Large outdoor advertisement identification display fault intelligent diagnosis and early warning method
By classifying the display faults of large outdoor advertising logos and setting a threshold function, combined with a multi-forktree early warning system, fast and accurate fault diagnosis is achieved, and the problem of slow fault diagnosis progress in the existing technology is solved.
Patent Information
- Application Number
- CN202411739652.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-29
- Publication Date
- 2025-03-21
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing large outdoor advertising logo display fault diagnosis technology is slow to identify and handle various hardware, software and network failures quickly and effectively.
By classifying large outdoor advertising logo display faults into hardware faults, software faults and network connection faults, and setting the first threshold function f12 and the second threshold function f23 of various parameters in each fault classification, collecting and judging the security types of parameters in real time, and creating a multi-forktree structure fault warning system to achieve rapid fault diagnosis and early warning.
This method significantly improves the efficiency and accuracy of fault diagnosis of large outdoor advertising logo display, can quickly identify and deal with faults, and reduces the time for manual verification.
Smart Images
Figure CN119669856A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to intelligent diagnosis and fault detection technology, and in particular to an intelligent diagnosis and early warning method for display faults of large outdoor advertising signs. Background Art
[0002] Large-scale outdoor advertising sign display technology usually refers to large display screens, billboards or other visual display devices used in public spaces such as cities, commercial areas or large event venues; these devices use advanced display technology to display advertising content, information or dynamic images to a wide audience, which can help businesses expand their marketing and benefit their profits.
[0003] With the development of large-scale outdoor advertising sign display technology, especially under the influence of various outdoor environments, the fault diagnosis technology of large-scale outdoor advertising sign display is particularly important; the existence of faults is not only in the failure of display screens and other display devices, but also in various hardware, software, network and sensor failures that may cause failures in large-scale outdoor advertising sign displays.
[0004] With the rapid development of intelligence and digitalization, fault diagnosis technology has also made certain progress, but it still needs to be continuously optimized and improved; existing diagnostic technologies often diagnose a classified fault separately, or require manual fault verification, which slows down the progress of fault diagnosis and wastes a lot of time. Summary of the invention
[0005] The purpose of the present invention is to overcome the above problems existing in the prior art and greatly improve its technical effect on the basis of the original technology; firstly, the present invention provides a large-scale outdoor advertising sign display fault intelligent diagnosis and early warning method, the method comprising:
[0006] Large outdoor advertising sign display failures are classified into hardware failures, software failures and network connection failures; the hardware failures include: power failures, display screen failures, sensor failures, fan and cooling system failures; the software failures include: software crashes, freezes and playback errors; the network connection failures include: failures caused by network connection bandwidth, delays and packet loss.
[0007] Set the first threshold function f of various parameters in each fault classification 12 and the second threshold function f 23 ; The first threshold function f 12 The parameters are divided into safe parameters and risky parameters. The second threshold function f 23 The parameters are divided into risky parameters and faulty parameters; the first threshold function f 12 Find the safety threshold of the parameter through the second threshold function f 23Calculate the fault threshold of the parameter; the safety threshold is the dividing point between the safety parameter and the risky parameter, and the fault threshold is the dividing point between the risky parameter and the fault parameter; the first threshold function f of the various parameters in each fault classification is set 12 and the second threshold function f 23 The various parameters are extracted by extracting the determining factors that can reflect the various fault safety states in each fault classification; the first threshold function f 12 The acquisition method is as follows: obtain the historical information of the parameters corresponding to the fault, select the safe parameters and the risky parameters in the historical information of the parameters according to the physical characteristics of the parameters; assign the parameters according to the size of the selected parameters, if the parameters are continuous, assign continuous values; if the parameters are discontinuous, assign classified values; for the continuous parameters and discontinuous parameters, respectively, obtain the first threshold function f by using the Newton interpolation algorithm and the BP neural network algorithm. 12 ; The second threshold function f 23 The acquisition method is as follows: obtain the historical information of the parameters corresponding to the fault, select the risky parameters and fault parameters in the historical information of the parameters according to the physical characteristics of the parameters; assign values to the parameters according to the sizes of the selected parameters, if the parameters are continuous, assign continuous values; if the parameters are discontinuous, assign classified values; obtain the first threshold function f for the continuous parameters and discontinuous parameters respectively by Newton interpolation algorithm and BP neural network algorithm. 23 ; The first threshold function f 12 Find the safety threshold of the parameter through the second threshold function f 23 Determining the fault threshold of the parameter includes: if it is a continuous parameter, then by calculating the first threshold function f 12 The safety threshold is determined by the intersection of the safety parameter and the risk parameter; if it is a discontinuous parameter, the first threshold function f is calculated 12 , that is, the input and output of the BP neural network model determine the safety threshold; the input and output of the BP neural network model determine the safety threshold means: according to the parameters of the adjacent numerical values input multiple times in the BP neural network model, the type of the parameter is judged by the output, and the safety threshold is obtained by combining the parameters of the adjacent numerical values input and the output; similarly, through the second threshold function f 23 The steps to find the fault threshold of the parameter are: if it is a continuous parameter, then by calculating the second threshold function f 23 The fault threshold is determined by the intersection of the risk parameter and the fault parameter; if it is a discontinuous parameter, the second threshold function f is calculated 23, that is, the input and output of the BP neural network model determine the fault threshold; the input and output of the BP neural network model determine the fault threshold means: according to the parameters of adjacent numerical values input multiple times in the BP neural network model, the type of parameters is judged by the output, and the fault threshold is obtained by combining the parameters of adjacent numerical values input and the output.
[0008] Collect the real-time parameters of each fault classification in real time, and bring the real-time parameters into the corresponding first threshold function f 12 and the second threshold function f 23 , determine the safety type of the real-time parameter, the safety type includes: safety, risk and failure, corresponding to the parameters belonging to safety parameters, risk parameters and failure parameters respectively; the real-time parameters are respectively brought into the corresponding first threshold function f 12 and the second threshold function f 23 The method comprises: determining a first threshold function f by analyzing the physical characteristics of the parameter and historical data. 12 The domain range of the real-time parameter is within the first threshold function f 12 Within the definition domain of , the real-time parameters are substituted into the first threshold function f 12 , judge the security type of the parameter according to the result; if the real-time parameter is not within the first threshold function f 12 Within the definition domain of , the real-time parameters are substituted into the second threshold function f 23 , determine the security type of the parameter based on the result.
[0009] A large-scale outdoor advertising sign display fault warning system is produced, and the large-scale outdoor advertising sign display fault warning system includes: a warning framework, warning information and whether to alarm; the warning framework is a multi-branch tree structure, the warning information is a hash value of the security type of each parameter, and whether to alarm refers to selecting whether to alarm through the value of the multi-branch tree root node; each branch position of the multi-branch tree structure is pre-filled with a hash value of the security type of a fixed parameter, and the security type of the parameter is assigned a value of 1 for a security parameter, a value of 2 for a risky parameter, and a value of 3 for a fault parameter; the warning framework is a multi-branch tree structure including: the multi-branch tree structure refers to a multi-branch tree framework, and the multi-branch tree framework is composed of multiple forms of frameworks, and one of the frameworks can be selected; the multi-branch tree framework is set to write the hash value of the security type of the fixed parameter at the corresponding branch position in real time, and the root value generated in real time can determine whether the large-scale outdoor advertising sign display is safe.
[0010] According to the hash value of the root node of the multi-branch tree, it is judged whether the large outdoor advertising sign is displayed safely; if it is safe, the monitoring is continued; if there is a risk, an alarm is issued, and the technicians choose whether to repair it; if there is a fault, an alarm is issued, and the technicians perform repairs; when the safety type of all parameters is 1, it means that the large outdoor advertising sign is displayed as safe; when the safety type of an individual parameter is 2, and the safety types of other parameters are all 1, it means that the large outdoor advertising sign is displayed as a risk; when the safety type of an individual parameter is 3, it means that the large outdoor advertising sign is displayed as a fault; the said judging whether the large outdoor advertising sign is displayed safely according to the hash value of the root node of the multi-branch tree includes: the fixed branch node of the multi-branch tree writes the hash value of the real-time safety type of the fixed parameter, so that the number of hash values of the root node is limited; However, since the number of parameters for various fault classifications exceeds ten, the number of different hash values of the root node is huge, and it is not easy to determine the corresponding position by simple comparison; therefore, when judging whether the large outdoor advertising sign is displayed safely by the hash value of the root node of the multi-branch tree, it is only necessary to record the root hash value of the fixed parameters written into the fixed branch nodes whose real-time safety type is all 1, to determine that the large outdoor advertising sign display at this time is free of risks and faults; there is no need to record the root hash value of any fixed parameter written into the fixed branch nodes whose real-time safety type is not 1; if the large outdoor advertising sign is displayed as having risks, then the risky parameters are determined by performing a hash search on the multi-branch tree, and then the cause of the risk is determined; if the large outdoor advertising sign is displayed as having faults, then the faulty parameters are determined by performing a hash search on the multi-branch tree, and then the cause of the fault is determined.
[0011] The beneficial effects of the present invention are:
[0012] The present invention provides a method for intelligent diagnosis and early warning of large-scale outdoor advertising sign display faults; the method sets a first threshold function f of various parameters in each fault classification. 12 and the second threshold function f 23 The safety type of the parameters is determined respectively, and whether the parameters are safety parameters, risky parameters or fault parameters. Therefore, the method refines the classification of large outdoor advertising sign display failures. By setting up a multi-branch tree early warning system, by writing the hash values corresponding to the parameter safety types of all fault classifications into the early warning framework, it is possible to directly determine in real time through the multi-branch tree root value whether there are risks or faults in the large outdoor advertising sign display, thereby greatly increasing the efficiency and accuracy of fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] Figure 1 The present invention is a flowchart of the method for intelligent diagnosis and early warning of large-scale outdoor advertising sign display faults.
[0014] Figure 2 Schematic diagram of a multi-branch tree structure warning framework displayed for any large outdoor advertising sign.
[0015] Figure 3 This is a schematic diagram of a multi-branch tree structure when a large outdoor advertising sign is displayed as safe.
[0016] Figure 4 A schematic diagram of a multi-branch tree structure when any large outdoor advertising sign is displayed as presenting a risk.
[0017] Figure 5 A schematic diagram of a multi-branch tree structure when any large outdoor advertising sign is displayed as a fault. DETAILED DESCRIPTION
[0018] The specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings; it should be understood that the specific embodiments given here are only used to illustrate and explain the present invention and cannot be used to limit the present invention.
[0019] It should be noted that many specific details are set forth in the following description to facilitate a full understanding of the present invention. However, the present invention may also have other implementations and variations thereof. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.
[0020] like Figure 1 As shown, according to the flowchart of the large outdoor advertising sign display fault intelligent diagnosis and early warning method of the embodiment of the present invention, the flowchart includes: step S100, classifying the large outdoor advertising sign display fault into hardware fault, software fault and network connection fault; the hardware fault includes: power failure, display screen failure, sensor failure, fan and cooling system failure; the software fault includes: software crash, freeze and playback error; the network connection fault includes: fault caused by bandwidth, delay and packet loss of the network connection; step S200, setting the first threshold function f of various parameters in each fault classification 12 and the second threshold function f 23 ; The first threshold function f 12 The parameters are divided into safe parameters and risky parameters. The second threshold function f 23 The parameters are divided into risky parameters and faulty parameters; the first threshold function f 12 Find the safety threshold of the parameter through the second threshold function f 23 Calculate the fault threshold of the parameter; the safety threshold is the dividing point between the safety parameter and the risky parameter, and the fault threshold is the dividing point between the risky parameter and the fault parameter; step S300, collect the real-time parameters of each fault classification in real time, and bring the real-time parameters into the corresponding first threshold function f 12 and the second threshold function f 23, determine the security type of the real-time parameter, the security type includes: safety, risk and failure, corresponding to the parameters belonging to safety parameters, risk parameters and failure parameters respectively; step S400, make a large outdoor advertising sign display fault warning system, the large outdoor advertising sign display fault warning system includes: warning framework, warning information and whether to alarm; the warning framework is a multi-branch tree structure, the warning information is the hash value of the security type of each parameter, and whether to alarm refers to selecting whether to alarm according to the value of the multi-branch tree root node; each branch position of the multi-branch tree structure is pre-filled with the hash value of the security type of the fixed parameter, the security type of the parameter is assigned to 1 as the security parameter, and the security type of the parameter is assigned to 1 as the existence parameter. The risk parameter is assigned a value of 2, and the parameter safety type is assigned a value of 3 for the fault parameter; step S500, based on the hash value of the root node of the multi-branch tree, determine whether the large outdoor advertising sign is displayed safe; if it is safe, continue to monitor; if there is a risk, an alarm is issued, and the technician chooses whether to perform repairs; if there is a fault, an alarm is issued, and the technician performs repairs; when the safety type of all parameters is 1, it means that the large outdoor advertising sign is displayed as safe; when the safety type of individual parameters is 2, and the safety types of other parameters are all 1, it means that the large outdoor advertising sign is displayed as having a risk; when the safety type of individual parameters is 3, it means that the large outdoor advertising sign is displayed as having a fault.
[0021] Among them, in step S100, various types of hardware failures correspond to different failure parameters, for example: power failures include: voltage instability, overvoltage and power overheating, etc.; software crash failures include: code errors, memory management problems and thread management problems, etc.
[0022] In step S200, various parameters are extracted by extracting the determining factors that can reflect various fault safety states in each fault classification; the first threshold function f 12 The acquisition method is as follows: obtain the historical information of the parameters corresponding to the fault, select the safe parameters and the risky parameters in the historical information of the parameters according to the physical characteristics of the parameters; assign the parameters according to the size of the selected parameters, if the parameters are continuous, assign continuous values; if the parameters are discontinuous, assign classified values; for the continuous parameters and discontinuous parameters, respectively, obtain the first threshold function f by using the Newton interpolation algorithm and the BP neural network algorithm. 12 ; The second threshold function f 23The acquisition method is as follows: obtain the historical information of the parameters corresponding to the fault, select the risky parameters and fault parameters in the historical information of the parameters according to the physical characteristics of the parameters; assign values to the parameters according to the sizes of the selected parameters, if the parameters are continuous, assign continuous values; if the parameters are discontinuous, assign classified values; obtain the first threshold function f for the continuous parameters and discontinuous parameters respectively by Newton interpolation algorithm and BP neural network algorithm. 23 .
[0023] By the first threshold function f 12 Method for finding the safety threshold of the parameter: If it is a continuous parameter, then by calculating the first threshold function f 12 The safety threshold is determined by the intersection of the safety parameter and the risk parameter; if it is a discontinuous parameter, the first threshold function f is calculated 12 , that is, the input and output of the BP neural network model determine the safety threshold; the input and output of the BP neural network model determine the safety threshold by: according to the parameters of the adjacent numerical values input multiple times in the BP neural network model, the type of the parameter is judged by the output, and the safety threshold is obtained by combining the parameters of the adjacent numerical values input and the output; through the second threshold function f 23 The steps to find the fault threshold of the parameter are: if it is a continuous parameter, then by calculating the second threshold function f 23 The fault threshold is determined by the intersection of the risk parameter and the fault parameter; if it is a discontinuous parameter, the second threshold function f is calculated 23 , that is, the input and output of the BP neural network model determine the fault threshold; the input and output of the BP neural network model determine the fault threshold means: according to the parameters of adjacent numerical values input multiple times in the BP neural network model, the type of parameters is judged by the output, and the fault threshold is obtained by combining the parameters of adjacent numerical values input and the output.
[0024] In step S300, the real-time parameters are respectively introduced into the corresponding first threshold function f 12 and the second threshold function f 23 The method comprises: determining a first threshold function f by analyzing the physical characteristics of the parameter and historical data. 12 The domain range of the real-time parameter is within the first threshold function f 12 Within the definition domain of , the real-time parameters are substituted into the first threshold function f 12 , judge the security type of the parameter according to the result; if the real-time parameter is not within the first threshold function f 12 Within the definition domain of , the real-time parameters are substituted into the second threshold function f 23 , determine the security type of the parameter based on the result.
[0025] In step S400, the multi-branch tree structure refers to a multi-branch tree framework, which is composed of various forms of frameworks, and one of the frameworks can be selected; the multi-branch tree framework is set to write a hash value of a secure type with fixed parameters at the corresponding branch position in real time, and the root value generated in real time can be used to determine whether the display of the large outdoor advertising sign is safe.
[0026] like Figure 2 As shown, it is a schematic diagram of the multi-branch tree structure warning framework displayed by any large outdoor advertising sign; in this schematic diagram, the bottom layer is the network connection failure parameters, which are network bandwidth, network delay and packet loss parameters, etc.; the second layer from the bottom is the software failure parameters, which are software crash parameters, software freeze parameters and playback error parameters, etc. Since software crash parameters, software freeze parameters and playback error parameters may be composed of multiple factors, these parameters are vectors composed of various factors, that is, most of the software failure parameters are composed of vectors; the third to last layer is the hardware failure parameters, which are display parameters, power parameters and sensor parameters, which are also composed of vectors; the top layer is the root node value; each branch position of the multi-branch tree structure is pre-filled with a hash value of a fixed parameter of a secure type, which means: for example, the branch position corresponding to the bandwidth parameter can only be written with a hash value of the secure type of the bandwidth parameter extracted in real time.
[0027] In step S500, a method for determining whether a large outdoor advertising sign is displayed safely is implemented according to a hash value of a root node of a multi-branch tree: a hash value of a real-time safety type of a fixed parameter is written into a fixed branch node of the multi-branch tree, so that the number of hash values of the root node is limited; however, since the number of parameters of various fault classifications exceeds ten, the number of different hash values of the root node is huge, and it is not easy to determine the corresponding position by simple comparison; therefore, when determining whether a large outdoor advertising sign is displayed safely according to the hash value of the root node of the multi-branch tree, it is only necessary to record the root hash value of the real-time safety type of the fixed parameters written into the fixed branch node as all 1, and determine that the large outdoor advertising sign display at this time does not have risks and faults; it is not necessary to record the root hash value of the real-time safety type of any fixed parameter written into the fixed branch node as not 1; if the large outdoor advertising sign is displayed as having risks, a hash search is performed on the multi-branch tree to determine the parameters having risks, and then determine the cause of the risks; if the large outdoor advertising sign is displayed as having faults, a hash search is performed on the multi-branch tree to determine the parameters having faults, and then determine the cause of the faults.
[0028] like Figure 3 As shown in the figure, it is a schematic diagram of the multi-branch tree structure when the large outdoor advertising sign is displayed as safe; Figure 2 On the basis of Figure 2 The hash value of 1 is written into the data frame corresponding to all parameters in the , and finally the root node value H is obtained k1; When verifying the parameter values extracted in real time, only the root node value is H k1 Only then can it be proved that large outdoor advertising signs are safe.
[0029] like Figure 4 As shown in the figure, it is a schematic diagram of the multi-branch tree structure when any large outdoor advertising sign is displayed as having a risk; Figure 2 On the basis of , the hash value of the security type corresponding to the parameter extracted in real time is written into the corresponding parameter frame. The written values are only the hash value of 1 and the hash value of 2, that is, there are only safe parameters and risky parameters; among them, the hash value of 2 is the last layer packet loss parameter and the software jamming parameter, that is, the parameters at the corresponding position are risky; and the root node value obtained is H k2 , it means that the large outdoor advertising sign display is unsafe and needs to be retrieved; k2 Searching further, we can find that the packet loss parameters and software jamming parameters are risky parameters;
[0030] like Figure 5 As shown in the figure, it is a schematic diagram of the multi-branch tree structure when any large outdoor advertising sign is displayed as a fault; Figure 2 On the basis of , the hash value of the security type corresponding to the parameter extracted in real time is written into the corresponding parameter frame. The written values include 1, 2 and 3, indicating that there are safe parameters and risky parameter failures in the graph; among them, the hash value of 2 is the last layer packet loss parameter and the second-to-last layer software jamming parameter, that is, the parameters at the corresponding position are risky; the hash value of 3 is the second-to-last layer software crash parameter, that is, the software crash parameter is a fault parameter; and the root node value obtained is H k3 , it means that the large outdoor advertising sign display is unsafe and needs to be retrieved; k2 Searching downwards, it can be found that packet loss parameters and software freeze parameters are risky parameters, and software crash parameters are fault parameters.
Claims
1. An intelligent diagnosis and early warning method for display failure of large outdoor advertising signs, characterized in that: The method comprises: S100, classifying the display failure of the large outdoor advertising sign into hardware failure, software failure and network connection failure; the hardware failure includes: power failure, display screen failure, sensor failure, fan and cooling system failure; the software failure includes: software crash, freeze and playback error; the network connection failure includes: failure caused by bandwidth, delay and packet loss of the network connection; S200, setting the first threshold function f of various parameters in each fault classification 12 and the second threshold function f 23 ; The first threshold function f 12 The parameters are divided into safe parameters and risky parameters. The second threshold function f 23 The parameters are divided into risky parameters and faulty parameters; the first threshold function f 12 Find the safety threshold of the parameter through the second threshold function f 23 Determine the fault threshold of the parameter; the safety threshold is the dividing point between the safety parameter and the risky parameter, and the fault threshold is the dividing point between the risky parameter and the faulty parameter; S300, collect real-time parameters of each fault classification in real time, and bring the real-time parameters into the corresponding first threshold function f 12 and the second threshold function f 23 , determine the safety type of the real-time parameter, the safety type includes: safe, risky and faulty, corresponding to the parameter belonging to a safe parameter, a risky parameter and a faulty parameter respectively; S400, manufacturing a large-scale outdoor advertising sign display fault warning system, the large-scale outdoor advertising sign display fault warning system comprising: a warning framework, warning information, and whether to alarm; the warning framework is a multi-branch tree structure, the warning information is a hash value of the security type of each parameter, and whether to alarm refers to selecting whether to alarm according to the value of the multi-branch tree root node; each branch position of the multi-branch tree structure is pre-filled with a hash value of the security type of a fixed parameter, the security type of the parameter is assigned a value of 1 for a security parameter, the security type of the parameter is assigned a value of 2 for a parameter with a risk, and the security type of the parameter is assigned a value of 3 for a fault parameter; S500. Determine whether the large outdoor advertising sign is safe based on the hash value of the root node of the multi-branch tree; if it is safe, continue to monitor; if there is a risk, issue an alarm, and the technicians will choose whether to perform repairs; if there is a fault, issue an alarm, and the technicians will perform repairs; when the safety type of all parameters is 1, it means that the large outdoor advertising sign is safe; when the safety type of some parameters is 2, and the safety types of other parameters are all 1, it means that the large outdoor advertising sign is at risk; when the safety type of some parameters is 3, it means that the large outdoor advertising sign is faulty.
2. The large-scale outdoor advertising sign display fault intelligent diagnosis and early warning method according to claim 1 is characterized in that: The first threshold function f of various parameters in each fault classification is set 12 and the second threshold function f 23 The various parameters are extracted by extracting the determining factors that can reflect the various fault safety states in each fault classification; the first threshold function f 12 The acquisition method is as follows: obtain the historical information of the parameters corresponding to the fault, select the safe parameters and the risky parameters in the historical information of the parameters according to the physical characteristics of the parameters; assign the parameters according to the size of the selected parameters, if the parameters are continuous, assign continuous values; if the parameters are discontinuous, assign classified values; for the continuous parameters and discontinuous parameters, respectively, obtain the first threshold function f by using the Newton interpolation algorithm and the BP neural network algorithm. 12 ; The second threshold function f 23 The acquisition method is as follows: obtain the historical information of the parameters corresponding to the fault, select the risky parameters and fault parameters in the historical information of the parameters according to the physical characteristics of the parameters; assign values to the parameters according to the sizes of the selected parameters, if the parameters are continuous, assign continuous values; if the parameters are discontinuous, assign classified values; obtain the first threshold function f for the continuous parameters and discontinuous parameters respectively by Newton interpolation algorithm and BP neural network algorithm. 23 .
3. The large-scale outdoor advertising sign display fault intelligent diagnosis and early warning method according to claim 1 is characterized in that: The first threshold function f 12 Find the safety threshold of the parameter through the second threshold function f 23 Determining the fault threshold of the parameter includes: if it is a continuous parameter, then by calculating the first threshold function f 12 The safety threshold is determined by the intersection of the safety parameter and the risk parameter; if it is a discontinuous parameter, the first threshold function f is calculated 12 , that is, the input and output of the BP neural network model determine the safety threshold; the input and output of the BP neural network model determine the safety threshold means: according to the parameters of the adjacent numerical values input multiple times in the BP neural network model, the type of the parameter is judged by the output, and the safety threshold is obtained by combining the parameters of the adjacent numerical values input and the output; similarly, through the second threshold function f 23 The steps to find the fault threshold of the parameter are: if it is a continuous parameter, then by calculating the second threshold function f 23 The fault threshold is determined by the intersection of the risk parameter and the fault parameter; if it is a discontinuous parameter, the second threshold function f is calculated 23 , that is, the input and output of the BP neural network model determine the fault threshold; the input and output of the BP neural network model determine the fault threshold means: according to the parameters of adjacent numerical values input multiple times in the BP neural network model, the type of parameters is judged by the output, and the fault threshold is obtained by combining the parameters of adjacent numerical values input and the output.
4. The large-scale outdoor advertising sign display fault intelligent diagnosis and early warning method according to claim 1 is characterized in that: The real-time parameters are respectively brought into the corresponding first threshold function f 12 and the second threshold function f 23 The method comprises: determining a first threshold function f by analyzing the physical characteristics of the parameter and historical data. 12 The domain range of the real-time parameter is within the first threshold function f 12 Within the definition domain of , the real-time parameters are substituted into the first threshold function f 12 , judge the security type of the parameter according to the result; if the real-time parameter is not within the first threshold function f 12 Within the definition domain of , the real-time parameters are substituted into the second threshold function f 23 , determine the security type of the parameter based on the result.
5. The large-scale outdoor advertising sign display fault intelligent diagnosis and early warning method according to claim 1 is characterized in that: The warning framework is a multi-branch tree structure including: the multi-branch tree structure refers to a multi-branch tree framework, which is composed of various forms of frameworks, and you can choose one of the frameworks; the multi-branch tree framework is set to write a hash value of a secure type with fixed parameters in real time at the corresponding branch position, and the root value generated in real time can determine whether the display of large outdoor advertising signs is safe.
6. The large-scale outdoor advertising sign display fault intelligent diagnosis and early warning method according to claim 1 is characterized in that: The method of judging whether the display of the large outdoor advertising sign is safe according to the hash value of the root node of the multi-branch tree includes: writing the hash value of the real-time safety type of the fixed parameter into the fixed branch node of the multi-branch tree, so that the number of hash values of the root node is limited; but because the number of parameters of various fault classifications exceeds ten, the number of different hash values of the root node is huge, and it is not easy to determine the corresponding position by simple comparison; therefore, when judging whether the display of the large outdoor advertising sign is safe according to the hash value of the root node of the multi-branch tree, it is only necessary to record the root hash value of the real-time safety type of the fixed parameters written into the fixed branch node as all 1, and determine that the display of the large outdoor advertising sign at this time is free of risks and faults; there is no need to record the root hash value of the real-time safety type of any fixed parameter written into the fixed branch node that is not 1; if the large outdoor advertising sign is displayed as having risks, then by performing a hash search on the multi-branch tree, the parameters having risks are determined, and then the causes of the risks are determined; if the large outdoor advertising sign is displayed as having faults, then by performing a hash search on the multi-branch tree, the parameters having faults are determined, and then the causes of the faults are determined.