perform all basic tasks in the next adaptive correction cycle, Safety represents the dynamic safety margin used to compensate for energy prediction errors, and Ewake represents the predicted energy threshold required for the wireless sensor node to perform the complete task in the next adaptive correction cycle.

[0100] Furthermore, the duty cycle control unit determines the task mode to be executed by the wireless sensor node in the next adaptive correction cycle based on the predicted cumulative energy Epredict, the dynamic wake-up energy threshold Ewake, the total energy consumption threshold Etotal, and the minimum monitoring energy threshold Emin.

[0101] When Epredict ≥ Ewake, it indicates that the predicted cumulative energy of the next adaptive correction cycle not only meets the energy requirements of all basic tasks but also covers the safety compensation amount corresponding to the energy prediction error. Therefore, the work cycle control unit determines the task mode of the next adaptive correction cycle as the complete task mode, controlling the wireless sensor node to perform temperature acquisition, vibration acquisition, data processing, and wireless communication tasks. When Etotal ≤ Epredict < Ewake, it indicates that the predicted cumulative energy of the next adaptive correction cycle meets the energy requirements corresponding to the overall energy consumption threshold but is insufficient to cover the safety compensation amount corresponding to the dynamic safety margin. Therefore, the work cycle control unit determines the task mode of the next adaptive correction cycle as the sampling buffer mode, controlling the wireless sensor node to perform temperature acquisition and vibration acquisition tasks, storing the acquisition results in the local buffer, and pausing the wireless communication task. When Emin≤Epredict<Etotal is satisfied, it indicates that the predicted cumulative energy of the next adaptive correction cycle is insufficient to support all basic tasks, but can meet the minimum energy requirements of the core monitoring task. Therefore, the working cycle control unit determines the task mode of the next adaptive correction cycle as the core monitoring mode, controls the wireless sensor node to perform key temperature monitoring tasks and low-frequency vibration sampling tasks, and suspends data processing and wireless communication tasks.

[0102] When Epredict < Emin, it indicates that the predicted cumulative energy of the next adaptive correction cycle is insufficient to meet the minimum energy requirements of the core monitoring task. Therefore, the working cycle control unit determines the task mode of the next adaptive correction cycle as deep sleep mode, and only keeps the vibration energy acquisition unit and necessary control circuits running.

[0103] After making the working mode decision, the working cycle control unit further analyzes the current cycle task execution results, including feedback parameters such as actual energy consumption (Euse), task completion rate (Rt), and communication success rate (Rc), and sends the feedback results to the energy management unit. The energy management unit writes the feedback parameters into the historical operation database as the basis for updating the energy prediction model, communication energy consumption model, and dynamic safety margin model for the next adaptive correction cycle. Simultaneously, the working cycle control unit adjusts the length of the next adaptive correction cycle (Tc_next) based on the current cycle task execution status and remaining energy storage level. When energy supply is sufficient for several consecutive cycles, the working cycle is shortened and the monitoring frequency is increased; when energy supply is insufficient for several consecutive cycles, the working cycle is extended and the node energy consumption level is reduced. Through this mechanism, the node can dynamically adjust its working frequency according to its actual energy supply capacity, achieving a balance between monitoring performance and energy consumption level.

[0104] In summary, this embodiment generates a dynamic safety margin through prediction deviation analysis and constructs a dynamic wake-up energy threshold using the overall energy consumption threshold and the dynamic safety margin. Based on this, a four-level task decision mechanism is established to achieve adaptive adjustment of the wireless sensor node's working mode and working cycle. This method effectively solves the problems of low energy utilization, high task execution failure rate, and poor environmental adaptability caused by traditional wireless sensor nodes using fixed wake-up thresholds and fixed working cycles, enabling wireless sensor nodes to maintain long-term stable operation under conditions of vibration energy fluctuations and changes in the communication environment.

[0105] like Figure 5The diagram illustrates the implementation flow of feedback correction and periodic adaptive update in this application. Embodiment two of this application: Based on the unchanged aspects of embodiment one, after the working mode decision and task execution, the wireless sensor node enters the feedback correction and periodic adaptive update phase. This embodiment dynamically corrects the aforementioned energy prediction model, communication energy consumption model, and adaptive correction cycle length by acquiring the execution results of various tasks within the current adaptive correction cycle and the actual energy changes, thereby achieving continuous optimization and long-term adaptive operation of the wireless sensor node's working cycle. Specifically, the working cycle control unit acquires task execution feedback information at the end of the current cycle. This feedback information includes parameters such as the actual energy consumption (Euse) of the current cycle, the remaining energy (Eremain) at the end of the cycle, the task completion rate (Rt), the communication success rate (Rc), the actual number of communications (Nc), and the actual number of retransmissions (Nr_real). The actual energy consumption (Euse) reflects the real energy consumed by the node in performing monitoring, processing, and communication tasks within the current cycle; the remaining energy (Eremain) reflects the node's current energy storage level; the task completion rate (Rt) reflects the degree of completion of the predetermined tasks for this cycle; and the communication success rate (Rc) reflects the link reliability during data upload. The aforementioned feedback parameters are sent to the energy management unit for unified storage and written into the historical operation database.

[0106] After receiving the feedback parameters, the energy management unit first assesses the accuracy of the energy prediction results for the current cycle. Specifically, it compares the predicted cumulative energy Epredict with the actual cumulative energy Eactual obtained after the current cycle ends to obtain the prediction deviation ΔE. Simultaneously, it establishes a historical sequence of prediction errors.

[0107] Herror={ΔE1,ΔE2,ΔE3,…,ΔEn},

[0108] Where n represents the number of consecutive statistical periods. The energy management unit analyzes the stability and accuracy of the current prediction model based on the historical sequence of prediction errors. When the prediction deviation remains within a small range for multiple consecutive periods, the current prediction model is considered to be operating stably; when the prediction deviation continues to increase for multiple consecutive periods, it is determined that the current prediction model deviates from the actual operating environment, and the prediction model needs to be corrected. In the specific correction process, the historical energy weighting coefficient λ1 and the vibration energy weighting coefficient λ2 in the prediction model are readjusted according to the current period's vibration frequency Fv, vibration amplitude Av, vibration energy input intensity Ie, and actual cumulative energy changes, so that the prediction results for the next period can be closer to the actual operating state, thereby improving the accuracy of the predicted cumulative energy calculation.

[0109] After revising the energy prediction model, the energy management unit performs feedback correction on the revised communication energy consumption based on the difference between the final revised communication energy consumption Efinal and the basic communication energy consumption Ec in the current adaptive correction cycle, as well as the actual communication success rate Rc, the actual number of retransmissions Nr_real, and the target communication distance Dt. The feedback correction updates the revised communication energy consumption for the next adaptive correction cycle based on the actual communication success rate Rc, the actual number of retransmissions Nr_real, and the target communication distance Dt in the current adaptive correction cycle. This ensures that the communication energy consumption model can adapt to changes in the current communication environment, thereby guaranteeing the accuracy of subsequent overall energy consumption threshold calculations.

[0110] Furthermore, the energy management unit calculates the cycle energy utilization rate η based on the actual energy supply capacity and task execution results for the current cycle, which can be expressed as:

[0111] η = Euse / Eactual

[0112] Here, η represents the periodic energy utilization rate. A low energy utilization rate indicates that the current node has sufficient energy reserves but the task execution frequency is too low; a high energy utilization rate indicates that the current node is operating under high load for a long time, posing a risk of insufficient energy. Therefore, the energy utilization rate parameter is used to further guide the adjustment in the next adaptive correction cycle.

[0113] Subsequently, the work cycle control unit determines the length of the next adaptive correction cycle, Tc_next, based on the task execution feedback results of the current adaptive correction cycle. Here, the current adaptive correction cycle length is denoted as Tc, and the next adaptive correction cycle length is denoted as Tc_next. The adaptive correction cycle length represents the time interval between two adjacent task executions of the wireless sensor node; when the adaptive correction cycle length decreases, the time interval between two adjacent task executions shortens, and the node monitoring frequency increases; when the adaptive correction cycle length increases, the time interval between two adjacent task executions lengthens, and the node monitoring frequency decreases.

[0114] The duty cycle control unit compares the difference between the actual accumulated energy (Eactual) of the current adaptive correction cycle and the dynamic wake-up energy threshold (Ewake) corresponding to the current adaptive correction cycle, and adjusts the length of the next adaptive correction cycle based on the difference.

[0115] When Eactual - Ewake > δ, it indicates that there is an energy surplus in the current adaptive correction cycle. At this time, the duty cycle control unit shortens the length of the next adaptive correction cycle to improve monitoring frequency and data real-time performance.

[0116] Tc_next = Tc - ΔT1,

[0117] Where Tc represents the current adaptive correction cycle length, Tc_next represents the next adaptive correction cycle length, and ΔT1 represents the cycle shortening amount, preferably 5% to 20% of the current adaptive correction cycle length Tc.

[0118] When -δ ≤ Eactual - Ewake ≤ δ, it indicates that the actual energy supply and task energy consumption demand of the current adaptive correction cycle are basically balanced. At this time, the work cycle control unit maintains the length of the next adaptive correction cycle unchanged.

[0119] Tc_next = Tc.

[0120] When the condition Eactual - Ewake < -δ is met, it indicates that there is insufficient energy in the current adaptive correction cycle. In this case, the work cycle control unit extends the length of the next adaptive correction cycle to increase the energy accumulation time between adjacent task executions.

[0121] Tc_next = Tc + ΔT2,

[0122] Wherein, ΔT2 represents the period extension amount, which is preferably taken as 10% to 30% of the current adaptive correction period length Tc.

[0123] Wherein, δ is the preset energy balance tolerance, used to avoid frequent adjustments to the length of the next adaptive correction cycle due to small fluctuations between the actual accumulated energy (Eactual) and the dynamic wake-up energy threshold (Ewake). The preset energy balance tolerance δ is taken as 5% to 10% of the total energy consumption threshold (Etotal) corresponding to the current adaptive correction cycle.

[0124] In the above manner, when Eactual-Ewake > δ, the node is determined to be in an energy surplus state, and the length of the next adaptive correction cycle is shortened by the cycle shortening amount ΔT1 to increase the monitoring frequency; when -δ ≤ Eactual-Ewake ≤ δ, the node is determined to be in an energy balance state, and the length of the next adaptive correction cycle remains unchanged; when Eactual-Ewake < -δ, the node is determined to be in an energy shortage state, and the length of the next adaptive correction cycle is extended by the cycle extension amount ΔT2 to accumulate more usable energy for the energy storage unit, thereby ensuring the long-term stable operation of the wireless sensing node.

[0125] After generating the parameters for the next adaptive correction cycle, the duty cycle control unit writes the updated cycle length, operating mode configuration parameters, and threshold parameters into the node's operation configuration area and saves them as the start parameters for the next cycle. When the next cycle begins, the energy state parameters, vibration state parameters, and communication state parameters are re-acquired, analyzed, and used for decision-making, thus forming a complete closed-loop operation mechanism.

[0126] In summary, this embodiment dynamically adjusts the energy prediction model, communication energy consumption model, and adaptive correction cycle length by acquiring feedback information from the current cycle task execution. It also generates parameters for the next adaptive correction cycle based on the matching relationship between the current energy supply capacity and the task energy consumption requirements, thus achieving continuous optimization and adaptive adjustment of the wireless sensor node's working cycle. By constructing a closed-loop control mechanism encompassing state perception, energy prediction, threshold generation, task decision-making, feedback correction, and cycle updates, the wireless sensor node can adapt to long-term vibration energy fluctuations, changes in the communication environment, and changes in equipment operating status. This effectively improves node energy utilization efficiency, task execution success rate, and operational stability, thereby achieving long-term self-powered intelligent monitoring functionality for industrial equipment condition monitoring scenarios.

Claims

1. An adaptive correction method for the duty cycle of wireless sensing nodes based on energy sensing, characterized in that, Includes the following steps: After the wireless sensor node enters the current adaptive correction cycle, it acquires energy state parameters, vibration state parameters, and communication state parameters. Based on the task type of the wireless sensor node, the node's running tasks are divided into temperature measurement tasks, vibration measurement tasks, data processing tasks, wireless communication tasks, and power management tasks, which are respectively used as basic tasks. Obtain the historical energy consumption corresponding to each basic task, determine the energy requirements of temperature measurement task, vibration measurement task, data processing task and power management task, determine the actual accumulated energy of the current cycle based on the energy storage unit voltage, charging current and historical accumulated energy, and determine the predicted accumulated energy of the next adaptive correction cycle based on the vibration state parameters and historical accumulated energy. The communication energy consumption corresponding to the wireless communication task is corrected according to the communication status parameters to obtain the corrected communication energy consumption. The corrected communication energy consumption is then summarized with the energy requirements of the temperature measurement task, vibration measurement task, data processing task, and power management task to generate the overall energy consumption threshold. A dynamic safety margin is determined based on the prediction deviation between the predicted accumulated energy and the actual accumulated energy in the historical adaptive correction cycle. The dynamic safety margin provides safety compensation for the overall energy consumption threshold and generates a dynamic wake-up energy threshold. Based on the predicted cumulative energy, total energy consumption threshold, and dynamic wake-up energy threshold of the current cycle, determine the mode of the wireless sensor node to perform a complete task, sampling buffer task, core monitoring task, or deep sleep task. Based on the actual energy consumption, task completion rate, and communication success rate generated during task execution, adjust the length of the adaptive correction cycle for the next adaptive correction cycle.

2. The adaptive correction method for the duty cycle of a wireless sensing node based on energy sensing as described in claim 1, characterized in that: The specific method for obtaining the energy state parameters, vibration state parameters, and communication state parameters is as follows: The wireless sensing node collects the current voltage value, charging current value, and historical accumulated energy of the energy storage unit through the energy management unit, and uses the current voltage value, charging current value, and historical accumulated energy together as the energy status parameter. The wireless sensing node collects the vibration frequency and amplitude of the bearing housing through the vibration acquisition unit, and uses the vibration frequency and amplitude as vibration state parameters. The wireless sensing node collects radio frequency power density, echo signal-to-noise ratio, historical communication success rate, historical retransmission count, and target communication distance through the communication monitoring unit, as communication status parameters of the current communication link status.

3. The adaptive correction method for the duty cycle of a wireless sensing node based on energy sensing as described in claim 1, characterized in that: The method for predicting the cumulative energy is as follows: The wireless sensing node acquires the vibration state parameters and historical accumulated energy within the current adaptive correction cycle, and calculates the vibration energy input intensity corresponding to the current adaptive correction cycle based on the vibration frequency and vibration amplitude in the vibration state parameters. Obtain the actual new energy records of the historical period corresponding to multiple consecutive historical adaptive correction periods, average the actual new energy records of the historical period to obtain the average new energy of the historical period, and determine the average new energy of the historical period as the historical predicted new energy. The vibration prediction new energy is determined based on the vibration energy input intensity corresponding to the current adaptive correction cycle and the length of the next adaptive correction cycle. The historical prediction new energy and the vibration prediction new energy are weighted and calculated to obtain the prediction new energy corresponding to the next adaptive correction cycle.

4. The adaptive correction method for the duty cycle of a wireless sensing node based on energy sensing as described in claim 1, characterized in that: The step of correcting the communication energy consumption corresponding to the wireless communication task based on communication status parameters includes: The RF power density and echo signal-to-noise ratio are normalized respectively, and the normalized RF power density and echo signal-to-noise ratio are weighted according to preset weights to obtain the current communication link quality parameters. The signal transmission status of the current communication link is determined based on the current communication link quality parameters, and the communication distance correction coefficient is determined in combination with the target communication distance. The basic communication energy consumption is dynamically adjusted based on the communication distance correction coefficient to obtain the corrected communication energy consumption. The wireless sensor node will use the corrected communication energy consumption as the energy consumption requirement of the wireless communication task corresponding to the current adaptive correction cycle to generate the overall energy consumption threshold.

5. The adaptive correction method for the duty cycle of a wireless sensing node based on energy sensing as described in claim 4, characterized in that: The step of correcting the communication energy consumption corresponding to the wireless communication task based on the communication state parameters further includes: The wireless sensor node obtains the historical communication success rate through the communication monitoring unit; The corrected communication energy consumption is compensated and corrected based on the historical communication success rate to obtain the final corrected communication energy consumption result. Among them, the historical communication success rate represents the stability of the current communication link in completing data transmission.

6. The adaptive correction method for the duty cycle of a wireless sensing node based on energy sensing as described in claim 4, characterized in that: The overall energy consumption threshold is obtained through the following process: Acquire the energy requirements for temperature measurement, vibration measurement, data processing, and power management tasks, and correct communication energy consumption. The energy requirements for temperature measurement, vibration measurement, data processing, power management, and corrective communication are summarized and calculated to obtain the overall energy consumption threshold corresponding to the next adaptive correction cycle.

7. The adaptive correction method for the duty cycle of a wireless sensing node based on energy sensing as described in claim 1, characterized in that: The dynamic safety margin is defined in the following process: Obtain the predicted cumulative energy and actual cumulative energy for each historical adaptive correction period; The prediction deviation is calculated based on the difference between the predicted cumulative energy and the actual cumulative energy in the same historical adaptive correction cycle, and the energy prediction error level corresponding to the current adaptive correction cycle is determined based on the prediction deviation. The corresponding dynamic safety margin is determined based on the energy prediction error level; The prediction deviations corresponding to several consecutive historical adaptive correction periods are statistically analyzed. The absolute value of each prediction deviation is calculated, and the absolute value of each prediction deviation is averaged. The resulting average absolute value of prediction deviation is used as the dynamic safety margin.

8. The adaptive correction method for the duty cycle of a wireless sensing node based on energy sensing as described in claim 1, characterized in that: The specific process for the dynamic wake-up energy threshold is as follows: Obtain the overall energy consumption threshold and dynamic safety margin corresponding to the current adaptive correction cycle; The dynamic safety margin serves as the safety compensation amount corresponding to the overall energy consumption threshold. The overall energy consumption threshold and the dynamic safety margin are superimposed to obtain the dynamic wake-up energy threshold.

9. The adaptive correction method for the duty cycle of a wireless sensing node based on energy sensing as described in claim 1, characterized in that: The specific method for determining the execution mode of the wireless sensor node—whether it performs a complete task, a sampling buffer task, a core monitoring task, or a deep sleep task—is as follows: Obtain the predicted cumulative energy and total energy consumption threshold, as well as the dynamic wake-up energy threshold, for the next adaptive correction cycle; The minimum monitoring energy threshold is determined based on the historical energy consumption corresponding to the core monitoring task, and the minimum monitoring energy threshold is less than the overall energy consumption threshold. The predicted cumulative energy is compared with the dynamic wake-up energy threshold, the overall energy consumption threshold, and the minimum monitoring energy threshold, respectively. When the predicted accumulated energy is not lower than the dynamic wake-up energy threshold, control the wireless sensor node to execute the complete task. When the predicted accumulated energy is lower than the dynamic wake-up energy threshold and not lower than the overall energy consumption threshold, control the wireless sensor node to perform a sampling buffer task. When the predicted cumulative energy is lower than the overall energy consumption threshold but not lower than the minimum monitoring energy threshold, control the wireless sensor node to perform the core monitoring task. When the predicted accumulated energy is below the minimum monitoring energy threshold, the wireless sensor node is controlled to enter a deep sleep state.

10. The adaptive correction method for the duty cycle of a wireless sensing node based on energy sensing as described in claim 1, characterized in that: The adaptive correction period length for the next adaptive correction period is adjusted based on the actual energy consumption, task completion rate, and communication success rate during task execution. The specific process is as follows: Obtain task execution feedback information within the current adaptive correction period; The task execution feedback information includes actual energy consumption, task completion rate, and communication success rate. The energy utilization status corresponding to the current adaptive correction cycle is determined based on the difference between the actual energy consumption and the overall energy consumption threshold, and the current task execution effect is determined based on the task completion rate and communication success rate. When the actual energy consumption of the current adaptive correction cycle is lower than the overall energy consumption threshold, it is determined that the current energy supply can meet the task execution requirements, and the length of the next adaptive correction cycle is shortened to shorten the time interval between two adjacent task executions and increase the monitoring frequency. When the actual energy consumption of the current adaptive correction cycle is equal to the total energy consumption threshold, the length of the next adaptive correction cycle remains unchanged. When the actual energy consumption of the current adaptive correction cycle is higher than the overall energy consumption threshold, it is determined that the current energy supply cannot stably meet the task execution requirements, and the length of the next adaptive correction cycle is extended to prolong the time interval between two adjacent task executions. The wireless sensor node generates the adaptive correction period length correction result corresponding to the next adaptive correction period based on the adjusted adaptive correction period length.

Images