Management platform for optimizing ladder networking device identity and data by using HyperledgerFabric

Abstract

The invention relates to the technical field of data transmission, and discloses a management platform for optimizing elevator networking equipment identity and data by using HyperledgerFabric, comprising an elevator abnormity identification unit which distributes a unique identity label for each elevator equipment, collects elevator equipment operation data, identifies abnormal data and obtains a final risk prediction value through a HyperledgerFabric block chain network; the maintenance request generation unit is used for collecting maintenance request book templates, calculating template characteristic values according to keywords, dividing template grades, determining template grades corresponding to the maintenance request book templates, determining the maintenance request book templates according to the matching degree of each template, and generating a maintenance request book; and the maintenance request sending unit is used for determining maintenance personnel and sending the maintenance request book to the maintenance personnel based on the block chain network. According to the invention, credible distribution is carried out based on the block chain network, and the transparency, timeliness and non-repudiation of maintenance task distribution are ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data transmission, in particular to a Hyperledger Fabric-based optimization of elevator networking equipment identity and data management platform. BACKGROUND

[0002] With the wide application of elevator equipment in residential, commercial and industrial buildings, the operation safety and maintenance efficiency of elevators have gradually become the focus of public attention. Traditional elevator maintenance relies on regular manual inspection and paper records, which has problems such as non-uniform equipment identity, difficulty in sharing operation data, information lag, and easy tampering of human records, and is difficult to meet the intelligent and refined requirements of modern city elevator safety management.

[0003] Although the Internet of Things technology has been gradually applied to the collection of elevator operation status in recent years, there is a lack of unified identity mechanism and trusted data platform, resulting in insufficient credibility of equipment data, making it difficult to efficiently cooperate between multiple management subjects. At the same time, abnormal judgment still relies on simple threshold judgment or rule system, which is difficult to cope with complex fault modes. Most of the existing maintenance requests are manually processed, and it is impossible to automatically generate matching maintenance request forms according to the actual operation risk of the elevator, affecting the response speed and maintenance quality.

[0004] Blockchain technology, especially Hyperledger Fabric consortium chain, has the characteristics of data tamper-proof, identity traceable, and controllable permissions, providing a new means for elevator equipment identity registration and operation data trusted management.

[0005] Therefore, it is necessary to design a Hyperledger Fabric-based optimization of elevator networking equipment identity and data management platform to solve the problems existing in the prior art. SUMMARY

[0006] In view of this, the present application provides a Hyperledger Fabric-based optimization of elevator networking equipment identity and data management platform, aiming to solve the problems of lack of unified identity management mechanism for elevator equipment, leading to difficulty in sharing equipment information across systems, tracking difficulty, low credibility of elevator operation data, easy tampering, inability to provide reliable basis for subsequent risk assessment and maintenance, insufficient means for identifying abnormal elevator data and risk assessment, inability to effectively quantify elevator state risk, and over-reliance on manual processing in existing maintenance request generation method, difficulty in accurate matching with actual operation risk, and untimely response.

[0007] The present application provides a Hyperledger Fabric-based optimization of elevator networking equipment identity and data management platform, comprising:

[0008] The elevator anomaly identification unit assigns a unique identity to each elevator device through a Hyperledger Fabric blockchain network, collects running data of each elevator device and stores it in the blockchain network, identifies abnormal data in the running data of the elevator device, and obtains an initial risk prediction value of the elevator device based on a random forest model; the initial risk prediction value is compared with a risk prediction threshold value, and when the initial risk prediction value is greater than or equal to the risk prediction threshold value, the initial risk prediction value is the final risk prediction value; when the initial risk prediction value is less than the risk prediction threshold value, the running frequency of the elevator in a set time period is identified in the running data of the elevator device, and the initial risk prediction value is adjusted according to the running frequency to obtain the final risk prediction value;

[0009] The maintenance request generation unit collects a plurality of maintenance request templates, extracts the keywords of each maintenance request template, calculates the template feature value according to the keywords, and divides the plurality of maintenance request templates into template levels according to the template feature value; the risk interval is confirmed according to the final risk prediction value; the template level corresponding to the maintenance request template is determined according to the risk interval, and after the corresponding template level is determined, the matching degree of the abnormal data and each template in the corresponding template level is obtained, the maintenance request template is determined according to the matching degree, the abnormal data is filled into the maintenance request template, and a maintenance request is generated.

[0010] The maintenance request sending unit determines the maintenance personnel according to the risk interval, and sends the maintenance request to the maintenance personnel based on the blockchain network.

[0011] Further, when the elevator anomaly identification unit obtains the initial risk prediction value of the elevator device based on the random forest model, it includes:

[0012] The elevator anomaly identification unit obtains historical running data in the running data, and divides the historical running data into a training set and a test set;

[0013] The best parameters of the random forest model are found by using cross-validation combined with grid search, and the random forest model is established;

[0014] The training set is used to fit the random forest model;

[0015] The test set is brought into the random forest model and the accuracy of the risk prediction value is calculated;

[0016] When the accuracy reaches a preset accuracy threshold, the initial risk prediction value of the current elevator device is obtained according to the abnormal data.

[0017] Furthermore, when the elevator anomaly identification unit identifies the elevator operating frequency within a set time period in the elevator equipment operating data and adjusts the initial risk prediction value according to the operating frequency to obtain the final risk prediction value, it includes:

[0018] The elevator anomaly identification unit compares the elevator operating frequency within the set time period with the operating frequency threshold. The operating frequency threshold includes a first operating frequency threshold and a second operating frequency threshold, wherein the first operating frequency threshold is less than the second operating frequency threshold.

[0019] When the operating frequency is less than the first operating frequency threshold, the elevator anomaly identification unit determines that the initial risk prediction value will not be adjusted and uses the initial risk prediction value as the final risk prediction value.

[0020] When the operating frequency is greater than or equal to the first operating frequency threshold and less than the second operating frequency threshold, the elevator anomaly identification unit determines to adjust the initial risk prediction value using the first adjustment coefficient to obtain the adjusted risk prediction value, and uses the adjusted risk prediction value as the final risk prediction value.

[0021] When the operating frequency is greater than or equal to the second operating frequency threshold, the elevator anomaly identification unit determines to adjust the initial risk prediction value using the second adjustment coefficient to obtain the adjusted risk prediction value, and uses the adjusted risk prediction value as the final risk prediction value.

[0022] Furthermore, when the maintenance request generation unit extracts keywords from each maintenance request template and calculates template feature values ​​based on the keywords, it includes:

[0023] The maintenance request generation unit calculates the term frequency and inverse document frequency of the keywords in the maintenance request template, and obtains the TF-IDF value of each keyword based on the term frequency and inverse document frequency.

[0024] The template feature value is obtained by weighted averaging of the TF-IDF values ​​of the keywords.

[0025] Furthermore, when the maintenance request generation unit classifies the several maintenance request templates into template levels based on the template feature values, it includes:

[0026] S1: Initialize K centroids, and assign each maintenance request template to the nearest centroid to form K clusters;

[0027] S2: Recalculate the centroid of each cluster;

[0028] S3: Repeat S1 and S2 until the centroid no longer changes or the preset number of repetitions is completed;

[0029] S4: Use the centroid of each cluster as the eigenvalue of the cluster.

[0030] Furthermore, when the maintenance request generation unit classifies the several maintenance request templates into template levels based on the template feature values, it also includes:

[0031] The maintenance request generation unit compares the feature value of the cluster with a first preset feature value and a second preset feature value respectively. The first preset feature value is less than the second preset feature value. Based on the comparison result, several maintenance request templates are divided into template levels.

[0032] When the feature value is less than or equal to the first preset feature value, the maintenance request template in the cluster is classified into the first template level;

[0033] When the feature value is greater than the first preset feature value and less than or equal to the second preset feature value, the maintenance request template in the cluster is classified into the second template level;

[0034] When the feature value is greater than the second preset feature value, the maintenance request template in the cluster is classified into the third template level;

[0035] The first template level indicates that the risk value in the maintenance request template is lower than the second template level, and the second template level indicates that the risk value in the maintenance request template is lower than the third template level.

[0036] Furthermore, when the maintenance request generation unit confirms the risk range based on the final risk prediction value, it includes:

[0037] The risk range includes a first risk range, a second risk range, and a third risk range;

[0038] When the final risk prediction value is less than or equal to the first risk prediction threshold, the maintenance request generation unit determines that the final risk prediction value is within the first risk range.

[0039] When the final risk prediction value is greater than the first risk prediction threshold and less than or equal to the second risk prediction threshold, the maintenance request generation unit determines that the final risk prediction value is within the second risk range.

[0040] When the final risk prediction value is greater than the second risk prediction threshold, the maintenance request generation unit determines that the final risk prediction value is within the third risk range;

[0041] The risk value in the first risk interval is lower than that in the second risk interval, and the risk value in the second risk interval is lower than that in the risk interval.

[0042] The first risk prediction threshold is less than the second risk prediction threshold.

[0043] Furthermore, the maintenance request generation unit determines the template level corresponding to the maintenance request template based on the risk range, including:

[0044] When the final risk prediction value is within the first risk range, the maintenance request generation unit determines the template level corresponding to the maintenance request template as the first template level.

[0045] When the final risk prediction value is within the second risk range, the maintenance request generation unit determines the template level corresponding to the maintenance request template as the second template level;

[0046] When the final risk prediction value is within the third risk range, the maintenance request generation unit determines the template level corresponding to the maintenance request template as the third template level.

[0047] Furthermore, the matching degree between the abnormal data and each template within the corresponding template level is obtained. When the maintenance request generation unit determines the maintenance request template based on the matching degree, it includes:

[0048] The maintenance request generation unit fills the abnormal data into each template within the corresponding template level one by one, and determines the matching degree by the proportion of abnormal data filled into each template.

[0049] Arrange the matching degree of each template within the corresponding template level from high to low, and select the maintenance request template with the highest matching degree.

[0050] Furthermore, when the maintenance request sending unit determines maintenance personnel based on the risk range, it includes:

[0051] The maintenance personnel include junior maintenance personnel, intermediate maintenance personnel, and senior maintenance personnel;

[0052] When the final risk prediction value is within the first risk range, the maintenance request sending unit determines that the identified maintenance personnel is the primary maintenance personnel;

[0053] When the final risk prediction value is within the second risk range, the maintenance request sending unit determines that the maintenance personnel are the intermediate maintenance personnel;

[0054] When the final risk prediction value is within the third risk range, the maintenance request sending unit determines that the maintenance personnel are the senior maintenance personnel.

[0055] Compared with existing technologies, the advantages of this invention are as follows: By assigning a unique identifier to each elevator device through the Hyperledger Fabric blockchain network and storing the device's operating data on the blockchain, the immutability, traceability, and unified management of device identity and data are achieved. The use of a random forest model for anomaly identification and risk prediction of elevator operating data enables accurate assessment of elevator operating status under multi-variable conditions, avoiding missed or incorrect judgments caused by traditional simple rule-based methods. The initial risk prediction value is adaptively adjusted based on the elevator's operating frequency within a set time period, making the risk assessment more closely reflect the actual usage status of the equipment. By analyzing keywords in historical maintenance request templates, feature values ​​are extracted and template levels are classified. After obtaining the final risk prediction value, the template with the highest matching degree is automatically selected and the corresponding abnormal data is filled in, achieving automatic generation of maintenance requests, reducing manual intervention costs, and improving response efficiency. Maintenance requests determine the corresponding maintenance personnel based on risk ranges and are reliably distributed based on the blockchain network, ensuring the transparency, timeliness, and non-repudiation of maintenance task distribution. Attached Figure Description

[0056] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0057] Figure 1 This is a functional block diagram of an elevator network device identity and data management platform optimized using Hyperledger Fabric, provided as an embodiment of the present invention. Detailed Implementation

[0058] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the disclosure to those skilled in the art. It should be noted that, unless otherwise specified, embodiments and features in the embodiments of the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0059] In view of the above problems, it is particularly necessary to design a platform that uses Hyperledger Fabric to optimize the identity and data management of elevator network devices.

[0060] In some embodiments of this application, see Figure 1As shown, a platform for optimizing the identity and data management of elevator network devices using Hyperledger Fabric includes:

[0061] The elevator anomaly detection unit, through the Hyperledger Fabric blockchain network, assigns a unique identifier to each elevator device, collects and stores the operating data of each elevator device on the blockchain network, identifies abnormal data in the elevator device's operating data, and obtains an initial risk prediction value for the elevator device based on a random forest model. The initial risk prediction value is compared with a risk prediction threshold. When the initial risk prediction value is greater than or equal to the risk prediction threshold, it becomes the final risk prediction value. When the initial risk prediction value is less than the risk prediction threshold, the unit identifies the elevator operating frequency within a set time period in the elevator device's operating data and adjusts the initial risk prediction value according to the operating frequency to obtain the final risk prediction value.

[0062] The maintenance request generation unit collects several maintenance request templates, extracts keywords from each template, calculates template feature values ​​based on the keywords, and classifies the templates into template levels based on the feature values. It then confirms the risk range based on the final risk prediction value, determines the corresponding template level based on the risk range, obtains the matching degree between abnormal data and each template within the corresponding template level, determines the maintenance request template based on the matching degree, fills the abnormal data into the maintenance request template, and generates a maintenance request.

[0063] The maintenance request sending unit determines the maintenance personnel based on the risk range and sends the maintenance request to the maintenance personnel based on the blockchain network.

[0064] Specifically, the elevator anomaly detection unit assigns a globally unique identity to each elevator device based on the Hyperledger Fabric blockchain network and establishes an on-chain association between device identity and data. The unit continuously collects multi-dimensional parameter information during elevator operation, including but not limited to elevator start / stop counts, door opening / closing times, current, voltage, car vibration data, operating speed, and floor response time. This operational data is encrypted and uploaded to the consortium blockchain network in block form, ensuring data immutability and traceability. The unit uses a constructed random forest model to extract features and classify the on-chain operational data, identifying anomalous data points and generating corresponding initial risk prediction values. A risk prediction threshold (based on historical risk prediction statistics) is set to distinguish between high-risk and low-risk elevator devices. The initial risk prediction value is compared with the preset risk prediction threshold. If the initial risk prediction value is lower than the threshold, it is considered that the prediction result may indicate an underestimation of risk. Directly basing maintenance strategies on this could lead to the neglect of potential hazards and a reduction in elevator safety. Therefore, it is necessary to further analyze elevator operation data within a set time period to reassess the actual operating status of the elevator, ensuring a more comprehensive and accurate risk assessment. This involves revising the initial risk prediction value to obtain a more valuable final risk prediction value. Furthermore, the initial risk value is weighted and adjusted based on the elevator's usage frequency (e.g., start-stop frequency, load change frequency) within a preset time period (e.g., 24 hours, 7 days) to form a final risk prediction value highly consistent with the actual elevator operating status. The maintenance request generation unit first collects various types of maintenance request templates, extracts core keywords from the templates using natural language processing technology, such as "door lock malfunction," "emergency stop failure," and "overspeed," and constructs a template feature space based on the keyword vectors. Cluster analysis is performed on all template features to divide them into several template levels, each corresponding to different severity or maintenance response levels. Based on the risk range into which the final risk prediction value falls, a template set corresponding to the level is selected, and the matching degree between the abnormal data and each template in the template set is calculated. The template with the highest similarity is prioritized for generating the maintenance request. During the generation process, identified anomalies are automatically filled and bound to template fields to form a structured maintenance request document, enabling automatic writing of maintenance content. The maintenance request sending unit automatically filters eligible maintenance personnel (e.g., qualification level) based on risk level and sends the maintenance request to the designated maintenance personnel's blockchain account node. All task distribution records are written to the blockchain ledger in the form of transactions, ensuring the transparency, traceability, and non-repudiation of the task flow process, providing technical support for subsequent operation and maintenance supervision and accountability.

[0065] Understandably, by leveraging the Hyperledger Fabric blockchain network to assign a unique identifier to each elevator, the problems of chaotic equipment identification, duplicate numbering, and difficulty in unified management in traditional elevator management are solved, enabling full lifecycle traceability and cross-system collaborative management of elevator equipment. Elevator operation data is collected in real time and stored on the blockchain, ensuring the security and integrity of data throughout the entire process of collection, transmission, storage, and use, preventing human tampering or data loss, and providing reliable evidence for subsequent risk assessment, accident analysis, and liability attribution. The system automatically matches the most suitable template based on abnormal data and fills in the maintenance request form, replacing manual filling and avoiding issues such as omissions and delayed responses, thus improving the speed and accuracy of the maintenance process. Maintenance personnel are automatically matched according to risk level and tasks are assigned via the blockchain, ensuring data consistency and security during task assignment. The blockchain record provides a full-process audit trail of maintenance task execution, enhancing management transparency and accountability.

[0066] In some embodiments of this application, when the elevator anomaly identification unit obtains the initial risk prediction value of the elevator equipment based on the random forest model, it includes:

[0067] The elevator anomaly identification unit obtains historical operating data from the operating data and divides the historical operating data into training set and test set;

[0068] The optimal parameters for the random forest model are found by combining cross-validation with grid search, and the random forest model is built.

[0069] Fit a random forest model using the training set;

[0070] The test set is fed into the random forest model and the accuracy of the risk prediction values ​​is calculated.

[0071] When the accuracy rate reaches the preset accuracy rate threshold, the initial risk prediction value of the current elevator equipment is obtained based on the abnormal data.

[0072] Specifically, the elevator anomaly detection unit extracts multi-dimensional feature information from the historical operating data of the elevator equipment and divides it into training and test sets. Then, using cross-validation combined with a grid search strategy, it optimizes multiple hyperparameters of the random forest model (e.g., number of trees, maximum depth, minimum number of samples) to determine the optimal parameter combination and build the model. The training set is used to fit and train the random forest model, and the test set is input into the model for prediction, calculating the accuracy of the risk prediction value. Only when the prediction accuracy reaches a preset accuracy threshold (e.g., 90%) is the currently collected anomaly data input into the model to obtain the initial risk prediction value for the current elevator equipment.

[0073] Understandably, employing cross-validation combined with grid search to optimize model parameters avoids overfitting or underfitting, improves the generalization ability of risk prediction, and ensures that the model maintains good accuracy under different elevator equipment and operating conditions. Compared to traditional threshold-based judgment methods, machine learning modeling can output risk probability values ​​based on complex, multi-dimensional operating data, transforming elevator operating risk from "whether it is abnormal" to "the degree of abnormality," providing a more refined basis for maintenance response grading.

[0074] In some embodiments of this application, when the elevator anomaly identification unit identifies the elevator operating frequency within a set time period in the elevator equipment operating data and adjusts the initial risk prediction value according to the operating frequency to obtain the final risk prediction value, the process includes:

[0075] The elevator anomaly detection unit compares the elevator operating frequency with the operating frequency threshold within a set time period. The operating frequency threshold includes a first operating frequency threshold and a second operating frequency threshold, and the first operating frequency threshold is less than the second operating frequency threshold.

[0076] When the operating frequency is less than the first operating frequency threshold, the elevator anomaly identification unit determines that the initial risk prediction value will not be adjusted and will be used as the final risk prediction value.

[0077] When the operating frequency is greater than or equal to the first operating frequency threshold and less than the second operating frequency threshold, the elevator anomaly identification unit determines to adjust the initial risk prediction value using the first adjustment coefficient to obtain the adjusted risk prediction value, and uses the adjusted risk prediction value as the final risk prediction value.

[0078] When the operating frequency is greater than or equal to the second operating frequency threshold, the elevator anomaly identification unit determines to adjust the initial risk prediction value using the second adjustment coefficient to obtain the adjusted risk prediction value, and uses the adjusted risk prediction value as the final risk prediction value.

[0079] Specifically, the elevator anomaly identification unit collects elevator operating frequency data within a set time period and compares this frequency with preset first and second operating frequency thresholds. Operating frequency reflects the intensity and frequency of elevator use. When the operating frequency is below the first threshold, it indicates infrequent elevator use and light equipment load; the elevator's operation has little impact on anomalies, and the initial risk prediction value output by the random forest model is directly used as the final risk prediction value, avoiding over-adjustment leading to misjudgment. When the operating frequency is between the first and second thresholds, it indicates the elevator is under moderate load; a first adjustment coefficient is introduced to fine-tune the initial risk prediction value to reflect the moderate impact of operating frequency on elevator anomalies. When the operating frequency exceeds the second threshold, it indicates frequent elevator use, increasing the potential risk of failure; a second adjustment coefficient is introduced to adjust the initial risk prediction value more significantly, making the final risk prediction value closer to the actual risk level. The operating frequency thresholds and adjustment coefficients are determined through actual statistics by collecting a large amount of elevator operating data under different usage scenarios, calculating the average number of runs per unit time, and combining indicators such as equipment failure rate and maintenance frequency.

[0080] Understandably, by introducing the operating frequency parameter, the impact of elevator usage intensity on equipment health can be dynamically reflected, compensating for errors caused by relying solely on static model predictions and improving the real-world fit of the prediction results. Setting multi-level adjustment strategies based on different operating frequencies allows elevators used frequently to receive more sensitive risk responses.

[0081] In some embodiments of this application, when the maintenance request generation unit extracts keywords from each maintenance request template and calculates template feature values ​​based on the keywords, it includes:

[0082] The maintenance request generation unit calculates the term frequency and inverse document frequency of keywords in the maintenance request template, and obtains the TF-IDF value of each keyword based on the term frequency and inverse document frequency;

[0083] Template feature values ​​are obtained by weighted averaging of the TF-IDF values ​​of the keywords.

[0084] Specifically, the maintenance request generation unit performs word segmentation on each maintenance request template to obtain a keyword set for that template. For each keyword, its term frequency (TF) is calculated, which is the ratio of the number of times the keyword appears in the template to the total number of words in the template. TF reflects the importance of the keyword in the template; the higher the frequency, the stronger the representativeness of the keyword for that template. To evaluate the distinguishing power of a keyword, the number of templates containing the keyword in the entire maintenance request template library is counted, and the inverse document frequency (IDF) is calculated. The IDF value is obtained by dividing the total number of templates by the number of templates containing the keyword and then taking the logarithm, reflecting the rarity of the keyword among all templates. Common words such as "elevator" and "malfunction" have low IDF, indicating lower distinguishing power; while highly specialized words such as "emergency stop failure" and "door lock malfunction" have high IDF, indicating stronger distinguishing power. Multiplying the term frequency (TF) by the inverse document frequency (IDF) yields the TF-IDF value for each keyword. TF-IDF combines the importance of the keyword in the current template and its distinguishing power in the entire template library, making it an effective indicator of keyword representativeness. To comprehensively reflect the overall characteristics of the template, the TF-IDF values ​​of all keywords in the template are weighted and averaged according to preset weights. The weights can be flexibly set based on factors such as the keyword's position in the template, its professionalism, or its historical matching effect, and the feature value of the template is calculated.

[0085] Understandably, by combining term frequency and inverse document frequency, representative and distinctive keywords in each maintenance request template can be effectively highlighted, avoiding interference from common words and achieving accurate feature characterization of the templates. Numerical template feature values ​​provide a scientific basis for subsequent template grading and anomaly data matching, helping to automatically identify the most suitable maintenance request template and improving the accuracy and response speed of maintenance request generation.

[0086] In some embodiments of this application, when the maintenance request generation unit classifies several maintenance request templates into template levels based on template feature values, it includes:

[0087] S1: Initialize K centroids, and assign each maintenance request template to the nearest centroid to form K clusters;

[0088] S2: Recalculate the centroid of each cluster;

[0089] S3: Repeat S1 and S2 until the centroid no longer changes or the preset number of repetitions is completed;

[0090] S4: Use the centroid of each cluster as the eigenvalue of the cluster.

[0091] In some embodiments of this application, when the maintenance request generation unit classifies several maintenance request templates into template levels based on template feature values, it further includes:

[0092] The maintenance request generation unit compares the feature value of the cluster with the first preset feature value and the second preset feature value respectively. If the first preset feature value is less than the second preset feature value, the unit classifies the template level according to the comparison result.

[0093] When the feature value is less than or equal to the first preset feature value, the maintenance request template in the cluster is classified as the first template level;

[0094] When the feature value is greater than the first preset feature value and less than or equal to the second preset feature value, the maintenance request template in the cluster is classified as the second template level;

[0095] When the feature value is greater than the second preset feature value, the maintenance request template in the cluster will be classified as the third template level;

[0096] The first template level indicates that the risk value in the maintenance request template is lower than the second template level, and the second template level indicates that the risk value in the maintenance request template is lower than the third template level.

[0097] Specifically, the maintenance request generation unit presets the number of clustering clusters K (for example, K = 3, corresponding to low, medium, and high risk levels respectively), and randomly or based on the eigenvalue distribution selects K initial centroids. The centroid represents the clustering center and is the representative point of the template eigenvalue within the cluster. For each maintenance request form template, calculate the distances from the calculated template eigenvalue to the K centroids; assign the template to the cluster where the nearest centroid is located to complete the first cluster division. Perform a weighted average calculation on all the template eigenvalues in each cluster to obtain a new cluster centroid and update the centroid position. Repeat the two steps of template assignment and centroid update until the centroid position no longer changes significantly (i.e., the clustering converges), or reaches the preset maximum number of iterations (such as 100 times) to ensure the stability of clustering. After clustering, use the centroid eigenvalue of each cluster as the overall eigenvalue of the cluster, representing the risk level of the template set. Preset two template level thresholds, namely the first preset eigenvalue (T1) and the second preset eigenvalue (T2), satisfying T1 < T2. The setting of the first preset eigenvalue (T1) and the second preset eigenvalue is combined with the actual risk control requirements of elevator maintenance to ensure that the threshold division can clearly reflect the business meaning of different risk levels, so that low-risk templates correspond to preventive maintenance, medium-risk templates correspond to regular maintenance, and high-risk templates correspond to emergency responses. If the cluster centroid eigenvalue ≤ T1, then all the templates within the cluster are classified into the first template level, indicating a low risk level; if T1 < cluster centroid eigenvalue ≤ T2, then it is classified into the second template level, indicating a medium risk level; if the cluster centroid eigenvalue > T2, then it is classified into the third template level, indicating a high risk level. The first template level corresponds to maintenance request form templates with lower risks and smaller failure probabilities, and is suitable for minor anomalies or preventive maintenance; the second template level represents medium risks and is suitable for abnormal situations that need to be processed in a timely manner but the risks are still controllable; the third template level represents high-risk templates, indicating that there are serious potential fault hazards and requires priority response and emergency maintenance.

[0098] It can be understood that the automatic classification of templates through the clustering algorithm avoids the subjectivity and low efficiency of traditional manual division, and improves the scientificity and consistency of template management. The grading method based on template eigenvalues can accurately reflect the risk levels corresponding to different templates, support the graded response and priority processing of maintenance requests. The clear template level division provides a basis for matching the most suitable template for abnormal data, reduces the matching error, and speeds up the generation speed of maintenance request forms.

[0099] In some embodiments of the present application, when the maintenance request generation unit confirms the risk interval according to the final risk prediction value, it includes:

[0100] The risk interval includes the first risk interval, the second risk interval, and the third risk interval;

[0101] When the final risk prediction value is less than or equal to the first risk prediction threshold, the maintenance request generation unit determines that the final risk prediction value is within the first risk range.

[0102] When the final risk prediction value is greater than the first risk prediction threshold and less than or equal to the second risk prediction threshold, the maintenance request generation unit determines that the final risk prediction value is within the second risk range.

[0103] When the final risk prediction value is greater than the second risk prediction threshold, the maintenance request generation unit determines that the final risk prediction value is within the third risk range.

[0104] The risk value in the first risk range is lower than that in the second risk range, and the risk value in the second risk range is lower than that in the risk range.

[0105] The first risk prediction threshold is less than the second risk prediction threshold.

[0106] Specifically, the maintenance request generation unit sets two risk prediction thresholds. These thresholds are determined by statistically analyzing historical operating data of a large amount of elevator equipment and corresponding risk events to identify the distribution characteristics of the predicted risk values. Data mining techniques are used to analyze the predicted value ranges corresponding to different risk levels, thereby defining reasonable threshold ranges. These include a first risk prediction threshold and a second risk prediction threshold, with the first threshold being less than the second. This setting is used to delineate risk level boundaries. When the final risk prediction value is less than or equal to the first risk prediction threshold, the elevator equipment is determined to be in the first risk range, indicating low equipment risk, and usually no emergency maintenance measures are required. When the final risk prediction value is greater than the first risk prediction threshold but less than or equal to the second risk prediction threshold, the equipment is determined to be in the second risk range, indicating moderate risk, and timely maintenance is recommended. When the final risk prediction value is greater than the second risk threshold, the equipment is determined to be in the third risk range, with the highest risk, requiring priority troubleshooting and maintenance intervention. The risk value of the first risk range is lower than that of the second risk range, and the risk value of the second risk range is lower than that of the third risk range, ensuring the hierarchical and progressive nature of risk classification.

[0107] Understandably, by setting a first and a second risk prediction threshold, elevator equipment risks are divided into three distinct ranges: low, medium, and high. This helps to quickly identify the equipment's condition, facilitating subsequent management and decision-making. Different risk ranges correspond to different maintenance strategies and priorities, enabling the rational allocation of maintenance resources. Low-risk equipment undergoes routine inspections, while medium- and high-risk equipment is given priority for intensive maintenance, thus improving maintenance efficiency.

[0108] In some embodiments of this application, the maintenance request generation unit determines the template level corresponding to the maintenance request template based on the risk range, including:

[0109] When the final risk prediction value is within the first risk range, the maintenance request generation unit determines the template level corresponding to the maintenance request template as the first template level.

[0110] When the final risk prediction value is within the second risk range, the maintenance request generation unit determines the template level corresponding to the maintenance request template as the second template level.

[0111] When the final risk prediction value is within the third risk range, the maintenance request generation unit determines the template level corresponding to the maintenance request template as the third template level.

[0112] Specifically, when the final risk prediction value of the elevator equipment falls within the first risk range, indicating a low risk, the maintenance request generation unit will select a first-level maintenance request template, suitable for routine inspections or preventative maintenance under low-risk conditions, avoiding resource waste. When the final risk prediction value falls within the second risk range, indicating a medium risk, the unit will automatically match a second-level maintenance request template for more targeted maintenance requests, ensuring timely handling of potential problems. When the final risk prediction value exceeds the second risk prediction threshold, entering the third risk range, indicating a high risk or significant potential for malfunction, the unit will select a third-level maintenance request template, triggering an emergency maintenance process to ensure safe equipment operation.

[0113] In some embodiments of this application, when obtaining the matching degree between abnormal data and each template within the corresponding template level, the maintenance request generation unit determines the maintenance request template based on the matching degree, including:

[0114] The maintenance request generation unit fills the abnormal data into each template within the corresponding template level one by one, and determines the matching degree by the proportion of abnormal data filled into each template.

[0115] Sort each template within the corresponding template level by matching degree from high to low, and select the maintenance request template with the highest matching degree.

[0116] Specifically, the maintenance request generation unit fills in the corresponding fields or placeholders in each maintenance request template within the corresponding template level with the collected abnormal data, one by one. The filling process ensures that the abnormal information is completely mapped to the template, reflecting the specific details of the problem. For each template, the ratio of successfully filled fields to the total number of fields required by the template is calculated, i.e., the filling ratio. This ratio serves as a matching indicator between the abnormal data and the template, reflecting the template's suitability for the abnormal data. The matching values ​​of all templates within the corresponding template level are sorted from highest to lowest to facilitate the selection of the most suitable template. The maintenance request template with the highest matching ratio is selected as the final template, ensuring that the abnormal data is covered and accurately expressed to the greatest extent possible, improving the relevance and effectiveness of the maintenance request.

[0117] In some embodiments of this application, when the maintenance request sending unit determines maintenance personnel based on a risk range, it includes:

[0118] Maintenance personnel include junior maintenance personnel, intermediate maintenance personnel, and senior maintenance personnel;

[0119] When the final risk prediction value is in the first risk range, the maintenance request sending unit determines the maintenance personnel as junior maintenance personnel.

[0120] When the final risk prediction value is in the second risk range, the maintenance request sending unit determines that the maintenance personnel are intermediate maintenance personnel.

[0121] When the final risk prediction value is in the third risk range, the maintenance request sending unit determines that the maintenance personnel are senior maintenance personnel.

[0122] Understandably, matching maintenance personnel with different qualifications and experience based on risk levels optimizes the allocation of human resources, avoids safety hazards caused by junior personnel handling high-risk issues, and reduces resource waste by senior personnel in low-risk scenarios. Assigning maintenance personnel with corresponding professional capabilities to different risk levels improves the professionalism and accuracy of problem handling, shortens fault response time, and enhances overall maintenance quality. High-risk elevators are handled by senior maintenance personnel, ensuring that complex or emergency faults are handled promptly and effectively, reducing the risk of safety accidents and protecting passenger safety.

[0123] In summary, by assigning a unique identifier to each elevator device through the Hyperledger Fabric blockchain network and storing the device's operational data on the blockchain, the immutability, traceability, and unified management of device identity and data are achieved. The use of a random forest model for anomaly identification and risk prediction of elevator operational data enables accurate assessment of elevator operating status under multivariate conditions, avoiding missed or false judgments caused by traditional simple rule-based methods. The initial risk prediction value is adaptively adjusted based on the elevator's operating frequency within a set time period, making the risk assessment more closely reflect the actual usage status of the equipment. By analyzing keywords in historical maintenance request templates, feature values ​​are extracted and template levels are categorized. After obtaining the final risk prediction value, the template with the highest matching degree is automatically selected and the corresponding anomaly data is filled in, achieving automatic generation of maintenance requests, reducing manual intervention costs, and improving response efficiency. Maintenance requests determine the corresponding maintenance personnel based on risk ranges and are reliably distributed based on the blockchain network, ensuring the transparency, timeliness, and non-repudiation of maintenance task distribution.

[0124] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program goods. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program goods embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0125] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program goods according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0126] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0127] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0128] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A platform for optimizing the identity and data management of elevator network devices using Hyperledger Fabric, characterized in that, include: The elevator anomaly identification unit assigns a unique identifier to each elevator device through the Hyperledger Fabric blockchain network, collects the operating data of each elevator device and stores it in the blockchain network, identifies abnormal data in the elevator device operating data, and obtains the initial risk prediction value of the elevator device based on the random forest model. The initial risk prediction value is compared with the risk prediction threshold. When the initial risk prediction value is greater than or equal to the risk prediction threshold, the initial risk prediction value is the final risk prediction value. When the initial risk prediction value is less than the risk prediction threshold, the elevator operating frequency within a set time period in the elevator equipment operation data is identified, and the initial risk prediction value is adjusted according to the operating frequency to obtain the final risk prediction value. The maintenance request generation unit collects several maintenance request templates, extracts keywords from each maintenance request template, calculates template feature values ​​based on the keywords, and classifies the several maintenance request templates into template levels based on the template feature values. The risk range is determined based on the final risk forecast value; Based on the risk range, determine the template level corresponding to the maintenance request form template. After determining the corresponding template level, obtain the matching degree between the abnormal data and each template within the corresponding template level. Based on the matching degree, determine the maintenance request form template, fill the abnormal data into the maintenance request form template, and generate the maintenance request form. The maintenance request sending unit determines the maintenance personnel based on the risk range and sends the maintenance request to the maintenance personnel based on the blockchain network.

2. The Hyperledger Fabric-optimized elevator network device identity and data management platform according to claim 1, characterized in that, When the elevator anomaly identification unit obtains the initial risk prediction value of the elevator equipment based on the random forest model, it includes: The elevator anomaly identification unit obtains historical operating data from the operating data and divides the historical operating data into a training set and a test set. The optimal parameters for the random forest model are found by combining cross-validation with grid search, and the random forest model is built. The training set is used to fit the random forest model; The test set is fed into the random forest model and the accuracy of the risk prediction values ​​is calculated. When the accuracy rate reaches a preset accuracy rate threshold, the initial risk prediction value of the elevator equipment is obtained based on the abnormal data.

3. The Hyperledger Fabric-optimized elevator network device identity and data management platform according to claim 2, characterized in that, The elevator anomaly identification unit identifies the elevator operating frequency within a set time period in the elevator equipment operation data, and adjusts the initial risk prediction value according to the operating frequency to obtain the final risk prediction value, including: The elevator anomaly identification unit compares the elevator operating frequency within the set time period with the operating frequency threshold. The operating frequency threshold includes a first operating frequency threshold and a second operating frequency threshold, wherein the first operating frequency threshold is less than the second operating frequency threshold. When the operating frequency is less than the first operating frequency threshold, the elevator anomaly identification unit determines that the initial risk prediction value will not be adjusted and uses the initial risk prediction value as the final risk prediction value. When the operating frequency is greater than or equal to the first operating frequency threshold and less than the second operating frequency threshold, the elevator anomaly identification unit determines to adjust the initial risk prediction value using the first adjustment coefficient to obtain the adjusted risk prediction value, and uses the adjusted risk prediction value as the final risk prediction value. When the operating frequency is greater than or equal to the second operating frequency threshold, the elevator anomaly identification unit determines to adjust the initial risk prediction value using the second adjustment coefficient to obtain the adjusted risk prediction value, and uses the adjusted risk prediction value as the final risk prediction value.

4. The Hyperledger Fabric-optimized elevator network device identity and data management platform according to claim 3, characterized in that, When the maintenance request generation unit extracts keywords from each maintenance request template and calculates template feature values ​​based on the keywords, it includes: The maintenance request generation unit calculates the term frequency and inverse document frequency of the keywords in the maintenance request template, and obtains the TF-IDF value of each keyword based on the term frequency and inverse document frequency. The template feature value is obtained by weighted averaging of the TF-IDF values ​​of the keywords.

5. The Hyperledger Fabric-optimized elevator network device identity and data management platform according to claim 4, characterized in that, When the maintenance request generation unit classifies several maintenance request templates into template levels based on the template feature values, it includes: S1: Initialize K centroids, and assign each maintenance request template to the nearest centroid to form K clusters; S2: Recalculate the centroid of each cluster; S3: Repeat S1 and S2 until the centroid no longer changes or the preset number of repetitions is completed; S4: Use the centroid of each cluster as the eigenvalue of the cluster.

6. The Hyperledger Fabric-optimized elevator network device identity and data management platform according to claim 5, characterized in that, When the maintenance request generation unit classifies several maintenance request templates into template levels based on the template feature values, it also includes: The maintenance request generation unit compares the feature value of the cluster with a first preset feature value and a second preset feature value respectively. The first preset feature value is less than the second preset feature value. Based on the comparison result, several maintenance request templates are divided into template levels. When the feature value is less than or equal to the first preset feature value, the maintenance request template in the cluster is classified into the first template level; When the feature value is greater than the first preset feature value and less than or equal to the second preset feature value, the maintenance request template in the cluster is classified into the second template level; When the feature value is greater than the second preset feature value, the maintenance request template in the cluster is classified into the third template level; The first template level indicates that the risk value in the maintenance request template is lower than the second template level, and the second template level indicates that the risk value in the maintenance request template is lower than the third template level.

7. The Hyperledger Fabric-optimized elevator network device identity and data management platform according to claim 6, characterized in that, When the maintenance request generation unit confirms the risk range based on the final risk prediction value, it includes: The risk range includes a first risk range, a second risk range, and a third risk range; When the final risk prediction value is less than or equal to the first risk prediction threshold, the maintenance request generation unit determines that the final risk prediction value is within the first risk range. When the final risk prediction value is greater than the first risk prediction threshold and less than or equal to the second risk prediction threshold, the maintenance request generation unit determines that the final risk prediction value is within the second risk range. When the final risk prediction value is greater than the second risk prediction threshold, the maintenance request generation unit determines that the final risk prediction value is within the third risk range; The risk value in the first risk interval is lower than that in the second risk interval, and the risk value in the second risk interval is lower than that in the risk interval. The first risk prediction threshold is less than the second risk prediction threshold.

8. The Hyperledger Fabric-optimized elevator network device identity and data management platform according to claim 7, characterized in that, The maintenance request generation unit determines the template level corresponding to the maintenance request template based on the risk range, including: When the final risk prediction value is within the first risk range, the maintenance request generation unit determines the template level corresponding to the maintenance request template as the first template level. When the final risk prediction value is within the second risk range, the maintenance request generation unit determines the template level corresponding to the maintenance request template as the second template level; When the final risk prediction value is within the third risk range, the maintenance request generation unit determines the template level corresponding to the maintenance request template as the third template level.

9. The Hyperledger Fabric-optimized elevator network device identity and data management platform according to claim 8, characterized in that, Obtaining the matching degree between the abnormal data and each template within the corresponding template level, the maintenance request generation unit determines the maintenance request template based on the matching degree, including: The maintenance request generation unit fills the abnormal data into each template within the corresponding template level one by one, and determines the matching degree by the proportion of abnormal data filled into each template. Arrange the matching degree of each template within the corresponding template level from high to low, and select the maintenance request template with the highest matching degree.

10. The Hyperledger Fabric-optimized elevator network device identity and data management platform according to claim 9, characterized in that, When the maintenance request sending unit determines maintenance personnel based on the risk range, it includes: The maintenance personnel include junior maintenance personnel, intermediate maintenance personnel, and senior maintenance personnel; When the final risk prediction value is within the first risk range, the maintenance request sending unit determines that the identified maintenance personnel is the primary maintenance personnel; When the final risk prediction value is within the second risk range, the maintenance request sending unit determines that the maintenance personnel are the intermediate maintenance personnel; When the final risk prediction value is within the third risk range, the maintenance request sending unit determines that the maintenance personnel are the senior maintenance personnel.

Images