LoRa-based equipment intelligent monitoring method and monitoring system

Abstract

The invention discloses a LoRa-based equipment intelligent monitoring method and system, and relates to the field of equipment intelligent monitoring, and the method comprises the steps: S1, obtaining the original detection data of equipment in real time through a data collection module disposed on a to-be-detected equipment power supply cable, and collecting the detection data into a matrix every 10 s, thereby forming a group of segmented detection matrixes Xi; s2, transmitting the detection matrix obtained in the S1 to a processing terminal based on a wireless communication network established by a LoRa wireless communication module; s3, the processing terminal processes the received detection matrix through a sectional type improved self-adaptive weighted data fusion algorithm to obtain test data # imgabs0 #; and S4, the operation state of the equipment is judged based on the test data # imgabs1 #, and intelligent monitoring of the equipment is realized. The invention provides an intelligent equipment monitoring method and system based on LoRa, and the method and system are based on a segmented improved adaptive weighted data fusion algorithm of a Grubbs criterion, and improve the accuracy of test data after fusion through employing a mode of two-time weighting processing.

Description

Technical Field

[0001] The present invention relates to the field of intelligent monitoring of equipment. More specifically, the present invention relates to a method and system for intelligent monitoring of equipment based on LoRa. Background Art

[0002] With the rapid development of intelligent and digital applications in industrial production, various types of testing equipment and control systems have emerged. However, they are faced with problems such as multiple terminal system manufacturers, difficult management, and scattered equipment operation data. It is difficult to form unified maintenance and management, and there is an urgent need to uniformly monitor and manage the health status of equipment. In addition, conventional testing equipment itself does not have a data transmission function, and cannot automatically and in real time transmit the health status of the equipment to the terminal for analysis and processing. For equipment with built-in transmission functions (such as equipment with industrial computers, chassis, etc.), corresponding communication media are also required to transmit the health status of the equipment to realize intelligent management and application of the equipment. The objective shortcomings of the existing technology are mainly reflected in the following aspects: 1. The existing equipment in the factory is rich in types, diverse in variety, large in number, widely distributed, and scattered in location, which makes equipment management inconvenient. In addition, the walls between buildings or rooms are thick, resulting in weak data transmission signals and high packet loss rates. If wired data transmission is adopted, not only will the laying area be large, requiring a lot of manpower, material resources, and financial resources, but also a large number of cables are easily affected by environmental humidity, rats, insects, and human damage. When the cables are damaged, it is not easy to locate the fault point, which will bring many invisible risks later.

[0003] 2. Conventional wireless transmission technologies include Bluetooth, ZigBee, Wi-Fi, etc., but these technologies have problems such as short transmission distance and weak confidentiality. It is difficult to transmit signals through walls, and data is easily lost.

[0004] 3. Existing algorithms generally use conventional adaptive data fusion algorithms to directly perform adaptive distribution weighting on data. However, they are sensitive to outliers, have large data fluctuations, are not flexible in weight adjustment, and have large errors. Summary of the invention

[0005] An object of the present invention is to solve at least the above problems and / or disadvantages and to provide at least the advantages which will be described hereinafter.

[0006] In order to achieve these purposes and other advantages of the present invention, a LoRa-based device intelligent monitoring method is provided, comprising: S1. The original detection data of the device is obtained in real time through the data acquisition module placed on the power supply cable of the device under test. The detection data is collected into a matrix every 10 seconds to form a segmented detection matrix X. i ; S2, the wireless communication network based on the LoRa wireless communication module will obtain the detection matrix X i Transmit to a processing terminal; S3, processing terminal segmented improved adaptive weighted data fusion algorithm for the received detection matrix X i Processing to obtain test data ; S4. Based on test data Determine the operating status of the equipment and realize intelligent monitoring of the equipment; Among them, the test data Obtained by the following formula: In the above formula, is the estimated value obtained by twice weighted data fusion processing on the i-th sample value, and n is the number of test data.

[0007] Preferably, in S2, the wireless communication network is configured to adopt any one of a point-to-point communication network, a many-to-one communication network, and a networking communication network.

[0008] Preferably, in S3, the test data The acquisition process includes: S30, the detection matrix X obtained by segmentation i Perform data conversion and calculation to obtain corresponding voltage and current values; S31, based on the improved Grubbs criterion, the voltage and current values ​​obtained in S30 are judged to complete the detection and elimination of abnormal values ​​and obtain a new test matrix Y j , expressed as: Among them, Y is the test data after removing outliers, j is the moment, M represents the number of test data; S32, after performing preliminary weighted calculation on the test matrix obtained in S31, a secondary adaptive weighted data fusion process is performed to obtain the test data ; Among them, in S32, the test matrix Y j The weight value corresponding to each test data in W j The following formula must be satisfied: In the above formula, α is the test matrix Y j The weighted value obtained by preliminary weighted calculation of each test data in the dataY j-1 is the previous test data, M represents the number of test data, For the test matrix Y j The estimated value is obtained by twice weighted data fusion processing of each test data; The calculation formula of the weighted value α is as follows: In the above formula, θ is the standard deviation of the test data, and 𝜅 is the mean of the test data.

[0009] Preferably, in S31, the method of detecting and eliminating the abnormal value is: S310, detection matrix X i Arrange in descending order to remove the maximum and minimum values ​​in the array; S310, calculate the detection matrix X according to the following formula i The mean value X eq , Standard Deviation σ and Grubbs statistic g i : S310, g i Compared with Grubbs' critical value T, if g i ≥ T The corresponding data is judged as an abnormal value and the abnormal data is removed. N is the detection matrix X i The number of samples in .

[0010] Preferably, in S32, the weight value W j Obtained by the following formula: In the above formula, is the mean square error of the jth test data, and σ 2 Obtained by the following formula: In the above formula, W j is the weight value of the jth test data, Y is the test data after removing outliers, E is the expected value, and when the weight value W jWhen is the optimal weight value, the corresponding minimum mean square error is It is characterized by the following formula: .

[0011] Preferably, in S4, if the test data If the device voltage and current values ​​are included, it is judged to be in operation state, and the test data Write the device running day, if the test data If only the device voltage value is included, the test data Write the device sleep log, if the test data If there is no voltage and current value, it is judged to be in the off state.

[0012] A monitoring system, which is applied to a LoRa-based device intelligent monitoring method, comprises: Multiple data acquisition modules arranged on the power supply cables of each device under test; Multiple LoRa transmission modules matched with corresponding data acquisition modules; Multiple LoRa receiving modules matched with corresponding LoRa transmitting modules; A processing terminal that processes and visualizes the data received by the LoRa receiving module; Among them, the data acquisition module is connected to the LoRa transmitting module through a 485 bus, and the LoRa receiving module is connected to the processing terminal through a serial port bus.

[0013] Preferably, the data acquisition module includes a voltage monitoring sensor, a current monitoring sensor, a circuit breaker monitoring sensor, and a temperature and humidity sensor.

[0014] The present invention has at least the following beneficial effects: Firstly, in order to ensure data reliability, the present invention proposes a piecewise improved adaptive weighted data fusion algorithm based on the Grubbs criterion, in which outliers are first eliminated, and the fusion result is not affected by the outliers. Compared with the prior art, it adopts a two-time weighted processing method and smoothes the data, making the fusion result more stable and improving the accuracy of the test data after fusion.

[0015] Secondly, based on the Internet of Things technology platform, the present invention will add a variety of sensor monitoring devices (such as voltage and current monitoring, circuit breaker monitoring, etc.) to the equipment to intelligently monitor the health status of each device, use LoRa communication technology to wirelessly transmit data to the terminal in real time, use algorithms to process data, build a visual display system, realize intelligent management, control and data analysis of various types of equipment, connect different types of equipment, and form an intelligent unified monitoring system for equipment.

[0016] Third, this patent adopts a LoRa wireless spread spectrum communication transmission technology, which has the characteristics of low power consumption, long-distance transmission, strong anti-interference ability, good confidentiality, strong penetration, etc. This technology can be used for point-to-point communication, many-to-one communication and networking communication, and can add transmission nodes at will. It has strong environmental adaptability and can quickly transmit device status data to the terminal for analysis, processing, visualization, and intelligent monitoring.

[0017] Other advantages, objectives and features of the present invention will be embodied in part through the following description, and in part will be understood by those skilled in the art through study and practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flow chart of the improved adaptive weighted data fusion algorithm based on the Grubbs criterion in the present invention; Figure 2 This is a flow chart of the visual interactive software on the processing terminal in the present invention; Figure 3 This is a schematic diagram of the work of realizing point-to-point communication through LoRa in the present invention; Figure 4 This is a working diagram of realizing many-to-one communication through LoRa in the present invention; Figure 5 This is a schematic diagram of the work of realizing networking communication through LoRa in the present invention; Figure 6 This is a principle block diagram of the LoRa-based device intelligent monitoring system in the present invention; Figure 7 The figure is a comparison diagram of the effects of smoothing data using the method of the present invention and the prior art. DETAILED DESCRIPTION

[0019] The present invention is further described in detail below in conjunction with the accompanying drawings so that those skilled in the art can implement the invention with reference to the description.

[0020] The present invention proposes a LoRa-based intelligent monitoring system for equipment, including several equipment data acquisition modules, LoRa wireless communication modules, data processing and visualization display terminals, and power supply modules. A data acquisition module is installed at the front end of the equipment to be tested to collect equipment health status information, and transmit the data to the LoRa wireless communication module through 485 communication. The LoRa wireless communication module then transmits the data to the terminal device through the serial port for data processing and visualization display. The overall system block diagram is shown in the figure. Figure 1 shown.

[0021] The data acquisition module is placed on the device to be tested. In order to facilitate data collection and ensure that the device is in its original state, the data acquisition module is installed on the 220V power supply cable. A data acquisition module is installed on the power supply cable of each device to be tested to collect equipment health status information. The data acquisition module includes voltage monitoring, current monitoring, circuit breaker monitoring, temperature and humidity sensors, etc. Among them, the voltage monitor is used to collect the input voltage of the device; the current monitor is used to collect the input current of the device; the circuit breaker monitoring is used to collect whether the device has a circuit breaker phenomenon; the temperature and humidity sensor is used to collect the temperature and humidity of the working environment of the device and transmit data in a timely manner. These data sensing devices are integrated into a data acquisition module to integrate equipment status data.

[0022] The LoRa wireless communication module includes a LoRa transmitting module and a LoRa receiving module, which are placed on the device under test and the PC end respectively, and are used to transmit or receive the status data of the device under test. In order to adapt to different application scenarios, the present invention proposes three LoRa wireless communication networking modes.

[0023] One is point-to-point communication, where the LoRa transmitter module transmits data directly to the LoRa receiver module, such as Figure 3 As shown. The data acquisition module and the LoRa transmitter module are connected, and the two use 485 communication mode, and both are placed on the device under test. The LoRa transmitter module is used to receive the data collected by the data acquisition module and transmit it to the next node. The LoRa transmitter module corresponds to the LoRa receiver module one by one, and is paired to achieve long-distance communication transmission. The LoRa receiver module is placed in the PC terminal and connected to the PC terminal to receive the data transmitted by the LoRa transmitter module, and transmit the data to the PC terminal through the serial port for algorithm data processing and visual display.

[0024] The second is many-to-one communication, where multiple LoRa transmitter modules transmit data to the LoRa gateway receiving module respectively. Figure 4 As shown. The data acquisition module and the LoRa transmitter module are connected, and the two use 485 communication mode, and both are placed on the device under test. The LoRa transmitter module is used to receive the data collected by the data acquisition module and transmit it to the next node. The data receiving end uses the LoRa gateway receiving module, which is designed to integrate several LoRa receiving modules, while achieving long-distance communication, it also reduces the data transmission cost. The LoRa gateway receiving module is placed on the PC terminal and connected to the PC terminal. It is used to receive the data transmitted by several LoRa transmitter modules, and transmit the data to the PC terminal through the serial port for algorithm data processing and visual display.

[0025] The third is networking communication. Multiple LoRa transmitter modules transmit data to the LoRa mid-end receiving module respectively, and several LoRa mid-end receiving modules then transmit the data to the LoRa gateway receiving module. Figure 5 As shown. The LoRa transmitter module and the data acquisition module are integrated into one, and the two use 485 communication mode. Both are installed on the device under test to receive the data collected by the data acquisition module and transmit the data to the next node. Several LoRa mid-end receiving modules are located between the LoRa transmitter module and the LoRa gateway receiving module, responsible for transferring data information and extending the communication transmission distance. The LoRa gateway receiving module is located at the data processing terminal, which is used to receive the data transmitted by the LoRa terminal receiving module over a long distance, and transmit the data to the terminal through serial communication for algorithm processing and visual display.

[0026] The data processing and visualization display terminal belongs to the PC side. The data processing side extracts the state information data of the device under test transmitted by LoRa wireless through base conversion, data analysis, data fitting, and algorithm processing. Since data transmission is easily affected by external interference, the accuracy of the data is affected, resulting in data anomalies. Figure 1 As shown, the present invention uses a segmented improved adaptive weighted data fusion algorithm based on the Grubbs criterion to filter out invalid and abnormal data in real time, extract valid data for data fitting and algorithm processing, and aims to improve test accuracy.

[0027] (1) Data conversion and calculation: extract the effective digits of current and voltage from the data transmitted by the data acquisition end, perform base conversion, and calculate the voltage and current values ​​of the equipment.

[0028] (2) Outlier detection and removal The segmented method refers to collecting the collected test data into a matrix at a predetermined interval (such as 10 seconds) to form a set of detection matrices X i : Among them, X is the detection data, X i is the detection data at the i-th moment, and N is the number of sample values ​​in the array.

[0029] Outlier detection and elimination uses the improved Grubbs criterion to arrange the test data in descending order, first eliminating the maximum and minimum values ​​in the array to avoid increasing errors; then calculate the mean X of the test data according to the following formula eq , Standard Deviation σ and Grubbs statistic g i : Grubbs statistic g iCompare with Grubbs' critical value T. The critical value can be obtained by looking up the Grubbs criterion critical value table. The present invention uses T=2.19. If the statistics corresponding to the collected test data g i ≥ T , it is considered that the data has a large error and should be eliminated; otherwise, the collected test data is normal data.

[0030] (3) Segment-wise improved adaptive weighted data fusion processing The test data after outlier detection is processed by adaptive weighted data fusion, the weight coefficient of each data is adjusted in real time, and then the data is weighted fused according to the weight value to further improve the accuracy of the test data.

[0031] The segmented improved adaptive weighted data fusion processing means that the present invention groups the real-time test data in a segmented manner, and performs adaptive weighted data fusion processing on each group of test data. The improved adaptive weighted data fusion algorithm processing flow proposed by the present invention is as follows: First, the detection matrix X i After outlier detection, a new test matrix is ​​formed Y j for: Among them, Y is the test data after removing outliers, j is the moment, M represents the number of test data; Secondly, based on the test matrix after exception handling Y j First, perform a preliminary weighted calculation on each test data, and then perform a secondary weighted data fusion to obtain the value to be estimated. , and the test matrix Y j Each test data and its weight value W j Satisfy the following formula: In the above formula, α is the test matrix Y j The weighted value obtained by preliminary weighted calculation of each test data in the data Y j-1 is the test data of the previous moment, M represents the number of test data, For the test matrix Y j The estimated value is obtained by performing two weighted data fusion processes on each test data.

[0032] Obviously, the present invention first performs a preliminary weighted calculation on the test data after abnormal processing. In order to allow the weighted value to change dynamically with the data, the present invention constructs a calculation formula for the weighted value α, as follows: In the above formula, θ is the standard deviation of the test data, and 𝜅 is the mean of the test data.

[0033] The average σ of the test data 2 for: The optimal weight value corresponding to each test data can be obtained by Lagrange multiplier method and extreme value theory. W j for: The optimal weight value W j The corresponding minimum mean square error is: Finally, substitute the optimal weight value W j , and the estimated value is obtained .

[0034] (4) Data averaging: To further improve data validity, the data is averaged. The test data after fusion of every 10 adaptive data is stored in an array, the maximum and minimum values ​​are filtered out, and mean filtering is performed to calculate the average value of the array samples. The algorithm used is as follows: in, For test data, is the estimated value obtained by twice weighted data fusion processing of the i-th sample value, n is the detection matrix X i The number of sample values ​​in .

[0035] (5) Data output: test data after algorithm processing , is the value obtained by the algorithm.

[0036] The processing terminal (also called PC terminal) for data processing and visual display is used to display the device status information in real time to improve the human-computer interaction of the present invention. The software design process of the processing terminal is as follows: Figure 2The following are shown: initialization program, serial communication pairing between the terminal and the LoRa module, obtaining the device status information transmitted by LoRa, performing base conversion, data analysis, data fitting and algorithm processing, extracting the device status information, and then judging the device's operating status, and writing the device operation and sleep logs. The visual display terminal adds functions such as historical operation record query, real-time display of device working status, display of device working location, device life expiration warning, and real-time display of ambient temperature and humidity.

[0037] The software of the processing terminal can calculate the power consumption of the device through the collected voltage and current values, and determine the state of the device. If the device voltage and current values ​​can be collected at the same time, that is, the device has power consumption, the device is in operation; if only the device voltage value can be collected, that is, the device is only powered on, the device is in sleep state; if there is no voltage and current value, the device is in shutdown state.

[0038] The power supply module is used to supply power to each module.

[0039] The present invention also includes the following beneficial technical effects: 1. The present invention not only facilitates the collection of device data by adding a data acquisition module to the 220V power supply cable of the device, but also ensures that the device is in its original state without being damaged, and has certain versatility; 2. The present invention proposes three LoRa wireless communication networking modes, namely point-to-point communication, many-to-one communication and networking communication, which can flexibly cope with different communication working environments; 3. The present invention proposes a piecewise improved adaptive weighted data fusion algorithm based on the Grubbs criterion, which first removes outliers, and the fusion result is not affected by outliers. Compared with the prior art, it uses a two-time weighted processing method to smooth the data, making the fusion result more stable and improving the accuracy of the test data after fusion; 4. The present invention develops a visual interface for monitoring the health status of equipment, performs wireless real-time monitoring, and can directly display basic information, health status, and operation status of equipment, so that managers can intuitively obtain the required information; 5. The present invention applies LoRa technology to intelligent equipment monitoring, realizes wireless, real-time, and long-distance monitoring of equipment status, and has a certain wall-penetrating effect, which greatly avoids the problem of signal attenuation during communication between buildings; 6. The implementation of this invention can be applied to the intelligent management, control and data analysis of various types of equipment, connecting different types of equipment to form an intelligent unified monitoring system for equipment, improving the scheduling capabilities of workshops and factories, and monitoring the operation and health status of equipment in real time. It has a wide range of application scenarios and great social significance; 7. The present invention applies LoRa technology to intelligent equipment monitoring, realizes wireless, real-time, and long-distance monitoring of equipment status, and has a certain wall-penetrating effect, which greatly avoids the problem of signal attenuation during communication between buildings.

[0040] Example: In order to verify the feasibility of the segmented improved adaptive weighted data fusion algorithm based on the Grubbs criterion proposed in the present invention, the device current data collected by the sensor is used as an example. 10 groups of collected current data are extracted for analysis, and the data are shown in Tables 1 and 2: Table 1 Table 2 The above data is applied to the method proposed in the present invention, and the data is processed for outliers based on the Grubbs criterion, and then substituted into the formula to calculate the weighted value 𝛼, and a weighted smoothing is performed, and finally an adaptive weighted data fusion process is performed. Compared with the existing traditional adaptive weighted data fusion algorithm, the error comparison diagram calculated by the two is as follows: Figure 7 As shown ( Figure 7 The horizontal axis represents the data sequence number in the test data, such as the first test data, and the vertical axis represents the mean square error). It can be seen that the segmented improved adaptive weighted data fusion algorithm based on the Grubbs criterion proposed in the present invention has a smaller error, a more stable fusion result, and a higher accuracy of the processed data.

[0041] The above solution is only an illustration of a preferred embodiment, but is not limited thereto. When implementing the present invention, appropriate replacement and / or modification can be performed according to user needs.

[0042] Although the embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and the embodiments. It can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily realized. Therefore, without departing from the general concept defined by the claims and equivalent scope, the present invention is not limited to the specific details and the illustrations shown and described here.

Claims

1. A LoRa-based device intelligent monitoring method, characterized in that: include: S1. The data acquisition module placed on the power supply cable of the device under test is used to obtain the original detection data of the device in real time. The detection data is collected into a matrix every 10 seconds to form a set of segmented detection matrices X i ; S2, the wireless communication network based on the LoRa wireless communication module will obtain the detection matrix X i Transmit to a processing terminal; S3, processing terminal segmented improved adaptive weighted data fusion algorithm for the received detection matrix X i Processing to obtain test data ; S4. Based on test data Determine the operating status of the equipment and realize intelligent monitoring of the equipment; Among them, the test data Obtained by the following formula: In the above formula, is the estimated value obtained by twice weighted data fusion processing on the i-th sample value, and n is the number of test data.

2. The LoRa-based intelligent monitoring method for equipment as claimed in claim 1, characterized in that: In S2, the wireless communication network is configured to adopt any one of a point-to-point communication network, a many-to-one communication network, and a networking communication network.

3. The LoRa-based intelligent monitoring method for equipment as claimed in claim 1, characterized in that: In S3, test data The acquisition process includes: S30, the detection matrix X obtained by segmentation i Perform data conversion and calculation to obtain corresponding voltage and current values; S31, based on the improved Grubbs criterion, the voltage and current values ​​obtained in S30 are judged to complete the detection and elimination of abnormal values ​​and obtain a new test matrix Y j , expressed as: Among them, Y is the test data after removing outliers, j is the moment, M represents the number of test data; S32, after performing preliminary weighted calculation on the test matrix obtained in S31, a secondary adaptive weighted data fusion process is performed to obtain the test data ; Among them, in S32, the test matrix Y j The weight value corresponding to each test data in W j The following formula must be satisfied: In the above formula, α is the test matrix Y j The weighted value obtained by preliminary weighted calculation of each test data in the data Y j-1 is the previous test data, M represents the number of test data, For the test matrix Y j The estimated value is obtained by twice weighted data fusion processing of each test data; The calculation formula of the weighted value α is as follows: In the above formula, θ is the standard deviation of the test data, and 𝜅 is the mean of the test data.

4. The LoRa-based intelligent monitoring method for equipment as claimed in claim 3, characterized in that: In S31, the method of detecting and eliminating the abnormal value is: S310, detection matrix X i Arrange in descending order to remove the maximum and minimum values ​​in the array; S310, calculate the detection matrix X according to the following formula i The mean value X eq , Standard Deviation σ and Grubbs statistic g i : S310, g i Compared with Grubbs' critical value T, if g i ≥ T The corresponding data is judged as an abnormal value and the abnormal data is removed. N is the detection matrix X i The number of samples in .

5. The LoRa-based device intelligent monitoring method according to claim 3, characterized in that: In S32, the weight value W j Obtained by the following formula: In the above formula, is the mean square error of the jth test data, and σ 2 Obtained by the following formula: In the above formula, W j is the weight value of the jth test data, Y is the test data after removing outliers, E is the expected value, and when the weight value W j When is the optimal weight value, the corresponding minimum mean square error is It is characterized by the following formula: 。 6. The LoRa-based device intelligent monitoring method according to claim 1, characterized in that: In S4, if the test data If the device voltage and current values ​​are included, it is judged to be in operation state, and the test data Write the device running day, if the test data If only the device voltage value is included, the test data Write the device sleep log, if the test data If there is no voltage and current value, it is judged to be in the shutdown state.

7. A monitoring system, applied to the LoRa-based device intelligent monitoring method as claimed in any one of claims 1 to 6, characterized in that: include: Multiple data acquisition modules arranged on the power supply cables of each device under test; Multiple LoRa transmission modules matched with corresponding data acquisition modules; Multiple LoRa receiving modules matched with corresponding LoRa transmitting modules; A processing terminal that processes and visualizes the data received by the LoRa receiving module; Among them, the data acquisition module is connected to the LoRa transmitting module through a 485 bus, and the LoRa receiving module is connected to the processing terminal through a serial port bus.

8. The monitoring system according to claim 7, characterized in that: The data acquisition module includes a voltage monitoring sensor, a current monitoring sensor, a circuit breaker monitoring sensor, and a temperature and humidity sensor.

Images