Electronic equipment abnormity early warning method and system based on big data analysis

By calculating the stability and degree of difference in the CART decision tree and dynamically adjusting the pruning threshold, the problem of inaccurate pruning in the decision tree is solved, the accuracy and efficiency of abnormal warning of electronic equipment is improved, and the reliability of the model is ensured. Resource optimization.

CN118981700AInactive Publication Date: 2024-11-19JINAN HAIJI TECH DEV CO LTD
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411471296.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-22
Publication Date
2024-11-19
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The CART decision tree algorithm has inaccuracy during the pruning process, which leads to overfitting the model and affects the accuracy and efficiency of early warning.

Method used

By calculating the stability and degree of difference of nodes in the decision tree, dynamically adjust the pruning threshold, optimize the decision tree structure, avoid overfitting and information loss, and ensure the reliability and efficiency of the model.

Benefits of technology

It improves the accuracy of pruning decisions of the decision tree, reduces the complexity of the model, reduces the consumption of computing resources and storage space, and enhances the reliability and efficiency of the early warning system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118981700A_ABST
    Figure CN118981700A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of data processing, in particular to an electronic equipment abnormity early warning method and system based on big data analysis. The method comprises the following steps: acquiring a decision tree of environment temperature data of space disinfection equipment; calculating node stability according to the stability feature and the stability weight of the data in any node; calculating the density of the nodes according to the distance between the data in the nodes and the nodes; calculating the difference degree between the node and other nodes according to the difference between the center data of the node and the center data of the other nodes and the node density of the other nodes; adjusting a node threshold according to the node stability and the difference degree between the node and other nodes; and according to the obtained final decision tree, carrying out abnormity early warning on the environment temperature data of the space disinfection equipment. According to the invention, abnormity early warning of the environment temperature data of the space disinfection equipment is more accurate.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to an electronic equipment abnormality early warning method and system based on big data analysis. Background Art

[0002] Space disinfection equipment is an electronic device widely used in public places such as hospitals, schools, and shopping malls. Its disinfection effect is directly related to public health and safety. Temperature, as an important factor affecting the disinfection effect, can directly affect the volatilization rate of the disinfectant, the disinfection effect, and the operating stability of the equipment. For example, some disinfectants may lose their disinfection activity in a low-temperature environment, affecting the disinfection effect. In a high-temperature environment, some chemical components may decompose, resulting in a decrease in the effect of the disinfectant. By monitoring the ambient temperature of the place where the disinfection equipment is used, managers can adjust the type and amount of disinfectants in time to ensure that the disinfection effect is maximized and the disinfection standards are met. At the same time, temperature monitoring can also provide data support for the management of disinfection work, help formulate a reasonable disinfection plan, and avoid waste of resources. For example, in the disinfection operation of a large hospital, the research team integrated temperature sensors on the disinfection equipment to monitor the ambient temperature of the operating room and public areas in real time. The monitoring data showed that the temperature in the operating room during peak hours was as high as 28°C, while the effective temperature range of the disinfectant was 20-25°C. Through the analysis of the data, the management personnel adjusted the concentration of the disinfectant in time and properly maintained the equipment to ensure the disinfection effect and safe operation of the equipment. Therefore, monitoring the ambient temperature of space disinfection equipment is not only an important means to improve the disinfection effect, but also the key to ensuring the safe operation of equipment and optimal allocation of resources.

[0003] As a classification algorithm, the CART decision tree algorithm can distinguish abnormal data from normal data to realize abnormal warning of equipment. For example, the Chinese patent application document with publication number CN116432123A discloses a method for early warning of electric energy meter faults based on the CART decision tree algorithm. The electric energy meter fault early warning model is constructed by the CART decision tree algorithm, and the preliminary early warning model is trained using the training set. Based on the constructed decision tree model, the innovative MMP pruning algorithm is used to select the fault with the minimum The subtree of the value is pruned, where The value can measure the degree of fit of the training data and the complexity of the model. However, the patent application document selects the minimum Pruning the subtree with the minimum value may cause the model to overfit the training data. Although it fits the training set well, it may perform poorly on the test set or new data. Values ​​may need to be compared between multiple subtrees, which can affect model efficiency when there are many nodes. Summary of the invention

[0004] In order to solve the problem of inaccurate pruning during decision tree construction, the present invention provides an electronic equipment abnormality warning method and system based on big data analysis.

[0005] In a first aspect, the present invention provides an electronic device abnormality warning method based on big data analysis, which adopts the following technical solution: Obtain ambient temperature data of space disinfection equipment and build a decision tree; During the decision tree construction process, calculate the stability of any of the constructed decision trees , For the The stability of the nodes, , Respectively The first node Data, The stable characteristics of the data, For the The first node The stable weight of each data, For the The number of data in the node; The first node The stability characteristics of the data and the Data and Nodes except The numerical differences of the rest of the data are negatively correlated, and the stability weight is negatively correlated with the distance between each data and the central data of the node; Calculate the difference between any node and other nodes , For the The difference between a node and other nodes, , Respectively Node, The central data of each node, No. The node density of nodes, is the number of nodes in the constructed decision tree; the node density is the negative correlation between the distance between all data in the node and the central data of the node; According to the ratio of node stability and degree of difference, the initial threshold of the decision tree is corrected to obtain the corrected node threshold. The decision tree is adjusted according to the corrected node threshold. Based on the adjusted decision tree, an abnormal warning is issued for the ambient temperature data of the space disinfection equipment.

[0006] The present invention dynamically adjusts the threshold value for node pruning by the stability of the node and the degree of difference from other nodes. This dynamism makes the pruning process more detailed, can better balance the degree of fitting and model complexity, improves the accuracy of pruning decisions, avoids decision tree imbalance and information loss caused by pruning, and thus ensures the reliability of model warning.

[0007] Preferably, the stability characteristic satisfies the expression: ; In the formula, for The first node The stable characteristics of the data, For the The first node The value of the data, For the Nodes except In addition to the data The value of the data, For the The number of data in each node, For The exponential function with base , is the absolute value symbol. For the The first node The data and The numerical difference of the data.

[0008] Using the mean of the differences between data to characterize node stability features helps to develop a more targeted pruning strategy. This approach can more finely identify which nodes are stable and which nodes may be unstable due to data variation, thereby adjusting the model structure more intelligently during the pruning process.

[0009] Preferably, the central data of the node is: the mean value of all the data in the node.

[0010] Preferably, the method for obtaining the stable weight of the data is: According to the distance between each data and the central data of the node, the initial weight of each data is calculated. The first node The initial weight of the data Satisfies the expression: ; In the formula, For the The first node The initial weight of the data, For the The first node The value of the data, For the The central data of each node, is the absolute value symbol; For the The first node The data and The distance of the central data of each node; According to the initial weights of all data in the node, the stable weights of each data are obtained to satisfy the expression: ; In the formula, For the The first node The stable weight of each data, For the The first node The initial weight of the data, For the The first node The data and The initial weight of the central data of each node, For the The number of data in a node.

[0011] By calculating the difference between each data and the node center data, the stability impact of the point on the overall characteristics of the node can be quantified. The introduction of the reciprocal makes the data closer to the center data in the node have a higher weight, and vice versa. Emphasizing the closeness to the center data can more effectively identify stable features.

[0012] Preferably, the node density satisfies the expression: ; In the formula, For the The node density, For the The first node The value of the data, For the The central data of each node, For the The number of data in nodes; among them, For the The first node The data and The distance between the center data of each node.

[0013] The difference between the data in the node and the central data reflects the deviation of the data from the node center. Using the inverse to represent this difference can effectively quantify the degree of data aggregation within the node. At the same time, in the process of decision tree pruning, the importance of the node can be evaluated by using the node density. Low-density nodes may contain a lot of noise or unimportant information. By calculating the density, it is possible to better decide on which nodes to prune, helping to improve the generalization ability of the model.

[0014] Preferably, the modified node threshold satisfies the expression: ; In the formula, For the The node threshold after the nodes are corrected, For the The stability of the nodes, For the The difference between a node and other nodes, is the initial threshold, is the logarithmic function with base 2, Preset hyperparameters.

[0015] By introducing the stability of the node and the degree of difference from other nodes, the threshold of each node can be adjusted according to its specific situation instead of using a fixed threshold. This makes the model more adaptable to different features and data distributions, and improves the modeling ability for complex data.

[0016] Preferably, the adjusting decision tree comprises the steps of: The information gain before and after the node split is calculated. If the information gain before and after the node split is lower than the node threshold after the node is corrected, it is pruned to obtain the adjusted decision tree.

[0017] Preferably, the abnormal warning of the ambient temperature data of the space disinfection equipment includes the steps of: If the decision tree contains less than If a node with data is found, the node is an abnormal node, the system triggers the early warning mechanism, the space disinfection equipment displays abnormal ambient temperature data, and reminds relevant personnel to handle it; among them, The number of data preset in the node.

[0018] In a second aspect, the present invention provides an electronic device abnormality warning system based on big data analysis, comprising a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned electronic device abnormality warning method based on big data analysis is implemented.

[0019] By adopting the above technical solution, the above-mentioned electronic device abnormality warning method based on big data analysis is generated into a computer program and stored in a memory to be loaded and executed by a processor, so that a terminal device is made according to the memory and the processor for easy use.

[0020] The present invention has the following technical effects: Through more sophisticated pruning control, unnecessary node splits are reduced, thereby simplifying the decision tree structure and reducing model complexity. This not only helps to improve prediction speed, but also reduces the consumption of computing resources and storage space, making the model more suitable for deployment and application in resource-constrained environments.

[0021] Accurate and timely response is crucial for the abnormal warning system of space disinfection equipment. Dynamic adjustment of pruning thresholds can ensure that the model can still accurately capture key information when the environment changes, issue warnings in a timely manner, reduce false alarms and missed alarms, and improve the overall reliability and efficiency of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0023] Figure 1 This is a method flow chart of an electronic device abnormality warning method based on big data analysis in an embodiment of the present invention. DETAILED DESCRIPTION

[0024] 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 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0025] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0026] The embodiment of the present invention discloses an electronic device abnormality warning method based on big data analysis, referring to Figure 1, comprising steps S1 to S6: S1: Obtain the ambient temperature data of the space disinfection equipment and build a decision tree.

[0027] It should be noted that the ambient temperature data of the space disinfection equipment refers to the ambient temperature in the place where the space disinfection equipment is used. By issuing abnormal warnings for the ambient temperature data of the space disinfection equipment, the impact of abnormal ambient temperature on the activity and effect of the disinfectant can be detected. When the ambient temperature data of the space disinfection equipment is abnormal, it will affect the activity of the disinfectant, and relevant personnel will need to readjust the concentration of the disinfectant to ensure the effective implementation of space disinfection.

[0028] In one embodiment, a temperature sensor is used to collect ambient temperature data of the room disinfection equipment, and the preset collection period is hours, the collection frequency is Hz (Hertz), a decision tree is constructed for the collected ambient temperature data of the space disinfection equipment. Among them, the decision tree is constructed by the CART decision tree algorithm, which is a well-known technology and will not be described in detail here.

[0029] It should be noted that It is a preset collection period, which is not specifically limited in the present invention and can be adjusted by the implementer according to the actual situation. For example, .

[0030] At this point, the collection of ambient temperature data of the space disinfection equipment is completed.

[0031] S2: Calculate the stability of any node in the decision tree.

[0032] It should be noted that the stability of a node reflects the consistency of the data within the node. When the stability is high, the node contributes more to the decision-making, and a higher pruning threshold can be set to retain these important branches; when the stability is low, the threshold can be lowered, pruning can be performed more actively, and unnecessary complexity can be reduced. Therefore, it is necessary to adjust the threshold for pruning the node according to the node stability.

[0033] In one embodiment, for any node, the stability of the node is calculated according to the stability characteristics and stability weight of the data in the node. The specific calculation formula is as follows: ; In the formula, For the The stability of the nodes, , For the The first node Data, The stable characteristics of the data, For the The first node The stable weight of each data, For the The number of data in the nodes. For the The stable feature weighted mean of the data in nodes, For the The smaller the value of the stable characteristic variance of the data in a node, the higher the stability of the data in the node, that is, the higher the stability of the node.

[0034] Among them, The first node Stable characteristics of the data The specific calculation formula is as follows: ; In the formula, For the The first node The stable characteristics of the data, For the The first node The value of the data, For the Nodes except In addition to the data The value of the data, For the The number of data in each node, For The exponential function with base , is the absolute value symbol. For the The first node The data and The numerical difference of the data; For the The first node The smaller the value is, the smaller the difference between the data and the rest of the data in the node except itself is, the smaller the difference between the data and the rest of the data in the node except itself is, and the higher the stability characteristic of the data is.

[0035] Among them, The central data of each node is: The mean value of all data in a node.

[0036] In another embodiment, The central data of a node may also be represented by the median of the numerical values ​​of the data in the node.

[0037] Among them, The first node The stable weight of the data The method to obtain is: According to the distance between each data and the central data of the node, the initial weight of each data is calculated. The first node The initial weight of the data Satisfies the expression: ; In the formula, For the The first node The initial weight of the data, For the The first node The value of the data, For the The central data of each node, is the absolute value symbol; For the The first node The data and The distance between the center data of the node, The first node The data and The smaller the distance between the center data of the nodes, the The first node The higher the initial weight of the data.

[0038] According to the initial weights of all data in the node, the stable weights of each data are obtained to satisfy the expression: ; In the formula, For the The first node The stable weight of each data, For the The first node The initial weight of the data, For the The first node The data and The initial weight of the central data of each node, For the The number of data in nodes; among them, For the The first node The normalization process of the initial weights of the data.

[0039] In another embodiment, The first node The initial weight of the data It can also be expressed as: ; In the formula, For the The first node The initial weight of the data, For the The first node The value of the data, For the The central data of each node, is the absolute value symbol, For An exponential function with base .

[0040] At this point, the stability calculation of any node in the decision tree is completed.

[0041] S3: Calculate the degree of difference between any node in the decision tree and other nodes.

[0042] It should be noted that in a decision tree, as the depth of the tree increases, the difference between nodes usually decreases gradually. When the difference between two nodes is very small, it means that the data sets they represent are very similar. Further splitting these nodes may not bring significant improvements in classification or prediction performance, but may increase the complexity of the model and lead to overfitting. In other words, in the process of adjusting the node threshold during node pruning, combined with the degree of difference between the node and other nodes, the accuracy of the node threshold can be further improved.

[0043] In one embodiment, the calculation formula for the difference between any node and other nodes is as follows: ; In the formula, For the The difference between a node and other nodes, , Respectively Node, The central data of each node, For the The node density of nodes, is the number of nodes in the constructed decision tree. For the The node and The central data difference of the nodes is larger. The node and The greater the difference between the nodes, the greater the The node density of nodes The larger the The less noise information a node contains, the better the The node and The greater the node density difference among nodes, the higher the reliability.

[0044] Among them, The node density of nodes The calculation formula is as follows: ; In the formula, For the The node density of nodes, For the The first node The value of the data, For the The central data of each node, For the The number of data in nodes; among them, For the The first node The data and The distance between the center data of the node At this point, the degree of difference between any node in the decision tree and other nodes has been calculated.

[0045] S4: Calculate the modified node threshold of any node according to the ratio of the stability of the node in the decision tree and the degree of difference with other nodes.

[0046] It should be noted that, in the process of correcting the threshold when pruning any node, if the stability of the node is higher and the difference between the node and other nodes is smaller, it means that the classification effect of the node is better and the necessity of not performing the next classification is greater, that is, the necessity of pruning is higher. In other words, a higher threshold needs to be set for the node to ensure that the information gain of the node after classification is as small as possible and is not reclassified.

[0047] In one embodiment, the calculation formula for the sth modified node threshold is as follows: ; In the formula, is the sth corrected node threshold, For the The stability of the nodes, For the The difference between a node and other nodes, is the initial threshold, is the logarithmic function with base 2, Preset hyperparameters.

[0048] It should be noted that It is a preset initial threshold value, which is not specifically limited in the present invention, but the value range of the initial threshold value is (0,1], and the implementer can adjust it according to the actual situation, for example , To preset hyperparameters, in the present invention .

[0049] At this point, the modified node threshold of any node in the decision tree is calculated.

[0050] S5: Adjust the decision tree.

[0051] In one embodiment, the information gain before and after the node split is calculated based on the revised node threshold of each node. If the information gain before and after the node split is lower than the revised node threshold of the node, pruning is performed on the node to obtain the final CART decision tree.

[0052] It should be noted that the calculation of information gain before and after node splitting is a well-known technique and will not be described in detail here.

[0053] S6: Based on the adjusted decision tree, abnormal warnings are issued for the ambient temperature data of the space disinfection equipment.

[0054] In one embodiment, if the number of data contained in the decision tree is less than If a node with data is detected, the node is considered to be an abnormal node, that is, an abnormality is detected in the ambient temperature data of the space disinfection equipment. At this time, the system triggers the early warning mechanism and reminds relevant personnel to handle it.

[0055] It should be noted that is the number of data preset in the node (the number of data preset is limited to a positive integer). The present invention does not specifically limit it, and the implementer can adjust it according to the actual situation. For example .

[0056] Through the above steps, the abnormal warning of the ambient temperature data of the space disinfection equipment is completed.

[0057] An embodiment of the present invention also discloses an electronic device abnormality warning system based on big data analysis, including a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, an electronic device abnormality warning method based on big data analysis according to the present invention is implemented.

[0058] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0059] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory, a dynamic random access memory, a static random access memory, an enhanced dynamic random access memory, a high bandwidth memory, a hybrid storage cube, etc., or any other medium that can be used to store the required information and can be accessed by an application, a module, or both. Any such computer storage medium may be part of the device or accessible or connectable to the device.

[0060] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0061] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. An electronic equipment abnormality warning method based on big data analysis, characterized in that: include: Obtain ambient temperature data of space disinfection equipment and build a decision tree; During the decision tree construction process, calculate the stability of any node in the constructed decision tree , For the The stability of the nodes, , Respectively The first node Data, The stable characteristics of the data, For the The first node The stable weight of each data, For the The number of data in the node; The first node The stability characteristics of the data and the Data and Nodes except The numerical differences of the rest of the data are negatively correlated, and the stability weight is negatively correlated with the distance between each data and the central data of the node; Calculate the difference between any node and other nodes , For the The difference between a node and other nodes, , Respectively Node, The central data of each node, For the The node density of nodes, is the number of nodes in the constructed decision tree; the node density is negatively correlated with the distance between all data in the node and the central data of the node; According to the ratio of node stability and degree of difference, the initial threshold of the decision tree is corrected to obtain the corrected node threshold. The decision tree is adjusted according to the corrected node threshold. Based on the adjusted decision tree, an abnormal warning is issued for the ambient temperature data of the space disinfection equipment.

2. According to claim 1, the electronic equipment abnormality early warning method based on big data analysis is characterized in that: The stability characteristics satisfy the expression: ; In the formula, For the The first node The stable characteristics of the data, For the The first node The value of the data, For the Nodes except In addition to the data The value of the data, For the The number of data in each node, For The exponential function with base , is the absolute value symbol, where For the The first node Data and The numerical difference of the data.

3. The electronic equipment abnormality early warning method based on big data analysis according to claim 1 is characterized in that: The central data of the node is the mean value of all the data in the node.

4. The electronic equipment abnormality early warning method based on big data analysis according to claim 1 is characterized in that: The method for obtaining the stable weight is: According to the distance between each data and the central data of the node, the initial weight of each data is calculated. The first node The initial weight of the data Satisfies the expression: ; In the formula, For the The first node The initial weight of the data, For the The first node The value of the data, For the The central data of each node, is the absolute value symbol; For the The first node Data and The distance of the central data of each node; According to the initial weights of all data in the node, the stable weights of each data are obtained to satisfy the expression: ; In the formula, For the The first node The stable weight of each data, For the The first node The initial weight of the data, For the The first node The data and The initial weight of the central data of each node, For the The number of data in a node.

5. The electronic equipment abnormality early warning method based on big data analysis according to claim 1 is characterized in that: The node density satisfies the expression: ; In the formula, For the The node density, For the The first node The value of the data, For the The central data of each node, For the The number of data in nodes; among them, For the The first node Data and The distance between the center data of each node.

6. The electronic equipment abnormality early warning method based on big data analysis according to claim 1 is characterized in that: The modified node threshold satisfies the expression: ; In the formula, For the The node threshold after the nodes are corrected, For the The stability of the nodes, For the The difference between a node and other nodes, is the initial threshold, is the logarithmic function with base 2, Preset hyperparameters.

7. The electronic equipment abnormality early warning method based on big data analysis according to claim 1 is characterized in that: The adjustment decision tree comprises the steps of: The information gain before and after the node split is calculated. If the information gain before and after the node split is lower than the node threshold after the node is corrected, it is pruned to obtain the adjusted decision tree.

8. The electronic equipment abnormality early warning method based on big data analysis according to claim 1 is characterized in that: The abnormal warning of the ambient temperature data of the space disinfection equipment includes the following steps: If the decision tree contains less than If a node with data is found, the node is an abnormal node, the system triggers the early warning mechanism, the space disinfection equipment displays abnormal ambient temperature data, and reminds relevant personnel to handle it; among them, The number of data preset in the node.

9. An electronic equipment abnormality warning system based on big data analysis, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, an electronic device abnormality warning method based on big data analysis according to any one of claims 1 to 8 is implemented.

Citation Information

Cited By

  • Multi-dimensional dynamic monitoring and early warning system and method for dangerous environment of explosion-proof unit

    CN120808557A

  • Multi-dimensional dynamic monitoring and early warning system and method for hazardous environments of explosion-proof units

    CN120808557B