A data collection method and system for sensor network

Abstract

The present invention discloses a data acquisition method and system for a sensor network, relating to the technical field of sensor networks. This method calculates by constructing a standard change formula and extracting a standard data set. The present invention can calculate the standard change rate V of the state parameters of each processing device respectively. On this basis, when the calculated standard change rate V i (t) exceeds the set change rate threshold Vth, the system determines that there is an abnormality in the operation of the device and enters the event trigger stage. In this stage, the operating frequency of the sensor module will increase by 50%, so as to collect more abnormal state parameters, such as power fluctuation P, vibration frequency F, and noise spectrum S. By calculating the abnormal eigenvalue Yc for these abnormal state parameters and summarizing them into an abnormal feature set O, the system's ability to identify device abnormalities is further enhanced, potential faults are discovered in a timely manner, thereby providing a basis for the subsequent fault response stage and reducing the risk of faults occurring.

Description

Technical Field

[0001] The present invention relates to the technical field of sensor networks, and in particular to a data collection method and system for sensor networks. Background Art

[0002] With the advent of the Industrial 4.0 era, the rapid development of intelligent manufacturing and Internet of Things technologies has led to the increasing application of sensor networks in industrial monitoring and data collection. Especially in the monitoring of production equipment and processing processes, sensor networks have become an important part of industrial automation systems. By connecting multiple sensor devices, sensor networks can realize real-time monitoring and data transmission of equipment status, which can effectively improve production efficiency and equipment health management. In this context, data collection methods for sensor networks, especially optimization under real-time data collection and event triggering mechanisms, have gradually become the research focus of academia and industry. Specifically, intelligent analysis and equipment status monitoring based on sensor network data collection enable processing equipment to continuously perceive and reflect abnormal conditions in real time during operation, providing an important basis for production maintenance and fault prediction.

[0003] In the Chinese invention application with application number 201610067183.8, a data collection method, gateway and data collection system for a sensor network are disclosed. The sensor network includes multiple sensor nodes. The data collection method includes: comparing the change amount of the data sent by each sensor node received within a preset time period with the corresponding preset value; judging whether there is data whose change amount reaches the corresponding preset value; and increasing the collection frequency of the sensor node corresponding to the data whose change amount reaches the corresponding preset value to the first collection frequency. The present invention uses the same gateway to control the collection frequency of different sensor nodes, so there will be no asynchronous upload of data, which is beneficial to the background server for data collection.

[0004] Analytical processing.

[0005] In combination with the prior art, the above application still has the following deficiencies:

[0006] At present, there are still some problems that cannot be ignored in the existing sensor network data collection methods. First, the traditional data collection method usually sets the sensor collection frequency to a fixed value, and lacks a dynamic adjustment mechanism for changes in the health status of the equipment. This may lead to the failure to increase the collection frequency in time when the equipment is abnormal, and the inability to fully capture the key changes in the operation of the equipment. Secondly, most of the existing monitoring systems use a "full data"-based processing method, which not only increases the burden of data transmission and storage, but also makes the system response slow, making it difficult to cope with complex and dynamically changing working environments. Finally, although some systems have introduced fault warning mechanisms, due to excessive reliance on equipment health status assessment standards and threshold settings, it may lead to delayed responses or misjudgments to some potential faults, affecting the accuracy and timeliness of equipment maintenance. Summary of the invention

[0007] In view of the deficiencies of the prior art, the present invention provides a data collection method and system for a sensor network, which solves the problems mentioned in the background technology.

[0008] To achieve the above objectives, the present invention is implemented by the following technical scheme: comprising the following steps:

[0009] S1. Set the sensor module to the lower limit acquisition mode of the operating frequency by default through the sensor network, record the state parameters of the processing equipment in real time, and build an embedded system at the same time. Transmit the state parameters of the processing equipment to the embedded system through the wireless communication network, and pre-process the state parameters in the embedded system to obtain a standard data set;

[0010] S2. Construct a standard change formula in the embedded system, extract the state parameters and input them into the standard change formula, calculate the standard change trend of the state parameters of each processing equipment respectively, obtain the standard change rate V of the parameter change, set the change threshold Vth and the standard change rate V of the parameter change for preliminary comparative evaluation, and enter the event triggering stage based on the preliminary comparative evaluation results;

[0011] S3. When any parameter change is found to be abnormal in the preliminary comparison and evaluation, the event triggering stage is entered. In the event triggering stage, the operating frequency of the sensor module is increased by 50% through the sensor network, the abnormal state parameters of the processing equipment are recorded, and the abnormal characteristic value Yc of each abnormal state parameter of the processing equipment is calculated and output based on the abnormal state parameters, and the abnormal characteristic value Yc is summarized to obtain the abnormal characteristic collection O;

[0012] S4, based on the acquired abnormal feature set O, perform summary calculation and comprehensive calculation to output the health status index H, and at the same time set the health threshold Hth and the health status index H to perform health assessment, and execute fault response measures based on the assessment results;

[0013] S5. After the fault response phase is completed, a response effectiveness mechanism is executed. The response effectiveness mechanism outputs a response effectiveness index E by comprehensively calculating the second health status indicator Hnew output after the fault response phase is completed and the health status indicator, and determines the effectiveness of the fault response measures based on the output result of the response effectiveness index E.

[0014] Preferably, said S1 includes S11 and S12;

[0015] S11, the sensor network includes a sensor module and a communication module, by installing the sensor module on the processing equipment, and using the sensor network to set the collection frequency of the sensor module to the lower limit collection mode of the operating frequency by default, and record the state parameters of the processing equipment at each frequency time node in real time;

[0016] The sensor module includes a power sensor, a vibration sensor, a temperature sensor and a noise sensor;

[0017] The state parameters include vibration amplitude A, temperature rise rate △T and noise increment △N;

[0018] S12, constructing an embedded system, wherein the embedded system includes a memory, an embedded operating system and a CPU processor;

[0019] Use the 5G communication network to connect the embedded system to the communication module of the sensor network, and transmit the real-time recorded state parameters of the processing equipment to the memory of the embedded system for data storage;

[0020] Then, the CPU processor extracts the state parameters of the processing equipment in the memory and performs preprocessing, wherein the preprocessing includes timestamp alignment and normalization processing to eliminate the time difference and dimension influence of the state parameters of the processing equipment and obtain a standard data set;

[0021] The standard data set includes the vibration amplitude A(t) at time t, the temperature rise rate ΔT(t) at time t, and the noise increment ΔN(t) at time t.

[0022] Preferably, S2 includes S21 and S22;

[0023] S21, constructing a standard change formula, extracting a standard data set and inputting it into the standard change formula, respectively calculating the standard change trend of the state parameter of each processing equipment, and obtaining the standard change rate V of the parameter change;

[0024] The standard change rate V is calculated and outputted by the following standard change formula;

[0025] ;

[0026] Where V i (t) represents the standard change rate V of the parameters in the i-th standard data set at time t, x i (t) represents the parameter of the i-th standard data set at time t, and △t represents the time interval;

[0027] S22, according to the abnormal upper limit values ​​of vibration, temperature rise and noise of the processing equipment, set the change rate threshold value Vth of the parameters in each standard data set respectively, and obtain the standard change rate Vth of the parameters in the i-th standard data set at time t i (t) respectively making preliminary comparative evaluation with the corresponding change rate threshold value Vth, analyzing the abnormal parameters of the processing equipment, and determining the execution of the event triggering stage based on the preliminary comparative evaluation results. The specific evaluation contents are as follows;

[0028] At time t, the standard change rate V of the parameters in the i-th standard data set i (t)>threshold value Vth of the parameter of the i-th standard data set i When the processing equipment is judged to be operating abnormally, the event triggering stage is executed;

[0029] At time t, the standard change rate V of the parameters in the i-th standard data set i (t) ≤ the change threshold Vth of the parameter of the i-th standard data set i , it is determined that the processing equipment is operating normally, and the current operating frequency lower limit acquisition mode is maintained.

[0030] Preferably, said S3 includes S31, S32 and S33;

[0031] S31, after the processing equipment is scheduled to run abnormally, the event triggering stage is executed, in which the operating frequency of the current sensor module is increased by 50% through the sensor network, and the sensor module is controlled to collect abnormal state parameters;

[0032] The abnormal state parameters include the power fluctuation P(t) at time t, the vibration frequency F(t) at time t, and the noise spectrum S(t) at time t;

[0033] The power fluctuation P(t) at time t is obtained by a power sensor, the vibration frequency F(t) at time t is obtained by a vibration sensor, and the noise spectrum S(t) at time t is obtained by a noise sensor;

[0034] Based on the acquired abnormal state parameters, the abnormal feature value Yc of each parameter in the abnormal state parameters is calculated and output respectively, and then the abnormal feature value Yc of each parameter is summarized into an abnormal feature set O;

[0035] The abnormal feature set O includes the abnormal feature value Yc of the power fluctuation P(t) at time t P（t） , the abnormal characteristic value Yc of the vibration frequency F(t) at time t F（t） and the abnormal characteristic value Yc of the noise spectrum S(t) at time t S（t）；

[0036] The abnormal characteristic value Yc of the power fluctuation P(t) at time t P（t） The output is calculated by the following algorithm formula;

[0037] ;

[0038] In the formula, Pmin represents the lower power limit of the processing equipment, and Pmax represents the upper power limit of the processing equipment. Indicates the power sensitivity adjustment coefficient, Pref indicates the power reference value, represents the first weight value, and ln represents the natural logarithm function.

[0039] Preferably, S32, the abnormal characteristic value Yc of the vibration frequency F(t) at time t F（t） The output is calculated by the following algorithm formula;

[0040] ;

[0041] In the formula, Fnom represents the standard vibration frequency of the processing equipment, and Fmax represents the upper limit of the vibration frequency of the processing equipment.

[0042] Preferably, S33, the abnormal characteristic value Yc of the noise spectrum S(t) at time t S（t） The output is calculated by the following algorithm formula;

[0043] ;

[0044] In the formula, Sstd represents the standard deviation of the noise of the processing equipment, which indicates the degree of noise fluctuation, Smax represents the upper limit of the noise spectrum of the processing equipment, and exp represents the exponential function. It represents the sensitivity adjustment coefficient of noise change, and is used to adjust the sensitivity of noise change of processing equipment.

[0045] Preferably, said S4 includes S41 and S42;

[0046] S41, based on all the parameters in the acquired abnormal feature set O, perform weighted summary calculation, output the health status index H of the processing equipment, and quantitatively analyze the health status of the processing equipment;

[0047] The health status indicator H is calculated and output by the following algorithm formula;

[0048] ;

[0049] Where H(t) represents the health status indicator at time t, n represents the total number of parameters in the abnormal feature set O, and W j represents the weight value of the jth parameter in the abnormal feature set O, j (t) represents the jth parameter in the abnormal feature set O at time t.

[0050] Preferably, S42, based on the overall health index of the processing equipment, the user sets a health threshold Hth, and then performs a health assessment on the health threshold Hth and the health status index H(t) at time t, analyzes the overall health status of the processing equipment, and triggers the fault response measures of the sensor network based on the health assessment result. The specific assessment content is as follows;

[0051] When the health status indicator H(t) at time t is less than the health threshold Hth, it means that the overall processing equipment monitored by the current sensor network is within the normal range, and it is automatically converted to the lower limit collection mode of the operating frequency;

[0052] When the health status indicator H(t) at time t ≥ the health threshold Hth, it means that the overall processing equipment monitored by the current sensor network is in an abnormal state. At this time, the first response information is sent through the embedded system to prompt the operator to check the processing equipment due to the abnormal equipment, and at the same time, the first response measure is executed. The first response measure increases the current operating frequency to 100% through the sensor network;

[0053] When the health status indicator H(t) at time t is greater than twice the health threshold Hth, it means that the overall processing equipment monitored by the current sensor network is in a risky state. At this time, the second response information is sent out through the embedded system to prompt the operator to perform maintenance and automatically execute the second response measure. The second response measure is to shut down the sensor network and record the complete status parameters before the shutdown.

[0054] Preferably, said S5 includes S51 and S52;

[0055] S51, after the second response measure is executed, restart the embedded system and the sensor network, output the second health status indicator Hnew after the fault response measure is executed, and calculate the difference between the second health status indicator Hnew and the health status indicator H(t) at time t to obtain the response effectiveness indicator E to analyze the effectiveness of the fault response mechanism;

[0056] The response effectiveness index E(t) is calculated and outputted by the following algorithm formula;

[0057] ;

[0058] In the formula, E(t) represents the effectiveness index at time t, Hnew(t) represents the second health status index at time t, and △t represents the time interval;

[0059] S52, based on the output result of the effectiveness index E(t) obtained at time t, the effectiveness of the fault response measures is determined, and the specific determination content is as follows;

[0060] When the effectiveness index E(t) at time t is ≥ 0, it means that the fault response measures are effective, and the sensor network is set to the lower limit acquisition mode of the operating frequency;

[0061] When the effectiveness indicator E(t) at time t is less than 0, it indicates that the fault response measure is invalid, and the second response measure is executed for the second time.

[0062] A data acquisition system for a sensor network, comprising a sensor network module, a state change analysis module, an event trigger module, a fault response module and a validity analysis module;

[0063] The sensor network module sets the sensor module to the lower limit acquisition mode of the operating frequency by default through the sensor network, records the state parameters of the processing equipment in real time, and constructs an embedded system at the same time, transmits the state parameters of the processing equipment to the embedded system through the wireless communication network, and pre-processes the state parameters in the embedded system to obtain a standard data set;

[0064] The state change analysis module constructs a standard change formula in the embedded system, extracts the state parameters and inputs them into the standard change formula, calculates the standard change trend of the state parameters of each processing equipment, obtains the standard change rate V of the parameter change, sets the change threshold Vth and the standard change rate V of the parameter change for preliminary comparison and evaluation, and enters the event triggering stage based on the preliminary comparison and evaluation results;

[0065] When the event trigger module finds that any parameter change is abnormal through preliminary comparison and evaluation, it enters the event trigger stage. In the event trigger stage, the operating frequency of the sensor module is increased by 50% through the sensor network, the abnormal state parameters of the processing equipment are recorded, and the abnormal characteristic value Yc of each abnormal state parameter of the processing equipment is calculated and output based on the abnormal state parameters, and the abnormal characteristic value Yc is summarized to obtain the abnormal feature collection O;

[0066] The fault response module performs a summary calculation based on the acquired abnormal feature set O to output a health status indicator H, sets a health threshold Hth and the health status indicator H to perform a health assessment, and executes a fault response measure based on the assessment result;

[0067] The effectiveness analysis module executes the response effectiveness mechanism after the fault response phase is completed. The response effectiveness mechanism outputs the response effectiveness index E by comprehensively calculating the second health status indicator Hnew output after the fault response phase is completed and the health status indicator, and determines the effectiveness of the fault response measures based on the output result of the response effectiveness index E.

[0068] The present invention provides a data collection method and system for a sensor network, which has the following beneficial effects:

[0069] (1) This method sets the acquisition frequency of the sensor module to the lower limit acquisition mode of the operating frequency by default through the sensor network, which can effectively reduce unnecessary data redundancy of the equipment and improve the efficiency of data acquisition. Based on the state parameters of the processing equipment collected by the sensor module, such as vibration amplitude A, temperature rise rate △T and noise increment △N, the embedded system performs preprocessing, including timestamp alignment and normalization, to eliminate the time difference and dimension influence of the state parameters of the processing equipment, and finally obtain a standard data set. This processing method makes the data consistent at different times, providing a reliable input for the subsequent calculation of the standard change formula, thereby ensuring the accuracy and stability of data analysis.

[0070] (2) This method constructs a standard change formula and extracts a standard data set for calculation. The present invention can calculate the standard change rate V of each processing equipment state parameter. On this basis, when the calculated standard change rate V i When (t) exceeds the set change rate threshold Vth, the system determines that the equipment is operating abnormally and enters the event triggering stage. In this stage, the operating frequency of the sensor module will increase by 50%, thereby collecting more abnormal state parameters, such as power fluctuation P, vibration frequency F, and noise spectrum S. By calculating the abnormal characteristic value Yc of these abnormal state parameters and summarizing them into an abnormal characteristic collection O, the system's ability to identify equipment abnormalities is further enhanced, potential faults are discovered in a timely manner, thereby providing a basis for the subsequent fault response stage and reducing the risk of faults.

[0071] (3) This method evaluates the overall health status of the processing equipment by comprehensively calculating the health status index H and combining it with the health threshold Hth. When the health status index H(t) exceeds the health threshold Hth, the system triggers the fault response measure, automatically prompts the operator to check the equipment, and increases the acquisition frequency of the sensor module to 100%, thereby ensuring that the equipment abnormality is identified in time. In extreme cases, when the health status index H(t) exceeds twice the health threshold Hth, the system enters a risk state and automatically executes the second response measure to ensure equipment safety through shutdown processing. This fault response mechanism greatly improves the fault detection capability of the equipment, ensures that the equipment can be handled in a timely and effective manner when problems occur, and reduces production stoppages and losses caused by equipment failures. Furthermore, after the fault response measures are executed, the second health status index Hnew is calculated, and the difference between it and the health status index H(t) is calculated to output the response effectiveness index E. This mechanism can quantitatively evaluate the effect of fault response. If the response effectiveness index E(t) at time t is ≥ 0, it means that the fault response measures are effective and the equipment has returned to normal operation. If the response effectiveness index E(t) at time t is < 0, it means that the fault response measures are invalid and the corresponding response measures need to be executed again. This adaptive mechanism can ensure that the health status of the equipment is always maintained at the optimal level, further improving the stability and reliability of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 A schematic diagram of the steps of a data collection method for a sensor network according to the present invention;

[0073] Figure 2 The present invention is a flow chart of a data acquisition system for a sensor network. DETAILED DESCRIPTION

[0074] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0075] Example 1

[0076] See also Figure 1 The present invention provides a data collection method for a sensor network. To achieve the above purpose, the present invention is implemented by the following technical solution: comprising the following steps:

[0077] S1. Set the sensor module to the lower limit acquisition mode of the operating frequency by default through the sensor network, record the state parameters of the processing equipment in real time, and build an embedded system at the same time. Transmit the state parameters of the processing equipment to the embedded system through the wireless communication network, and pre-process the state parameters in the embedded system to obtain a standard data set;

[0078] S2. Construct a standard change formula in the embedded system, extract the state parameters and input them into the standard change formula, calculate the standard change trend of the state parameters of each processing equipment respectively, obtain the standard change rate V of the parameter change, set the change threshold Vth and the standard change rate V of the parameter change for preliminary comparative evaluation, and enter the event triggering stage based on the preliminary comparative evaluation results;

[0079] S3. When any parameter change is found to be abnormal in the preliminary comparison and evaluation, the event triggering stage is entered. In the event triggering stage, the operating frequency of the sensor module is increased by 50% through the sensor network, the abnormal state parameters of the processing equipment are recorded, and the abnormal characteristic value Yc of each abnormal state parameter of the processing equipment is calculated and output based on the abnormal state parameters, and the abnormal characteristic value Yc is summarized to obtain the abnormal characteristic collection O;

[0080] S4, based on the acquired abnormal feature set O, perform summary calculation and comprehensive calculation to output the health status index H, and at the same time set the health threshold Hth and the health status index H to perform health assessment, and execute fault response measures based on the assessment results;

[0081] S5. After the fault response phase is completed, the response effectiveness mechanism is executed. The response effectiveness mechanism outputs the response effectiveness index E by comprehensively calculating the second health status indicator Hnew output after the fault response phase is completed and the health status indicator, and determines the effectiveness of the fault response measures based on the output result of the response effectiveness index E.

[0082] In this embodiment, the method records the state parameters of the processing equipment in real time by setting the acquisition frequency of the sensor module to the lower limit acquisition mode of the operating frequency in the sensor network. These state parameters are transmitted to the embedded system through the wireless communication network, and are preprocessed in the embedded system to obtain a standard data set. Then, in the embedded system, based on the standard change formula, the standard change trend of the state parameters of each processing equipment is calculated to obtain the standard change rate V of the parameter change. By comparing and evaluating with the set change threshold Vth, if any parameter change is found to be abnormal, the event triggering stage is entered. At this stage, the operating frequency of the sensor module is increased by 50% through the sensor network, and the abnormal feature vector is collected, the abnormal feature value Yc of the abnormal state parameter is calculated, and the abnormal feature collection O is summarized.

[0083] Based on the abnormal feature set O, the system further summarizes, calculates and outputs the health status indicator H, and compares and evaluates it with the preset health threshold Hth. According to the health evaluation results, fault response measures are executed to ensure the normal operation of the equipment. When the fault response is completed, the system calculates the second health status indicator Hnew through the response effectiveness mechanism, and performs difference calculation with the health status indicator H to generate the response effectiveness indicator E, which is used to determine the effectiveness of the fault response measures. Through this method, the state parameter changes of the processing equipment can be monitored in real time, and the response mechanism can be quickly triggered when an abnormality occurs. By increasing the acquisition frequency of the sensor module, more abnormal feature data can be obtained, thereby improving the accuracy and timeliness of fault diagnosis. Through the dynamic calculation of the health status indicator and the comparative evaluation of the health threshold, it is possible to accurately identify whether the equipment is in a normal or abnormal state, and take corresponding fault response measures to ensure the stable operation of the processing equipment. The addition of the response effectiveness mechanism further verifies the effect of the fault response measures, ensures that the health management of the equipment is more accurate and efficient, and dynamically adjusts the acquisition frequency of the sensor network, effectively reducing the problem of excessive energy consumption of the sensor module during large-scale acquisition.

[0084] Example 2

[0085] Specifically: S1 includes S11 and S12;

[0086] S11, the sensor network includes a sensor module and a communication module. By installing the sensor module on the processing equipment, and using the sensor network to set the collection frequency of the sensor module to the lower limit collection mode of the operating frequency by default, the state parameters of the processing equipment at each frequency time node are recorded in real time;

[0087] The sensor module includes power sensor, vibration sensor, temperature sensor and noise sensor;

[0088] The state parameters include vibration amplitude A, temperature rise rate △T and noise increment △N;

[0089] S12. Build an embedded system, which includes a memory, an embedded operating system and a CPU processor;

[0090] Use the 5G communication network to connect the embedded system to the communication module of the sensor network, and transmit the real-time recorded state parameters of the processing equipment to the memory of the embedded system for data storage;

[0091] Then, the CPU processor extracts the state parameters of the processing equipment in the memory and performs preprocessing, which includes timestamp alignment and normalization processing to eliminate the time difference and dimension influence of the state parameters of the processing equipment and obtain a standard data set;

[0092] The standard data set includes the vibration amplitude A(t) at time t, the temperature rise rate △T(t) at time t, and the noise increment △N(t) at time t.

[0093] In this embodiment, the method monitors the state parameters of the processing equipment in real time based on the sensor modules in the sensor network. The acquisition frequency of these sensor modules is set to the lower limit acquisition mode of the operating frequency by default, which can ensure continuous monitoring of the equipment status at low energy consumption. The monitored state parameters include key indicators such as vibration amplitude A, temperature rise rate △T and noise increment △N, which are used to reflect the operating status of the equipment. Secondly, the embedded system is connected to the sensor network through the 5G communication network, and the real-time recorded equipment state parameters are transmitted to the system's memory for data storage. The CPU processor in the embedded system extracts these data and preprocesses them, including timestamp alignment and normalization, to ensure that the parameters collected at different time points are consistent and eliminate time differences and dimensional effects. The preprocessed data forms a standard data set. The implementation of this step can not only monitor the state parameters of the processing equipment in real time and comprehensively, but also ensure high-precision acquisition and analysis of the equipment status through data preprocessing and standardization. In the low-energy acquisition mode, the status data of the equipment can be continuously recorded, reducing the excessive consumption of system resources, and providing accurate and comparable basic data for subsequent health assessment and fault warning.

[0094] Example 3

[0095] Specifically: S2 includes S21 and S22;

[0096] S21, constructing a standard change formula, extracting a standard data set and inputting it into the standard change formula, respectively calculating the standard change trend of the state parameter of each processing equipment, and obtaining the standard change rate V of the parameter change;

[0097] The standard rate of change V is calculated using the following standard change formula:

[0098] ;

[0099] Where V i (t) represents the standard change rate V of the parameters in the i-th standard data set at time t, x i (t) represents the parameter of the i-th standard data set at time t, △t represents the time interval,

[0100] Among them, x 1 (t) = A(t), x 2 (t) = △T(t), x 3 (t) = △N(t);

[0101] S22, according to the abnormal upper limit values ​​of vibration, temperature rise and noise of the processing equipment, set the change rate threshold value Vth of the parameters in each standard data set respectively, and obtain the standard change rate Vth of the parameters in the i-th standard data set at time t i (t) respectively making preliminary comparative evaluation with the corresponding change rate threshold value Vth, analyzing the abnormal parameters of the processing equipment, and determining the execution of the event triggering stage based on the preliminary comparative evaluation results. The specific evaluation contents are as follows;

[0102] At time t, the standard change rate V of the parameters in the i-th standard data set i (t)>threshold value Vth of the parameter of the i-th standard data set i When , it indicates that the vibration, temperature rise or noise data of the current equipment is abnormal, and the processing equipment is judged to be operating abnormally. At this time, the event triggering stage is executed;

[0103] At time t, the standard change rate V of the parameters in the i-th standard data set i (t) ≤ the change threshold Vth of the parameter of the i-th standard data set i , it indicates that the vibration, temperature rise or noise data of the current equipment is normal, and it is determined that the processing equipment is operating normally. At this time, the current operating frequency lower limit collection mode is maintained.

[0104] In this embodiment, the method calculates the change trend of the device state parameters collected by the sensor through the standard change formula to obtain the standard change rate V of each parameter. This process ensures the accurate capture of the dynamic operation of the equipment, so that the subtle changes in the equipment state can be monitored in real time. Subsequently, based on the abnormal upper limit values ​​of key parameters such as vibration, temperature rise and noise, a corresponding change rate threshold Vth is set for each state parameter. When the standard change rate V of the device state parameter is i (t) When the change rate threshold Vth is exceeded, the device is judged to be operating abnormally and an event response is triggered; when the standard change rate Vth of the device status parameter is exceeded, the device is judged to be operating abnormally and an event response is triggered; i (t) When the change rate threshold Vth is not exceeded, the equipment is judged to be operating normally and the existing acquisition mode is maintained. This method of setting and comparing the change rate threshold Vth effectively avoids misjudgment caused by parameter fluctuations and ensures the accuracy and real-time nature of abnormal identification. Through this standard change analysis and threshold judgment mechanism, this method can timely capture potential abnormal changes in processing equipment, greatly improving the sensitivity and accuracy of equipment status monitoring.

[0105] Example 4

[0106] Specifically: S3 includes S31, S32 and S33;

[0107] S31, after the processing equipment is scheduled to run abnormally, the event triggering stage is executed. In the event triggering stage, the operating frequency of the current sensor module is increased by 50% through the sensor network, and the sensor module is controlled to collect abnormal state parameters;

[0108] The abnormal state parameters include the power fluctuation P(t) at time t, the vibration frequency F(t) at time t, and the noise spectrum S(t) at time t;

[0109] The power fluctuation P(t) at time t is obtained by a power sensor, the vibration frequency F(t) at time t is obtained by a vibration sensor, and the noise spectrum S(t) at time t is obtained by a noise sensor;

[0110] Based on the acquired abnormal state parameters, the abnormal characteristic value Yc of each parameter in the abnormal state parameters is calculated and output respectively, and then the abnormal characteristic value Yc of each parameter is summarized into an abnormal characteristic set O;

[0111] The abnormal feature set O includes the abnormal feature value Yc of the power fluctuation P(t) at time t P（t） , the abnormal characteristic value Yc of the vibration frequency F(t) at time t F（t） and the abnormal characteristic value Yc of the noise spectrum S(t) at time t S（t）；

[0112] Abnormal characteristic value Yc of power fluctuation P(t) at time t P（t） The output is calculated by the following algorithm formula;

[0113] ;

[0114] In the formula, Pmin represents the lower power limit of the processing equipment, and Pmax represents the upper power limit of the processing equipment. Represents the power sensitivity adjustment coefficient, which is used to introduce sensitivity to changes in device power and is usually adjusted according to the device type. Pref represents the power reference value, which indicates the power level of the device under normal operating conditions. represents the first weight value, and ln represents the natural logarithm function.

[0115] S32, abnormal characteristic value Yc of the vibration frequency F(t) at time t F（t） The output is calculated by the following algorithm formula;

[0116] ;

[0117] In the formula, Fnom represents the standard vibration frequency of the processing equipment, and Fmax represents the upper limit of the vibration frequency of the processing equipment.

[0118] S33, abnormal characteristic value Yc of the noise spectrum S(t) at time tS（t） The output is calculated by the following algorithm formula;

[0119] ;

[0120] In the formula, Sstd represents the standard deviation of the noise of the processing equipment, which indicates the degree of noise fluctuation, Smax represents the upper limit of the noise spectrum of the processing equipment, and exp represents the exponential function. It represents the sensitivity adjustment coefficient of noise change, and is used to adjust the sensitivity of noise change of processing equipment.

[0121] In this embodiment, when the abnormal operation of the processing equipment is initially determined, the method actively increases the operating frequency of the sensor module by 50% through the event triggering stage, so as to collect and monitor the abnormal state parameters of the equipment at a higher frequency, including power fluctuation P (t), vibration frequency F (t) and noise spectrum S (t). By further collecting these abnormal state parameters, the abnormal characteristic value Yc of each parameter is calculated according to different types of sensor data. The calculation of these abnormal characteristic values ​​is based on the difference between the actual operating value of the equipment and its standard range. This series of abnormal detection and characteristic value calculation methods effectively improves the sensitivity and accuracy of abnormal states. By aggregating each abnormal characteristic value Yc to form an abnormal characteristic collection O, the equipment status can be comprehensively evaluated and potential faults can be accurately identified. This mechanism can not only detect and respond to the abnormal state of the equipment in a timely manner, but also provide accurate fault diagnosis information, helping operators to quickly take effective treatment measures, thereby improving the maintenance efficiency and stability of the equipment.

[0122] Example 5

[0123] Specifically: S4 includes S41 and S42;

[0124] S41, based on all the parameters in the acquired abnormal feature set O, perform weighted summary calculation, output the health status index H of the processing equipment, and quantitatively analyze the health status of the processing equipment;

[0125] The health status indicator H is calculated and output by the following algorithm formula;

[0126] ;

[0127] Where H(t) represents the health status index at time t, n represents the total number of parameters in the abnormal feature set O, where n=3, and W j represents the weight value of the jth parameter in the abnormal feature set O, j (t) represents the jth parameter in the abnormal feature set O at time t.

[0128] S42, based on the overall health index of the processing equipment, the user sets a health threshold Hth, and then performs a health assessment on the health threshold Hth and the health status index H(t) at time t, analyzes the overall health status of the processing equipment, and triggers the fault response measures of the sensor network based on the health assessment results. The specific assessment contents are as follows;

[0129] When the health status indicator H(t) at time t is less than the health threshold Hth, it means that the overall processing equipment monitored by the current sensor network is within the normal range, and it is automatically converted to the lower limit collection mode of the operating frequency;

[0130] When the health status indicator H(t) at time t ≥ the health threshold Hth, it means that the overall processing equipment monitored by the current sensor network is in an abnormal state. At this time, the first response information is sent through the embedded system to prompt the operator to check the processing equipment due to the abnormal equipment, and the first response measures are executed at the same time. The first response measures increase the current operating frequency to 100% through the sensor network;

[0131] When the health status indicator H(t) at time t is greater than twice the health threshold Hth, it means that the overall processing equipment monitored by the current sensor network is in a risky state. At this time, the second response information is sent out through the embedded system to prompt the operator to perform maintenance and automatically execute the second response measure. The second response measure is to shut down the sensor network and record the complete status parameters before the shutdown.

[0132] In this embodiment, the method effectively enhances the intelligent maintenance capability of the processing equipment by introducing the health status assessment and fault response mechanism. In the specific implementation process, the health status index H is calculated by weighted aggregation based on all the parameters in the abnormal feature set O previously collected, and the overall health status of the processing equipment is quantified in real time. The health status index H combines key parameters such as power fluctuation, vibration frequency and noise spectrum of the equipment, and each parameter is assigned a different weight value W according to its importance in fault detection. j. This weighted calculation method ensures that the contribution of each parameter to the overall health status can be accurately reflected. On this basis, a dynamic health assessment is performed by setting the health threshold Hth. When the health status indicator H(t) is lower than the set health threshold Hth, the equipment is considered to be operating normally and the low-frequency acquisition mode is maintained; when the health status indicator H(t) reaches or exceeds the health threshold Hth, the processing equipment is judged to be abnormal and the fault response measures are triggered, the operator is notified to check and the acquisition frequency is automatically increased to further confirm the abnormal condition. If the health status indicator H(t) exceeds twice the health threshold, the equipment is further judged to be in a risky state, and the equipment is automatically shut down and the complete status parameters are recorded to provide accurate data for equipment maintenance. Through this multi-level evaluation and response mechanism based on health status, the equipment operation status can be monitored in real time, and targeted fault response measures can be taken according to different health levels. This not only improves the ability to detect faults early, but also enables rapid shutdown and data recording measures when equipment risks increase, minimizing equipment damage and downtime.

[0133] Example 6

[0134] Specifically: S5 includes S51 and S52;

[0135] S51, after the second response measure is executed, restart the embedded system and the sensor network, output the second health status indicator Hnew after the fault response measure is executed, and calculate the difference between the second health status indicator Hnew and the health status indicator H(t) at time t to obtain the response effectiveness indicator E to analyze the effectiveness of the fault response mechanism;

[0136] The response effectiveness index E(t) is calculated and output by the following algorithm formula;

[0137] ;

[0138] In the formula, E(t) represents the effectiveness index at time t, Hnew(t) represents the second health status index at time t, and △t represents the time interval;

[0139] S52, based on the output result of the effectiveness index E(t) obtained at time t, the effectiveness of the fault response measures is determined, and the specific determination content is as follows;

[0140] When the effectiveness index E(t) at time t is ≥ 0, it means that the fault response measures are effective, and the sensor network is set to the lower limit acquisition mode of the operating frequency;

[0141] When the effectiveness indicator E(t) at time t is less than 0, it indicates that the fault response measure is invalid, and the second response measure is executed for the second time.

[0142] In this embodiment, the method is further strengthened and verified by the effectiveness of the fault response measures. After the second response measures are executed, the embedded system and the sensor network are restarted, and the second health status indicator Hnew is calculated and output, so that the actual effect of the fault response measures can be evaluated. The response effectiveness index E(t) is obtained by difference calculation, which reflects the degree of change of the health status before and after the fault response measures. If the effectiveness index E(t) is greater than or equal to 0, it indicates that the fault response measures are effective and the equipment status has been restored. At this time, the system resets the sensor network to the lower limit acquisition mode of the operating frequency and resumes normal monitoring. If the effectiveness index E(t) is less than 0, it indicates that the fault response measures are invalid, and the second response measures will be executed twice to ensure the recovery of the health status of the equipment. Through this process, not only the automation and precision of the fault response can be realized, but also the fault handling strategy can be further optimized through effectiveness judgment. The introduction of this mechanism improves the flexibility and adaptability in the fault handling process, making equipment maintenance more intelligent and refined. In addition, the continuous feedback based on the health status and effectiveness evaluation also enhances the real-time monitoring and self-adjustment capabilities of the equipment status, thereby effectively improving the stability and production efficiency of the equipment operation.

[0143] Example 7

[0144] See also Figure 2 , a data acquisition system for a sensor network, comprising a sensor network module, a state change analysis module, an event trigger module, a fault response module and a validity analysis module;

[0145] The sensor network module sets the sensor module to the lower limit acquisition mode of the operating frequency by default through the sensor network, records the state parameters of the processing equipment in real time, and builds an embedded system at the same time. The state parameters of the processing equipment are transmitted to the embedded system through the wireless communication network, and the state parameters are pre-processed in the embedded system to obtain a standard data set;

[0146] The state change analysis module constructs a standard change formula in the embedded system, extracts the state parameters and inputs them into the standard change formula, calculates the standard change trend of the state parameters of each processing equipment, obtains the standard change rate V of the parameter change, and then sets the change threshold Vth to perform a preliminary comparative evaluation with the standard change rate V of the parameter change, and enters the event triggering stage based on the preliminary comparative evaluation results;

[0147] When the event trigger module finds any parameter change abnormally through preliminary comparison and evaluation, it enters the event trigger stage. In the event trigger stage, the sensor module's operating frequency is increased by 50% through the sensor network, the abnormal state parameters of the processing equipment are recorded, and the abnormal characteristic value Yc of each abnormal state parameter of the processing equipment is calculated and output based on the abnormal state parameters, and the abnormal characteristic value Yc is summarized to obtain the abnormal feature set O;

[0148] The fault response module performs a summary calculation based on the acquired abnormal feature set O to output the health status indicator H, sets the health threshold Hth and the health status indicator H for health assessment, and executes fault response measures based on the assessment results;

[0149] The effectiveness analysis module executes the response effectiveness mechanism after the fault response phase is completed. The response effectiveness mechanism outputs the response effectiveness index E by comprehensively calculating the second health status indicator Hnew output after the fault response phase is completed and the health status indicator, and determines the effectiveness of the fault response measures based on the output result of the response effectiveness index E.

[0150] While the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that many changes, modifications, substitutions and variations can be made to the embodiments without departing from the principles and spirit of the invention.

Claims

1. A data collection method for a sensor network, characterized in that: The following steps are involved: S1. Set the sensor module to the lower limit acquisition mode of the operating frequency by default through the sensor network, record the state parameters of the processing equipment in real time, and build an embedded system at the same time. Transmit the state parameters of the processing equipment to the embedded system through the wireless communication network, and pre-process the state parameters in the embedded system to obtain a standard data set; S2. Construct a standard change formula in the embedded system, extract the state parameters and input them into the standard change formula, calculate the standard change trend of the state parameters of each processing equipment respectively, obtain the standard change rate V of the parameter change, set the change threshold Vth and the standard change rate V of the parameter change for preliminary comparative evaluation, and enter the event triggering stage based on the preliminary comparative evaluation results; S3. When any parameter change is found to be abnormal in the preliminary comparison and evaluation, the event triggering stage is entered. In the event triggering stage, the operating frequency of the sensor module is increased by 50% through the sensor network, the abnormal state parameters of the processing equipment are recorded, and the abnormal characteristic value Yc of each abnormal state parameter of the processing equipment is calculated and output based on the abnormal state parameters, and the abnormal characteristic value Yc is summarized to obtain the abnormal characteristic collection O; The S3 includes S31, S32 and S33; S31, after the processing equipment is scheduled to run abnormally, the event triggering stage is executed, wherein the event triggering stage controls the operating frequency of the current sensor module to increase by 50% through the sensor network, and controls the sensor module to collect abnormal state parameters; The abnormal state parameters include the power fluctuation P(t) at time t, the vibration frequency F(t) at time t, and the noise spectrum S(t) at time t; The power fluctuation P(t) at time t is obtained by a power sensor, the vibration frequency F(t) at time t is obtained by a vibration sensor, and the noise spectrum S(t) at time t is obtained by a noise sensor; Based on the acquired abnormal state parameters, the abnormal characteristic value Yc of each parameter in the abnormal state parameters is calculated and output respectively, and then the abnormal characteristic value Yc of each parameter is summarized into an abnormal characteristic set O; The abnormal feature set O includes the abnormal feature value Yc of the power fluctuation P(t) at time t P（t） , the abnormal characteristic value Yc of the vibration frequency F(t) at time t F（t） and the abnormal characteristic value Yc of the noise spectrum S(t) at time t S（t）； The abnormal characteristic value Yc of the power fluctuation P(t) at time t P（t） The output is calculated by the following algorithm formula; ； In the formula, Pmin represents the lower power limit of the processing equipment, and Pmax represents the upper power limit of the processing equipment. Indicates the power sensitivity adjustment coefficient, Pref indicates the power reference value, represents the first weight value, ln represents the natural logarithm function; S4, based on the acquired abnormal feature set O, perform summary calculation and comprehensive calculation to output the health status index H, and at the same time set the health threshold Hth and the health status index H to perform health assessment, and execute fault response measures based on the assessment results; S5. After the fault response phase is completed, a response effectiveness mechanism is executed. The response effectiveness mechanism outputs a response effectiveness index E by comprehensively calculating the second health status indicator Hnew output after the fault response phase is completed and the health status indicator, and determines the effectiveness of the fault response measures based on the output result of the response effectiveness index E.

2. A data collection method for a sensor network according to claim 1, characterized in that: Said S1 includes S11 and S12; S11, the sensor network includes a sensor module and a communication module, by installing the sensor module on the processing equipment, and using the sensor network to set the collection frequency of the sensor module to the lower limit collection mode of the operating frequency by default, and record the state parameters of the processing equipment at each frequency time node in real time; The sensor module includes a power sensor, a vibration sensor, a temperature sensor and a noise sensor; The state parameters include vibration amplitude A, temperature rise rate △T and noise increment △N; S12, constructing an embedded system, wherein the embedded system includes a memory, an embedded operating system and a CPU processor; Use the 5G communication network to connect the embedded system to the communication module of the sensor network, and transmit the real-time recorded state parameters of the processing equipment to the memory of the embedded system for data storage; Then, the CPU processor extracts the state parameters of the processing equipment in the memory and performs preprocessing, wherein the preprocessing includes timestamp alignment and normalization processing to eliminate the time difference and dimension influence of the state parameters of the processing equipment and obtain a standard data set; The standard data set includes the vibration amplitude A(t) at time t, the temperature rise rate ΔT(t) at time t, and the noise increment ΔN(t) at time t.

3. A data collection method for a sensor network according to claim 2, characterized in that: The S2 includes S21 and S22; S21, constructing a standard change formula, extracting a standard data set and inputting it into the standard change formula, respectively calculating the standard change trend of the state parameter of each processing equipment, and obtaining the standard change rate V of the parameter change; The standard change rate V is calculated and outputted by the following standard change formula; ； Where V i (t) represents the standard change rate V of the parameters in the i-th standard data set at time t, x i (t) represents the parameter of the i-th standard data set at time t, and △t represents the time interval; S22, according to the abnormal upper limit values ​​of vibration, temperature rise and noise of the processing equipment, set the change rate threshold value Vth of the parameters in each standard data set respectively, and obtain the standard change rate Vth of the parameters in the i-th standard data set at time t i (t) respectively making preliminary comparative evaluation with the corresponding change rate threshold value Vth, analyzing the abnormal parameters of the processing equipment, and determining the execution of the event triggering stage based on the preliminary comparative evaluation results. The specific evaluation contents are as follows; At time t, the standard change rate V of the parameters in the i-th standard data set i (t)>threshold value Vth of the parameter of the i-th standard data set i When the processing equipment is judged to be operating abnormally, the event triggering stage is executed; At time t, the standard change rate V of the parameters in the i-th standard data set i (t) ≤ the change threshold Vth of the parameter of the i-th standard data set i , it is determined that the processing equipment is operating normally, and the current operating frequency lower limit acquisition mode is maintained.

4. The data collection method for a sensor network according to claim 1, characterized in that: S32, the abnormal characteristic value Yc of the vibration frequency F(t) at time t F（t） The output is calculated by the following algorithm formula; ； In the formula, Fnom represents the standard vibration frequency of the processing equipment, and Fmax represents the upper limit of the vibration frequency of the processing equipment.

5. The data collection method for a sensor network according to claim 1, characterized in that: S33, the abnormal characteristic value Yc of the noise spectrum S(t) at time t S（t） The output is calculated by the following algorithm formula; ； In the formula, Savg represents the average value of the noise signal, Sstd represents the standard deviation of the noise of the processing equipment, and represents the degree of noise fluctuation, Smax represents the upper limit of the noise spectrum of the processing equipment, and exp represents the exponential function. It represents the sensitivity adjustment coefficient of noise change, and is used to adjust the sensitivity of noise change of processing equipment.

6. The data collection method for a sensor network according to claim 1, characterized in that: The S4 includes S41 and S42; S41, based on all the parameters in the acquired abnormal feature set O, perform weighted summary calculation, output the health status index H of the processing equipment, and quantitatively analyze the health status of the processing equipment; The health status indicator H is calculated and output by the following algorithm formula; ； Where H(t) represents the health status indicator at time t, n represents the total number of parameters in the abnormal feature set O, and W j represents the weight value of the jth parameter in the abnormal feature set O, j (t) represents the jth parameter in the abnormal feature set O at time t.

7. A data collection method for a sensor network according to claim 6, characterized in that: S42, based on the overall health index of the processing equipment, the user sets a health threshold Hth, and then performs a health assessment on the health threshold Hth and the health status index H(t) at time t, analyzes the overall health status of the processing equipment, and triggers the fault response measures of the sensor network based on the health assessment results. The specific assessment contents are as follows; When the health status indicator H(t) at time t is less than the health threshold Hth, it means that the overall processing equipment monitored by the current sensor network is within the normal range, and it is automatically converted to the lower limit collection mode of the operating frequency; When the health status indicator H(t) at time t ≥ the health threshold Hth, it means that the overall processing equipment monitored by the current sensor network is in an abnormal state. At this time, the first response information is sent through the embedded system to prompt the operator to check the processing equipment due to the abnormal equipment, and at the same time, the first response measure is executed. The first response measure increases the current operating frequency to 100% through the sensor network; When the health status indicator H(t) at time t is greater than twice the health threshold Hth, it means that the overall processing equipment monitored by the current sensor network is in a risky state. At this time, the second response information is sent out through the embedded system to prompt the operator to perform maintenance and automatically execute the second response measure. The second response measure is to shut down the sensor network and record the complete status parameters before the shutdown.

8. A data collection method for a sensor network according to claim 7, characterized in that: The S5 includes S51 and S52; S51, after the second response measure is executed, restart the embedded system and the sensor network, output the second health status indicator Hnew after the fault response measure is executed, and calculate the difference between the second health status indicator Hnew and the health status indicator H(t) at time t to obtain the response effectiveness indicator E to analyze the effectiveness of the fault response mechanism; The response effectiveness index E(t) is calculated and outputted by the following algorithm formula; ； In the formula, E(t) represents the effectiveness index at time t, Hnew(t) represents the second health status index at time t, and △t represents the time interval; S52, based on the output result of the effectiveness index E(t) obtained at time t, the effectiveness of the fault response measures is determined, and the specific determination content is as follows; When the effectiveness index E(t) at time t is ≥ 0, it means that the fault response measures are effective, and the sensor network is set to the lower limit acquisition mode of the operating frequency; When the effectiveness indicator E(t) at time t is less than 0, it indicates that the fault response measure is invalid, and the second response measure is executed for the second time.

9. A data collection system for a sensor network, applied to a data collection method for a sensor network according to any one of claims 1 to 8, characterized in that: It includes sensor network module, state change analysis module, event trigger module, fault response module and effectiveness analysis module; The sensor network module sets the sensor module to the lower limit acquisition mode of the operating frequency by default through the sensor network, records the state parameters of the processing equipment in real time, and constructs an embedded system at the same time, transmits the state parameters of the processing equipment to the embedded system through the wireless communication network, and pre-processes the state parameters in the embedded system to obtain a standard data set; The state change analysis module constructs a standard change formula in the embedded system, extracts the state parameters and inputs them into the standard change formula, calculates the standard change trend of the state parameters of each processing equipment, obtains the standard change rate V of the parameter change, sets the change threshold Vth and the standard change rate V of the parameter change for preliminary comparison and evaluation, and enters the event triggering stage based on the preliminary comparison and evaluation results; When the event trigger module finds that any parameter change is abnormal through preliminary comparison and evaluation, it enters the event trigger stage. In the event trigger stage, the operating frequency of the sensor module is increased by 50% through the sensor network, the abnormal state parameters of the processing equipment are recorded, and the abnormal characteristic value Yc of each abnormal state parameter of the processing equipment is calculated and output based on the abnormal state parameters, and the abnormal characteristic value Yc is summarized to obtain the abnormal feature collection O; The fault response module performs a summary calculation based on the acquired abnormal feature set O to output a health status indicator H, sets a health threshold Hth and the health status indicator H to perform a health assessment, and executes a fault response measure based on the assessment result; The effectiveness analysis module executes the response effectiveness mechanism after the fault response phase is completed. The response effectiveness mechanism outputs the response effectiveness index E by comprehensively calculating the second health status indicator Hnew output after the fault response phase is completed and the health status indicator, and determines the effectiveness of the fault response measures based on the output result of the response effectiveness index E.

Images