A solar insect killer lamp Internet of Things fault detection method based on prior knowledge

Abstract

The invention discloses a solar insect-killing lamp Internet of Things fault detection method based on prior knowledge, which belongs to the technical field of distributed fault detection of agricultural Internet of Things equipment, and comprises: obtaining historical data of corresponding faults of solar insect-killing lamp Internet of Things nodes, and establishing a fault rule base; performing threshold parameter and correlation function relationship detection on current measurement values; judging whether the current measurement values ​​meet preset threshold conditions; if so, outputting an abnormal value result "1"; otherwise, outputting an abnormal value result "0"; judging whether the current measurement values ​​meet preset correlation function relationships; obtaining parameter threshold ranges corresponding to different fault data by establishing a fault rule base, obtaining linear function relationships between various parameters, detecting faults that cannot pass the threshold detection through correlation function relationships, and performing sliding window detection through multiple byte caches, thereby greatly reducing the cache share when running a fault self-checking program.

Description

Technical Field

[0001] The invention relates to a solar insecticidal lamp Internet of Things fault detection method based on prior knowledge, belonging to the technical field of distributed fault detection of agricultural Internet of Things equipment. Background Art

[0002] Solar insect killer lamp is an environmentally friendly physical pest control device. When combined with IoT technology, it can achieve functions such as accurate pest counting, pest outbreak area positioning, and equipment energy management. However, solar insect killer lamp IoT nodes are deployed in complex and changeable field scenes, which are easily affected by factors such as aging, theft, and destruction, which eventually lead to node failures and are not conducive to their stable operation.

[0003] Early fault detection and management is a key technology to ensure stable operation of equipment, but current research often adopts the method of transmitting the data collected by the equipment to the background for fault detection, that is, centralized fault detection. This fault detection method has the disadvantages of poor real-time detection and serious data redundancy, and is not suitable for scenarios such as solar insect killer lamp Internet of Things where computing resources are limited and nodes are widely deployed in farmland and other scenarios with poor network signals. The information disclosed in this background technology section is only intended to increase the understanding of the overall background of the present invention, and should not be regarded as an admission or any form of implication that the information constitutes prior art that is already known to ordinary technicians in the field. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a solar insect killer lamp Internet of Things fault detection method based on prior knowledge, which solves the current problem of low efficiency of solar insect killer lamp Internet of Things node fault detection and the need to transmit a large amount of data.

[0005] In order to achieve the above-mentioned purpose / solve the above-mentioned technical problems, the present invention is implemented by adopting the following technical solutions: In the first aspect, a solar insect killer lamp Internet of Things fault detection method based on prior knowledge comprises:

[0006] Get the current measurement value;

[0007] Determine whether the current measurement value meets the preset parameter threshold; if so, output the abnormal value result "1"; otherwise, output the abnormal value result "0";

[0008] Determine whether the current measurement value satisfies the preset correlation function relationship; if so, output the abnormal value result "1"; otherwise, output the abnormal value result "0";

[0009] The output abnormal value result is updated to the sliding window. When the number of "0"s in the sliding window is greater than or equal to 3, the corresponding fault information is output according to the fault information label corresponding to the sliding window.

[0010] Furthermore, the method for setting the preset threshold specifically includes:

[0011] Obtain historical data of corresponding faults of solar insect killer lamp IoT nodes;

[0012] Get the parameter threshold range corresponding to different historical data;

[0013] Mark and label the parameter threshold ranges corresponding to different faults, and describe the fault characteristics;

[0014] The described fault parameter thresholds are stored in the fault rule base.

[0015] Furthermore, the historical data includes: various sensor parameters, voltage and current of the total power supply, voltage and current of the metal mesh, voltage of the insect trap, on current and off current of the insect trap, voltage and current of the solar energy, light intensity, voltage of the rain control module, wind direction and wind speed.

[0016] Furthermore, the method for setting the preset correlation function relationship specifically includes:

[0017] Obtain parameters of two or more components of a solar insect killer lamp under normal conditions;

[0018] Compare the acquired parameters, and obtain and store the linear function relationship between the parameters.

[0019] Furthermore, the size of the sliding window is one byte, and different numbers of "0" in the sliding window correspond to different fault information labels.

[0020] In a second aspect, a solar insect killer lamp Internet of Things fault detection device based on prior knowledge includes a processor and a storage medium; the storage medium is used to store instructions;

[0021] The processor is used to operate according to the instructions to execute the steps of the above method.

[0022] In a third aspect, a computer-readable storage medium stores a computer program, which implements the steps of the above method when executed by a processor.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] The present invention establishes a fault rule base to obtain the parameter threshold range corresponding to different fault data, and obtains the linear function relationship between the parameters according to the parameters of two or more components of the solar insecticidal lamp under normal circumstances. The faults that cannot be detected by the threshold are detected through the correlation function relationship, and sliding window detection is performed through multiple byte caches. The corresponding fault information is output according to the abnormal value results, thereby greatly reducing the cache share when running the fault self-check program, and solving the problem that the current fault detection efficiency of the solar insecticidal lamp Internet of Things node is low and a large amount of data needs to be transmitted. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 A flow chart of a method for detecting faults of a solar insect killer lamp Internet of Things based on prior knowledge provided by an embodiment of the present invention;

[0026] Figure 2 A schematic diagram of a fault rule base of a solar insect killer lamp Internet of Things fault detection method based on prior knowledge provided by an embodiment of the present invention;

[0027] Figure 3 A schematic diagram of labels corresponding to various faults in a method for detecting faults of a solar insect killer lamp Internet of Things based on prior knowledge provided by an embodiment of the present invention;

[0028] Figure 4 is the correlation function between the light intensity value and the photovoltaic panel current;

[0029] Figure 5 A comparison chart of the fault detection results of a solar insect killer lamp Internet of Things fault detection method based on prior knowledge provided in an embodiment of the present invention and two related algorithms and six machine learning algorithms.

[0030] Figure 6 A comparison chart of the relationship between the fault detection model size and running time of a solar insect killer lamp Internet of Things fault detection method based on prior knowledge provided in an embodiment of the present invention and two related algorithms and six machine learning algorithm fault detection. DETAILED DESCRIPTION

[0031] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0032] Embodiment 1

[0033] like Figure 1 As shown, a solar insect killer lamp Internet of Things fault detection method based on prior knowledge includes:

[0034] Obtain historical data of the corresponding fault of the solar insect killer lamp IoT node; historical data includes: various sensor parameters, voltage and current of the total power supply, voltage and current of the metal mesh, voltage of the insect trap, current when the insect trap is turned on and off, voltage and current of the solar energy, light intensity, voltage of the rain control module, wind direction and wind speed;

[0035] like Figure 2 As shown, this embodiment collects historical data from August 14, 2021 to October 4, 2021 by actually deploying 7 solar insect killer lamp IoT nodes;

[0036] Obtain parameter threshold ranges corresponding to different historical data; mark and label parameter threshold ranges corresponding to different faults, and describe fault characteristics; store the described fault parameter thresholds into the fault rule library;

[0037] Get the current measurement value;

[0038] Determine whether the current measurement value meets the preset parameter threshold; specifically:

[0039] For faults that can be detected by threshold parameters, such as abnormal voltage values ​​when an insect trap lamp or metal mesh is open-circuited, and abnormal changes in current values ​​after an insect trap lamp is damaged, the thresholds preset in the fault rule library are used to perform abnormal value judgment; determine whether the current measured value meets the preset threshold conditions; if so, output the abnormal value result "1"; otherwise, output the abnormal value result "0";

[0040] Determine whether the current measurement value satisfies the preset correlation function relationship, specifically:

[0041] For some faults that cannot be detected and judged by threshold parameters, detection is performed through correlation function relationships, including:

[0042] Obtain parameters of two or more components of a solar insect killer lamp under normal conditions;

[0043] Compare the measured parameters and obtain the linear function relationship between the parameters;

[0044] Substitute the current measured value into the corresponding linear function; for example, the light intensity value threshold on a rainy day is different from that on a sunny day. Figure 4 It indicates the mismatch between the light intensity value and the current value of the photovoltaic panel; the corresponding relationship can be expressed as:

[0045]

[0046] Where x represents the light intensity value, and y represents the photovoltaic panel current; the formula shows that when the light intensity value is less than 100000, there is a linear function relationship between the photovoltaic panel current value and the light intensity value, and when the light intensity value is greater than or equal to 100000, the photovoltaic panel current reaches a maximum value of 3200Lux; judge whether the current measurement value meets the preset correlation function relationship; if so, output the abnormal value result "1"; otherwise, output the abnormal value result "0";

[0047] After the sensor collects the latest data, the "1" and "0" obtained by the parameter threshold judgment and the correlation function relationship are updated in a stacked manner to the sliding window of the corresponding fault label. The number of sliding windows is Figure 6 The number of corresponding fault labels in is consistent; the size of each sliding window is one byte, that is, eight binary bits, to store eight recent judgment results, thereby greatly reducing the cache ratio when running the fault self-check program; for each sliding window, when there are three or more "0", the fault label corresponding to the sliding window is output; such as Figure 3 As shown, for example, for F7 fault, by comparing the matching relationship between the photovoltaic panel current value and the light intensity value, and inputting it into the sliding window, when there are three "0s", the fault F7 outputs the corresponding fault information;

[0048] Send fault information to the background or user end through ZigBee devices to reduce the transmission of unnecessary data.

[0049] like Figure 5 and Figure 6 As shown in the figure, in order to analyze the advantages of the present invention, the same data is applied to two similar fault self-detection methods TinyD2 and TrusDet, and is also compared with six classic machine learning methods, including K-nearest neighbor (KNN), linear regression (LR), stochastic gradient descent (SGD), naive Bayes (NB), support vector machine (SVM), gradient boosted tree (GDBT) and random forest (RF). Figure 3 As shown in the figure, the method of the present invention has good performance in accuracy, false detection rate and missed detection rate; the KNN method, RF method and GDBT method are all higher than the method of the present invention in terms of model size and running time;

[0050] Therefore, the method of the present invention can optimize the model size and running time while ensuring accuracy, so that the method of the present invention is suitable for devices with limited computing and storage resources such as solar insect killer lamp Internet of Things nodes.

[0051] Embodiment 2

[0052] A solar insect-killing lamp Internet of Things fault detection device based on prior knowledge, comprising a processor and a storage medium; the storage medium is used to store instructions;

[0053] The processor is configured to operate according to the instructions to perform the steps of the following method:

[0054] Get the current measurement value;

[0055] Determine whether the current measurement value meets the preset parameter threshold; if so, output the abnormal value result "1"; otherwise, output the abnormal value result "0";

[0056] Determine whether the current measurement value satisfies the preset correlation function relationship; if so, output the abnormal value result "1"; otherwise, output the abnormal value result "0";

[0057] The output abnormal value result is updated to the sliding window. When the number of "0"s in the sliding window is greater than or equal to 3, the corresponding fault information is output according to the fault information label corresponding to the sliding window.

[0058] Embodiment 3

[0059] A computer readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the following method:

[0060] Get the current measurement value;

[0061] Determine whether the current measurement value meets the preset parameter threshold; if so, output the abnormal value result "1"; otherwise, output the abnormal value result "0";

[0062] Determine whether the current measurement value satisfies the preset correlation function relationship; if so, output the abnormal value result "1"; otherwise, output the abnormal value result "0";

[0063] The output abnormal value result is updated to the sliding window. When the number of "0"s in the sliding window is greater than or equal to 3, the corresponding fault information is output according to the fault information label corresponding to the sliding window.

[0064] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A solar insect killer lamp Internet of Things fault detection method based on prior knowledge, characterized in that: include: Get the current measurement value; Determine whether the current measurement value meets the preset parameter threshold; If it is satisfied, the abnormal value result "1" is output; Otherwise, the output abnormal value result is "0"; The method for setting the preset parameter threshold specifically includes: Obtain historical data of the solar insect killer lamp IoT node under corresponding faults; the historical data includes: various sensor parameters, voltage and current of the total power supply, voltage and current of the metal mesh, voltage of the insect trap lamp, current when the insect trap lamp is turned on and off, voltage and current of the solar energy, light intensity, voltage of the rain control module, wind direction and wind speed; Get the parameter threshold range corresponding to different historical data; Mark and label the parameter threshold ranges corresponding to different faults, and describe the fault characteristics; The described fault parameter thresholds are stored in a fault rule base; Determining whether the current measurement value satisfies a preset correlation function relationship includes: obtaining parameters of two or more components of the solar insect killer lamp under normal circumstances; Compare the acquired parameters, obtain and store the linear function relationship between the parameters: Get the linear function relationship between the light intensity value and the photovoltaic panel current under normal conditions: ； Where x represents the light intensity value, and y represents the photovoltaic panel current; the formula shows that when the light intensity value is less than 100000, there is a linear function relationship between the photovoltaic panel current value and the light intensity value, and when the light intensity value is greater than or equal to 100000, the photovoltaic panel current reaches a maximum value of 3200Lux; determine whether the current measurement value meets the preset correlation function relationship; If it is satisfied, the abnormal value result "1" is output; otherwise, the abnormal value result "0" is output; The output abnormal value result is updated to the sliding window, the size of the sliding window is one byte, and different numbers of "0" in the sliding window correspond to different fault information labels. When the number of "0" appearing in the sliding window is greater than or equal to 3, the corresponding fault information is output according to the fault information label corresponding to the sliding window.

2. A solar insecticidal lamp Internet of Things fault detection device based on prior knowledge, characterized in that: It includes a processor and a storage medium; the storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to claim 1 .

3. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to claim 1 are implemented.

Images