An edge fault analysis system and method for an embedded street lamp controller

Abstract

The present invention discloses an edge fault analysis system and method of an embedded street lamp controller, including a fault diagnosis module, the input end of the fault diagnosis module is connected to the output end of the control host and the terminal device, and the output end of the fault diagnosis module is connected to the input end of the storage module and the display module; the fault diagnosis module collects the working parameters and fault data information of the terminal device, generates a diagnostic analysis report and a maintenance plan after analysis and processing, and stores them in the storage module, or sends them to the display module to inform maintenance personnel. The present invention can quickly extract features and accurately analyze the cause of the fault; it can distinguish the level of the fault cause to facilitate self-repair and maintenance, and timely corresponding analysis and maintenance work.

Description

Technical Field

[0001] The present invention relates to the technical field of edge fault analysis, and in particular to an edge fault analysis system and method for an embedded street lamp controller. Background Art

[0002] In recent years, with the increasing application of street light controllers, higher requirements have been placed on the reliability and stability of controllers. As an indispensable component in smart city lighting, a controller failure during operation will have a great impact on the entire street lighting system. If the fault cannot be predicted and handled in time, it will often cause more serious equipment damage or safety accidents, resulting in high maintenance costs and serious economic losses, and even safety problems. Existing controller fault diagnosis technology has the following problems:

[0003] (1) Traditional diagnostic methods require manual feature extraction, which is time-consuming and produces unstable diagnostic results;

[0004] (2) Convolutional neural network diagnosis methods that do not require manual feature extraction require a large amount of computing resources and a long training time, which conflicts with the real-time response requirements of fault diagnosis;

[0005] (3) Fault analysis of faulty equipment is not graded, self-repair and maintenance capabilities are poor, and the repair cycle is long. Summary of the invention

[0006] To solve the above problems, the present invention provides an edge fault analysis system and method for an embedded street lamp controller, which can quickly extract features and accurately analyze the cause of the fault; classify the cause of the fault to facilitate self-repair and maintenance, and timely respond to analysis and maintenance work.

[0007] The technical solution of the present invention is: an edge fault analysis system of an embedded street lamp controller, including a fault diagnosis module, the input end of the fault diagnosis module is connected to the output end of the control host and the terminal device, and the output end of the fault diagnosis module is connected to the input end of the storage module and the display module; the fault diagnosis module collects the working parameters and fault data information of the terminal device, generates a diagnostic analysis report and a maintenance plan after analysis and processing, and stores them in the storage module, or sends them to the display module to notify the maintenance personnel.

[0008] Preferably, in the present technical solution, the fault diagnosis module includes an acquisition unit, which acquires the working parameters and status information of the faulty equipment and sends it to the data unit for temporary storage, or is converted by the signal processing unit and input into the fault diagnosis model. The fault diagnosis model generates a fault analysis report and a maintenance plan after analysis and processing, and sends it to the control host through the protocol unit.

[0009] Preferably, in the present technical solution, the acquisition unit includes an acquisition device for acquiring operating parameters and operating conditions of the faulty equipment, and the operating conditions at least include basic equipment parameters and working environment information.

[0010] Preferably, in the present technical solution, the fault diagnosis model includes a training unit and a fault diagnosis unit. The training model in the training unit obtains the data signal output by the signal processing unit, performs training and learning in the training model, and sends the training and learning results to the fault diagnosis unit. The fault analysis model in the fault diagnosis unit analyzes and processes the results, makes a fault level judgment, and generates a fault report and a maintenance plan and sends them out.

[0011] Preferably, in the present technical solution, the data signal of the fault analysis model obtains fault level differentiation through fault discrimination, obtains an alarm result based on the fault level differentiation, and the result of merging the alarm result with the diagnostic result obtained through fault diagnosis is transmitted to the control host through the protocol unit.

[0012] Preferably, in the present technical solution, when distinguishing the fault level, the fault analysis model calculates the fault level according to the difference between the normal target and the abnormality.

[0013] Another object of the present invention is to provide an analysis method of an edge fault analysis system of an embedded street lamp controller, characterized in that it comprises the following steps:

[0014] S100, collecting data information of faulty equipment;

[0015] S200, storing fault data information and performing feature extraction;

[0016] S300, inputting feature extraction into a training module for training and learning;

[0017] S400, judging whether the equipment failure level and maintenance cycle meet the requirements;

[0018] S500, the fault analysis model generates a fault diagnosis analysis report and a maintenance plan;

[0019] S600, output report and plan.

[0020] Preferably, in the present technical solution, in step S300, after obtaining the characteristic extraction value, it is loaded into the fault diagnosis model together with the parameter information of the equipment configuration for training and learning, and then the fault diagnosis model is used on the control host to perform real-time diagnosis of equipment faults. After the diagnosis is completed, the labeled data will be uploaded to the database of the data unit.

[0021] Preferably, in the present technical solution, the step S300 comprises the following steps:

[0022] S310, calculating the difference and generating an adjustment strategy;

[0023] S320, selecting a device working behavior according to the adjustment strategy;

[0024] S330, the terminal device executes the working behavior, obtains new parameters, and observes the next state information;

[0025] S340, selecting new working parameters and status information for training;

[0026] S350, updating the working parameters and behaviors of the faulty device;

[0027] S360, determine whether the new working state of the terminal equipment is normal. If it is normal, end the training and output the fault analysis and maintenance results. If it is abnormal, repeat the training and learning. After repeating the training and learning for 5 times, if it is still abnormal, stop the training and report.

[0028] Preferably, in the present technical solution, in step S500, different fault diagnosis analysis reports and maintenance plans are reported according to different safety levels.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] The present invention is trained and learned in the training model, and can quickly extract features and accurately analyze the cause of the fault; by comparing the parameter differences, the difference range is calculated, and the fault level is defined according to different ranges, and different fault levels are processed, which is conducive to self-repair and maintenance, and timely corresponding analysis and maintenance work. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is an architectural diagram of a fault analysis system of the present invention;

[0032] Figure 2 is an architectural diagram of a fault diagnosis module of the present invention;

[0033] Figure 3 is a working connection diagram of the fault diagnosis of the present invention;

[0034] Figure 4 It is a work flow chart of the fault analysis system of the present invention;

[0035] Figure 5 It is a work flow chart of the fault diagnosis module of the present invention. DETAILED DESCRIPTION

[0036] 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.

[0037] like Figure 1 As shown, an edge fault analysis system of an embedded street lamp controller of the present invention includes a fault diagnosis module 200, the input end of the fault diagnosis module 200 is connected to the output end of the control host 300 and the terminal device 100, the output end of the fault diagnosis module 200 is connected to the input end of the storage module 400 and the display module 500, the fault diagnosis module 200 collects the working parameters and fault data information of the terminal device, generates a diagnosis analysis report and a maintenance plan after analysis and processing, and stores them in the storage module 400, or sends them to the display module 500 to inform the maintenance personnel.

[0038] Further, such as Figure 2 and Figure 3 As shown, the fault diagnosis module 200 includes a collection unit 210. The collection unit 210 collects the working parameters and status information of the faulty equipment and sends it to the data unit 240 for temporary storage, or is converted by the signal processing unit 220 and input into the fault diagnosis model 230. The fault diagnosis model 230 generates a fault analysis report and a maintenance plan after analysis and processing, and sends it to the control host 300 through the protocol unit 250.

[0039] More specifically, if Figure 3 As shown, the acquisition unit 210 includes an acquisition device 211, which is used to collect the operating parameters and working conditions of the faulty equipment. The working conditions at least include basic equipment parameters and working environment information. The basic parameters include working voltage, current, signal input, output state, acquired control equipment parameters, etc. The working environment information includes information such as location, temperature and humidity. The signal processing unit 220 is used to convert the data collected by the acquisition unit 210 into data signals; the data unit 240 is provided with a database 241 to provide data storage, query, retrieval and processing; the fault diagnosis model 230 at least includes a training unit 231 and a fault diagnosis unit 232. The training model 2311 in the training unit 231 obtains the data signal output by the signal processing unit 220, performs training and learning in the training model 2311, and sends the training and learning results to the fault diagnosis unit 232, which is analyzed and processed by the fault analysis model 2321 in the fault diagnosis unit 232, performs fault level judgment, and generates a fault report and a maintenance plan to send out.

[0040] It should be noted that Figure 3In the method flow shown, the data signal of the fault analysis model is distinguished by fault discrimination to obtain a fault level, and an alarm result is obtained according to the fault level distinction, and the result of merging the alarm result with the diagnosis result obtained by fault diagnosis is transmitted to the control host through the protocol unit; wherein the alarm result includes the channel number, the location of the mechanical equipment measurement point, the time point of data collection, the signal type, the threshold value, and the threshold trigger result; the diagnosis result includes the channel number, the location of the mechanical equipment measurement point, the time point of data collection, the recognition result, and the diagnosis rate; the threshold trigger result is used to express whether it is triggered, which can be expressed by yes / no, etc.

[0041] When distinguishing the fault level, the fault analysis model 2321 calculates the fault level based on the difference between the normal target and the abnormality. If the difference is less than a certain size (target setting value), it is considered to have reached the vicinity of the normal target, and positive feedback is output. It is judged that the equipment has successfully performed the task, and the fault level is 0, and a safe operation signal is output; if the difference exceeds a certain size (target setting value), it is considered that the equipment has failed, and negative feedback is output. It is judged to be an equipment failure, and different numerical values ​​are output according to the size of the difference to indicate the fault level; for example, the target value setting value can be defined as Q, and when the difference D exceeds 20%, 50%, 80%, and 100% of the target setting value, the output fault levels are defined as 1, 2, 3, and 4, respectively, and the larger the output value, the higher the fault level.

[0042] The present invention provides an analysis method for an edge fault analysis system of an embedded street lamp controller, as follows Figure 4 As shown, the following steps are included:

[0043] S100, collecting data information of faulty equipment;

[0044] Specifically, basic equipment parameters and working environment information are collected through collection equipment, where the basic parameters include working voltage, current, signal input, output status, collected control equipment parameters, etc., and the working environment information includes location, temperature and humidity, etc. More equipment status terminal measurement sensors can also be added to collect information such as the location of the terminal device and light intensity.

[0045] S200, storing fault data information and performing feature extraction;

[0046] Specifically, the collected data information is stored in a data unit, the data information is converted into a data format, and then data features are extracted.

[0047] S300, inputting feature extraction into a training module for training and learning;

[0048] In step S300, after obtaining the characteristic extraction value, it is loaded into the fault diagnosis model together with the parameter information of the device configuration for training and learning, and then the fault diagnosis model is used on the control host to perform real-time diagnosis of device faults. After the diagnosis is completed, the labeled data will be uploaded to the database of the data unit. The specific process includes the following steps:

[0049] S310, calculating the difference and generating an adjustment strategy;

[0050] In the fault analysis module, the target setting value can be set to Q, x represents a parameter at time t, which can be collected, and the objective function can be defined as Q(x), thus the difference function can be obtained as After the calculation is completed, the result parameters are transmitted to the data unit, and the parameter adjustment range corresponding to the difference function is found in the data unit by looking up the table;

[0051] S320, selecting a device working behavior according to the adjustment strategy;

[0052] Specifically, the parameter adjustment range is converted into a work behavior data control signal, and the equipment work control information is generated and sent;

[0053] S330, the terminal device executes the working behavior, obtains new parameters, and observes the next state information;

[0054] Specifically, the terminal device receives the working behavior digital signal, obtains new working state parameters, adjusts the working state by itself, and automatically uploads the operating parameters after the adjusted working state to the training model;

[0055] S340, selecting new working parameters and status information for training;

[0056] S350, updating the working parameters and behaviors of the faulty device;

[0057] S360, determine whether the new working state of the terminal equipment is normal. If it is normal, end the training and output the fault analysis and maintenance results. If it is abnormal, repeat the training and learning. After repeating the training and learning for 5 times, if it is still abnormal, stop the training and report.

[0058] S400, judging whether the equipment failure level and maintenance cycle meet the requirements;

[0059]

[0060] When the difference of D(x) is in the range of 50≥D(x)>20, a positive feedback of 1 is given; when it is in the range of 80≥D(x)>50, a negative feedback of -2 is given; when it is in the range of 100≥D(x)>80, a negative feedback of 3 is given; when D(x)>100, a negative feedback of 3 is given; in other cases, the feedback is 0, and the fault is regarded as being at a safe level.

[0061] S500, the fault analysis model generates a fault diagnosis analysis report and a maintenance plan;

[0062] Specifically, the system reports different fault diagnosis analysis reports and maintenance plans according to different safety levels.

[0063] S600, output report and plan.

[0064] In summary, the present invention can quickly extract features and accurately analyze the cause of the fault; it can classify the causes of the fault to facilitate self-repair and maintenance, and timely corresponding analysis and maintenance work.

[0065] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0066] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. An edge fault analysis system for an embedded street light controller, characterized in that: It includes a fault diagnosis module, the input end of which is connected to the output end of the control host and the terminal device, and the output end of which is connected to the input end of the storage module and the display module; the fault diagnosis module collects the working parameters and fault data information of the terminal device, generates a diagnosis analysis report and a maintenance plan after analysis and processing, and stores them in the storage module, or sends them to the display module to inform the maintenance personnel; The fault diagnosis module includes: the acquisition unit collects the working parameters and status information of the faulty equipment, sends them to the data unit for temporary storage, or inputs them into the fault diagnosis model after conversion by the signal processing unit, and generates a fault analysis report and maintenance plan after analysis and processing by the fault diagnosis model, which are sent to the control host through the protocol unit; The fault diagnosis model at least includes a training unit and a fault diagnosis unit. The training model in the training unit obtains the data signal output by the signal processing unit, performs training and learning in the training model, and sends the training and learning results to the fault diagnosis unit. The fault analysis model in the fault diagnosis unit analyzes and processes the results, determines the fault level, and generates a fault report and a maintenance plan to send out. The data signal of the fault analysis model is distinguished by fault discrimination to obtain fault level, and the alarm result is obtained according to the fault level distinction, and the result of merging the alarm result with the diagnosis result obtained by fault diagnosis is transmitted to the control host through the protocol unit; wherein, the alarm result includes the channel number, the location of the mechanical equipment measurement point, the time point of data collection, the signal type, the threshold value, and the threshold trigger result; the diagnosis result includes the channel number, the location of the mechanical equipment measurement point, the time point of data collection, the recognition result, and the diagnosis rate; the threshold trigger result is used to express whether it is triggered; When distinguishing the fault level, the fault analysis model calculates the fault level based on the difference between the normal target and the abnormality. If the difference is less than the target setting value, it is considered to have reached the normal target, and positive feedback is output. It is judged that the equipment has successfully performed the task, and the fault level is 0, and a safe operation signal is output; if the difference exceeds the target setting value, it is considered that the equipment has failed, and negative feedback is output. It is judged to be a device failure, and different values ​​are output according to the size of the difference to indicate the fault level. The larger the output value, the higher the fault level. In the fault analysis model, the target setting value is set to Q, x represents a parameter at time t, and the objective function is defined as Q(x). The difference function is obtained as follows: After the calculation is completed, the result parameters are transmitted to the data unit, and the parameter adjustment range corresponding to the difference function is found in the data unit by looking up the table; the device working behavior is selected according to the adjustment strategy; the terminal device receives the working behavior digital signal, obtains the new working state parameters, adjusts the working state by itself, and automatically sends the operation after the adjustment of the working state to the terminal device. Upload parameters to the training model; select new working parameters and status information for training; update the working parameters and behaviors of the faulty equipment; determine whether the new working status of the terminal equipment is normal. If it is normal, end the training and output the fault analysis and maintenance results. If it is abnormal, repeat the training and learning. After repeated training and learning, if it is still abnormal, stop the training and report; determine whether the equipment fault level and maintenance cycle meet the requirements; When the difference of D(x) is in the range of 50≥D(x)>20, a positive feedback of 1 is given; when it is in the range of 80≥D(x)>50, a negative feedback of -2 is given; when it is in the range of 100≥D(x)>80, a negative feedback of 3 is given; when D(x)>100, a negative feedback of 3 is given; in other cases, the feedback is 0, and the fault is regarded as being at a safe level; the fault analysis model reports different fault diagnosis analysis reports and maintenance plans according to different safety levels, and outputs reports and plans.

2. The edge fault analysis system of the embedded street light controller according to claim 1, characterized in that: The fault diagnosis module includes a collection unit, which collects the working parameters and status information of the faulty equipment and sends them to the data unit for temporary storage, or is converted by the signal processing unit and input into the fault diagnosis model. The fault diagnosis model generates a fault analysis report and maintenance plan after analysis and processing, and sends them to the control host through the protocol unit.

3. The edge fault analysis system of the embedded street light controller according to claim 2, characterized in that: The collection unit includes a collection device for collecting operating parameters and working conditions of the faulty equipment, and the working conditions at least include basic equipment parameters and working environment information.

4. The analysis method of the edge fault analysis system of the embedded street lamp controller according to any one of claims 1 to 3, characterized in that: The steps include: S100, collecting data information of faulty equipment; S200, storing fault data information and performing feature extraction; S300, extracting features and inputting them into a training model for training and learning; S400, judging whether the equipment failure level and maintenance cycle meet the requirements; S500, the fault analysis model generates a fault diagnosis analysis report and a maintenance plan; S600, output report and plan.

5. The analysis method according to claim 4, characterized in that In step S300, after the characteristic extraction value is obtained, it is loaded onto the fault diagnosis model together with the parameter information of the device configuration for training and learning. The fault diagnosis model is then used on the control host to perform real-time diagnosis of the device fault. After the diagnosis is completed, the labeled data is uploaded to the database of the data unit.

6. The analysis method according to claim 4, characterized in that The step S300 includes the following steps: S310, calculating the difference and generating an adjustment strategy; S320, selecting a device working behavior according to the adjustment strategy; S330, the terminal device executes the working behavior, obtains new parameters, and observes the next state information; S340, selecting new working parameters and status information for training; S350, updating the working parameters and behaviors of the faulty device; S360, determine whether the new working state of the terminal equipment is normal. If it is normal, end the training and output the fault analysis and maintenance results. If it is abnormal, repeat the training and learning. After repeating the training and learning for 5 times, if it is still abnormal, stop the training and report.

7. The analysis method according to claim 4, characterized in that In step S500, different fault diagnosis analysis reports and maintenance plans are reported according to different safety levels.

Images