Auxiliary decision-making method, system and product for fault processing of platform screen door

Abstract

The invention discloses a platform shielding door fault processing auxiliary decision-making method, system and product, and relates to the technical field of urban rail transit platform shielding doors, and the method comprises the steps: obtaining shielding door fault information which is collected and transmitted to a station subsystem through a mobile terminal subsystem, the shielding door fault information comprises the fault type, the fault location, the fault time and the fault phenomenon of the shielding door; and fault processing measures corresponding to each role are obtained based on the shielding door fault information and a shielding door fault processing knowledge graph, the shielding door fault processing knowledge graph is constructed according to the obtained shielding door fault processing related data, and the roles comprise station attendants, station comprehensive controllers and driving dispatchers. According to the method, information circulation among the driving dispatcher, the station comprehensive controller and the station operator is electronized, and fault processing measures which should be taken by different roles are automatically derived through the knowledge graph, so that errors caused by manual judgment are reduced, and the efficiency and accuracy of fault processing are improved.

Description

Technical Field

[0001] The present application relates to the technical field of platform screen doors for urban rail transit, and in particular to a method, system and product for auxiliary decision-making for fault handling of platform screen doors. Background Art

[0002] Urban rail transit plays a vital role in the transportation network of modern cities. It not only significantly alleviates the pressure of urban traffic, but also provides citizens with convenient and efficient travel options. As a key component of this system, the urban rail transit platform screen door builds a safety barrier between the urban rail transit platform and the train, effectively preventing passengers from accidentally falling onto the track. At the same time, it also creates a more comfortable waiting environment for the urban rail transit platform, reducing the impact of noise and airflow generated by the train running on passengers.

[0003] The platform shield door system integrates multiple professional knowledge such as machinery, communication and electromechanical equipment, and has many interfaces and complex structures with multiple systems such as signal systems, low-voltage power distribution systems and integrated monitoring systems. Due to the closed space and large flow of people in urban rail transit stations, there are many uncontrollable factors, which makes the platform shield door system face many potential failure risks. Once the platform shield door fails, it will not only affect the normal operation of urban rail transit at the faulty station and adjacent lines, but may also threaten the safety of passengers.

[0004] At present, in the process of handling platform screen door failures, train dispatchers and station controllers, station controllers and station attendants rely on dispatch telephones and walkie-talkies to obtain information through verbal communication; at the same time, the handling of platform screen door system failures and the adjustment of urban rail transit operations after the handling are mostly dependent on the experience and judgment of train dispatchers; this method has obvious limitations in terms of efficiency and accuracy. Summary of the invention

[0005] The purpose of this application is to provide a platform screen door fault handling auxiliary decision-making method, system and product, which can improve the accuracy and efficiency of platform screen door fault handling.

[0006] To achieve the above objectives, this application provides the following solutions:

[0007] In a first aspect, the present application provides an auxiliary decision-making method for handling platform screen door faults, comprising:

[0008] Acquire the platform screen door fault information sent by the station subsystem; the platform screen door fault information is input by the station attendant into the mobile terminal subsystem for collection and sent to the station subsystem, and the platform screen door fault information includes the platform screen door fault type, platform screen door fault location, platform screen door fault time and platform screen door fault phenomenon;

[0009] Based on the platform screen door fault information and the platform screen door fault processing knowledge graph, the fault processing measures corresponding to each role are obtained respectively; the platform screen door fault processing knowledge graph is constructed according to the obtained platform screen door fault processing related data, and the roles include the station attendant, the station integrated controller and the train dispatcher;

[0010] The fault handling measures corresponding to the station attendant are sent to the mobile terminal subsystem, the fault handling measures corresponding to the station integrated controller are sent to the station subsystem, and the fault handling measures corresponding to the train dispatcher are displayed.

[0011] In a second aspect, the present application provides an auxiliary decision-making system for urban rail transit platform screen door fault handling, comprising:

[0012] The mobile terminal subsystem is used to collect the platform screen door fault information input by the station attendant and send it to the station subsystem; the platform screen door fault information includes the platform screen door fault type, platform screen door fault location, platform screen door fault time and platform screen door fault phenomenon;

[0013] The station subsystem is used to send the platform screen door fault information to the dispatching center subsystem;

[0014] The dispatch center subsystem is used to:

[0015] Based on the platform screen door fault information and the platform screen door fault processing knowledge graph, the fault processing measures corresponding to each role are obtained respectively; the platform screen door fault processing knowledge graph is constructed according to the obtained platform screen door fault processing related data, and the roles include the station attendant, the station integrated controller and the train dispatcher;

[0016] The fault handling measures corresponding to the station attendant are sent to the mobile terminal subsystem, the fault handling measures corresponding to the station integrated controller are sent to the station subsystem, and the fault handling measures corresponding to the train dispatcher are displayed.

[0017] In a third aspect, the present application provides a computer program product, including a computer program, which, when executed by a processor, implements the above-mentioned auxiliary decision-making method for platform screen door fault processing.

[0018] According to the specific embodiments provided in this application, this application discloses the following technical effects:

[0019] The present application provides a method, system and product for assisting decision-making in handling platform screen door faults, which obtains screen door fault information sent by the station subsystem, which is input by the station attendant into the mobile terminal subsystem and sent to the station subsystem, and includes the fault type, fault location, fault time and fault phenomenon of the screen door; based on the fault information and the knowledge graph of screen door fault handling, the fault handling measures corresponding to each role are obtained respectively, wherein the knowledge graph of screen door fault handling is constructed based on the acquired data related to screen door fault handling, and the roles include station attendant, station integrated controller and train dispatcher; the fault handling measures corresponding to the station attendant are sent to the mobile terminal subsystem, the fault handling measures corresponding to the station integrated controller are sent to the station subsystem, and the fault handling measures corresponding to the train dispatcher are displayed. The present application electronicizes the information flow between the train dispatcher, the station integrated controller and the station attendant through information interaction between the mobile terminal subsystem, the station subsystem and the dispatching center subsystem, thereby improving the response speed and efficiency of fault handling. Moreover, a knowledge graph is constructed based on the relevant data of platform shield door fault handling, and targeted fault handling measures are provided for different roles (such as station attendants, station integrated controllers, and train dispatchers) through the knowledge graph. Therefore, the most suitable handling measures can be obtained through systematic knowledge deduction, thereby improving the accuracy and efficiency of fault diagnosis and handling. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0021] Figure 1 This is an application environment diagram of an auxiliary decision-making method for platform screen door fault processing in one embodiment of the present application;

[0022] Figure 2 A flowchart of an auxiliary decision-making method for handling platform screen door failures provided in one embodiment of the present application;

[0023] Figure 3 A schematic diagram of the process of constructing a knowledge graph for handling faults of a platform shielding door provided in an embodiment of the present application;

[0024] Figure 4 A schematic diagram of the knowledge graph structure for handling faults of a platform shielding door provided in one embodiment of the present application;

[0025] Figure 5 A schematic diagram of functional modules of an auxiliary decision-making system for platform screen door fault processing provided by an embodiment of the present application; Figure 6 A schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0026] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0027] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0028] The auxiliary decision-making method for platform screen door fault processing provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be set up separately, integrated on the server 104, or placed on the cloud or other servers. The terminal 102 can send the screen door fault information to the server 104. After the server 104 receives the screen door fault information, the server 104 obtains the fault handling measures corresponding to each role based on the screen door fault information and the screen door fault processing knowledge map; the fault handling measures corresponding to the station attendant are sent to the mobile terminal subsystem, the fault handling measures corresponding to the station integrated controller are sent to the station subsystem, and the fault handling measures corresponding to the train dispatcher are carried out. In addition, in some embodiments, the auxiliary decision-making method for platform screen door fault processing can also be implemented by the server 104 or the terminal 102 alone, such as the terminal 102 can directly process the screen door fault information and the screen door fault processing knowledge map, or the server 104 can obtain the screen door fault information and the screen door fault processing knowledge map from the data storage system and process them.

[0029] The terminal 102 is a dispatch center subsystem. The server 104 can be implemented by an independent server or a server cluster composed of multiple servers, or a cloud server.

[0030] In an exemplary embodiment, Figure 2 As shown, a method for assisting decision-making in handling platform screen door faults is provided. The method is executed by a computer device, and can be executed by a computer device such as a terminal or a server alone, or can be executed by a terminal and a server together. In the embodiment of the present application, the method is applied to Figure 1 The server 104 in the example is used for explanation, and the steps include the following steps 201 to 205.

[0031] in:

[0032] Step 201, obtain the PSD fault information sent by the station subsystem; the PSD fault information is input by the station attendant into the mobile terminal subsystem and sent to the station subsystem, and the PSD fault information includes the PSD fault type, PSD fault location, PSD fault time and PSD fault phenomenon. By adopting a distributed system architecture, the information flow between the train dispatcher, the station integrated controller and the station attendant is electronic, specifically including the following contents.

[0033] When the train stops and needs to depart, the platform door system transmits the platform door information to the Automatic Train Supervision (ATS), and at the same time, the platform door system transmits the platform door status information to the Computers Interlocking (CI), which then forwards it to the ATS. The ATS transmits the two types of information to the auxiliary decision system, which receives the fault information, generates a fault warning pop-up window, and pushes it to the dispatch center subsystem, station subsystem, and mobile terminal subsystem. At the same time, the auxiliary decision system automatically calls the monitoring screen closest to the platform door and displays it on the auxiliary information interface of the dispatch center subsystem.

[0034] After receiving the fault warning information, the station attendant goes to the site to confirm the fault situation and handle the fault, and feedbacks the on-site fault information and handling situation to the station integrated controller through the mobile terminal subsystem. The station integrated controller feedbacks the on-site fault information and the station attendant's on-site handling progress to the dispatching center subsystem through the station subsystem.

[0035] Step 202, based on the platform shielding door fault information and the platform shielding door fault handling knowledge graph, obtain multiple fault handling plans corresponding to each role respectively; the platform shielding door fault handling knowledge graph is constructed according to the acquired platform shielding door fault handling related data, and the roles include station attendants, station integrated controllers and train dispatchers; specifically includes the following contents.

[0036] Step 221, based on the acquired data related to the shielding door fault processing, a shielding door fault processing knowledge graph is constructed. The specific process is as follows: Figure 3 shown.

[0037] The construction of knowledge graph mainly includes four parts: data collection, knowledge extraction, knowledge fusion and knowledge storage. Data collection is the basis of knowledge graph construction. The original data directly affects the quality of subsequent steps. Knowledge extraction depends on the collected data, and its output becomes the input of the knowledge fusion stage. Knowledge fusion solves the problems of duplication and conflict of content in multiple data formats and increases the reliability of knowledge graph. Knowledge storage saves the current state of knowledge so that the knowledge can be quickly accessed when applied, and supports the continuous updating and development of knowledge graph.

[0038] At the data layer, the data related to the processing of shielding door faults is collected through web crawlers and manual screening methods, and the data related to the processing of shielding door faults is classified according to the data format to obtain structured data, semi-structured data and unstructured data, and then the data is cleaned to delete useless and repeated content. The specific content of the data related to the processing of shielding door faults is shown in Table 1:

[0039] Table 1 Description of relevant data for shield door fault processing

[0040]

[0041]

[0042] Among them, by adding the authority and responsibility regulations of different work roles in the knowledge graph, as shown in Table 2, corresponding information and support are provided to each role during the fault handling process. For example, station attendants need operation guides for preliminary fault troubleshooting and handling, station integrated controllers need information on fault classification and preliminary analysis, and train dispatchers need information on train operation adjustments. In this way, it is ensured that each role can obtain the key information required to perform its authority and responsibilities in a timely manner, thereby improving the pertinence and practicality of auxiliary information.

[0043] Table 2 Description of the authority and responsibilities of job roles

[0044]

[0045] In the knowledge extraction stage, for structured data, D2R technology is used to perform knowledge extraction operations on structured data to obtain the first data result, that is, converting the content of the relational database into RDF format data of the semantic web standard, so as to promote data interoperability and reusability; for semi-structured data, including XML and JSON files, the wrapper method is used to perform knowledge extraction operations on semi-structured data to obtain the second data result. The wrapper method can effectively parse these data and extract valuable information from them; unstructured data includes unorganized text and audio, and it is challenging to mine useful knowledge from them. Therefore, natural processing technology is used to automatically perform knowledge extraction operations on unstructured data to obtain the third data result, that is, identifying and extracting key entities and relationship information from the original content. Among them, the first data result, the second data result and the third data result are all RDF format data.

[0046] Further, in this embodiment, the knowledge extraction of unstructured data includes two parts: entity recognition and relationship extraction. Among them, in the entity recognition part, a joint model Bi-LSTM-CRF-DL is designed as an entity recognition model to identify entities and add role labels; the entity recognition model includes a bidirectional long-term and short-term memory network (Bilateral long-term and short-term memory network, Bi-LSTM), a conditional random field (Condition random field, CRF) and a fully connected layer (Dense Layer, DL) connected in sequence. By adding a fully connected layer after the CRF layer, the Bi-LSTM-CRF model with fusion word embedding is used to extract entities from the text, and then the fully connected layer is used to realize multi-label classification, add role labels to each entity, clarify the role of the entity in fault handling, and realize entity recognition and role labeling in the same framework. For example, the local control panel is an entity, which is operated by the station attendant, so its role is marked as a station attendant. The entity categories and definitions in this embodiment are shown in Table 3. The specific process includes the following contents.

[0047] Table 3 Entity categories and definitions

[0048]

[0049] The Skip-gram model in Word2Vec is used to vectorize words in unstructured data to obtain word vectors for entities, word vectors for role labels, and word vectors for authority and responsibility. The word vectors for entities, word vectors for role labels, and word vectors for authority and responsibility are fused using the weighted average method to obtain a unified first vector. The weights of different vectors are dynamically adjusted according to the context and semantic relevance of the entity to improve the semantic representation of the entity in the knowledge graph.

[0050] In the Skip-gram model, given a central word and its context words, the model's goal is to maximize the conditional probability of the context words. This conditional probability can be calculated using the softmax function:

[0051]

[0052] in, The central word w O The word vector of is the context word w i The word vector of is the context word w i word vector in softmax; W is the size of the vocabulary; v′ w are the word vectors of all words in the vocabulary; T represents the transpose operation.

[0053] The model is trained by maximizing the following objective function:

[0054]

[0055] Where T is the total number of training samples; k is the size of the context window; The central word w O The context word at position i+j.

[0056] After obtaining the word vectors of entities, role labels, and authority responsibilities, we use the weighted average method to merge these vectors into a unified first vector v f :

[0057]

[0058] Among them, v e 、w e are the word vectors and weights of the entities; v r 、w r are the word vectors and weights of role labels; v d 、w d They are the word vectors of authority and responsibility respectively.

[0059] In order to ensure that the length of the fused vector is moderate, it is normalized to ensure that all vectors have the same scale, which helps to improve the stability of the subsequent entity recognition model. Finally, the normalized second vector is input into the joint model.

[0060] In the joint model, Bi-LSTM provides context-dependent representation of word vectors, CRF processes the dependencies of word sequences to accurately identify entity boundaries and types, and the fully connected layer receives the output of Bi-LSTM-CRF and predicts the role of each entity.

[0061] Bi-LSTM is composed of a forward LSTM, a backward LSTM and a linear change layer. The forward LSTM is responsible for the operation of the input information in the forward direction, and the backward LSTM operates the input information in the reverse direction to generate context-related hidden states. The LSTM model consists of multiple consecutive memory cells. The memory cell at position t is sensitive to the input sequence X = (X1, X2, ···, X n ) t The text vector x t After processing, the hidden layer output h is obtained t , the specific calculation process is as follows:

[0062]

[0063] Among them, t 、f t 、i t are the output gate, forget gate, and input gate at time t; tanh and σ are activation functions; W f , W i ,Wo,W C are the weight parameter matrices of the forget gate, output gate, input gate and memory cell respectively; b f 、b i 、b o 、b C are the bias vectors of the forget gate, output gate, input gate and memory cell respectively; h t , C t is the output and state of the memory cell, and ⊙ represents the XOR operation.

[0064] The Bi-LSTM module first calculates the forward layer in sequence The backward layer calculates in reverse order Then the outputs in the two directions are concatenated to obtain the hidden layer output h at position t t for: The function of the BI-LSTM module is to obtain the sentence label probability matrix P by concatenating the outputs of the forward LSTM layer and the reverse LSTM layer and passing them through a linear transformation layer, which serves as the input of the CRF module.

[0065] The CRF module is spliced ​​after the network layer as the inference layer, and processes the output of Bi-LSTM to obtain the optimal sequence that maximizes the objective function, so that the annotation result of the entire sentence is the most reasonable. Its objective function not only considers the state feature function of the input, but also includes the label transfer feature function to optimize the final annotation result. Let y = [y1, y2, ..., y n-1 ,y n ] is the input label matrix to be predicted, then the CRF score of the sequence is:

[0066]

[0067] Among them, P is the score matrix output by the network layer, with a size of n×m, where m is the number of different tags. According to the above entity category definition, m is 8; The yth word corresponding to the tth word in the sentence t The scores of the labels, A is the matrix of conversion scores, is the transition feature matrix used to store the probability of transition between all labels.

[0068] The fully connected layer maps the entity sequence in the output of the CRF layer to the role label space, and converts the linearly transformed value into the probability of each role through the Sigmoid activation function.

[0069] Fully connected layer linear change formula:

[0070] z=W fc ·h t '+b fc ;

[0071] Among them, W fc is a weight matrix of size (m, c), where c is the number of role categories, i.e. 3; h t 'Entity label output by the CRF layer; b fc is a bias vector of length c.

[0072] When training the model, the weight matrix is ​​randomly initialized using a uniform distribution, and the bias vector is initialized to a zero vector. In each training batch, the loss between the predicted output and the true label is calculated using the binary cross entropy loss function, and the gradient is calculated using the back propagation algorithm. Then, the weight matrix and bias vector are updated based on these gradients.

[0073] When training the model, the weight matrix is ​​randomly initialized using a uniform distribution, and the bias vector is initialized to a zero vector. In each training batch, the loss between the predicted output and the true label is calculated using the binary cross entropy loss function, and the gradient is calculated using the back propagation algorithm. Then, the weight matrix and bias vector are updated based on these gradients.

[0074] In addition, in the relationship extraction part, the relationship between entities is extracted by using Text Convolutional Neural Networks (TextCNN) as a relationship extraction model, and the authority and responsibility relationship of a specific role is added. By using the second vector as input in the TextCNN model, the semantic information of word embedding is used to improve the accuracy of relationship extraction, and the authority and responsibility relationship is added between the entities extracted in the relationship extraction part, such as "responsible", "authorized to operate", etc., to reflect their interaction in fault handling. The key relationship categories and definitions in this embodiment are shown in Table 4.

[0075] Table 4 Key relationship categories and definitions

[0076] Relationship Category Relationship Definition Positional Relationship Positional relationship between fault phenomenon and fault location Select Relationship The selection relationship between fault type and treatment measures Character Trait Relationship The correspondence between personnel and authority and responsibilities Affiliation The relationship between fault phenomena and fault impacts

[0077] Finally, the BERT model is used to perform knowledge fusion operations on the first data results, the second data results, and the third data results, and the descriptive information about the same entity or concept in data from multiple different sources is integrated to obtain the knowledge graph of shielding door fault processing, such as Figure 4 As shown; at the same time, the above-constructed shielding door fault processing knowledge graph is stored in the form of Neo4j graph database.

[0078] Step 2022: Based on the shielding door fault information, query the shielding door fault handling knowledge graph for information related to the reported fault type, including historical cases and standard operating procedures, and use the rules and reasoning engine in the knowledge graph to automatically retrieve and infer multiple fault handling plans corresponding to different roles.

[0079] Step 203, according to the multiple fault handling plans corresponding to each role, a recommendation strength is assigned to each fault handling plan based on a clustering algorithm, and the fault handling measures corresponding to each role are determined, thereby reducing the decision-making pressure and time consumption of the staff in an emergency. Specifically, the following contents are included.

[0080] Step 2031, obtain historical fault handling data; based on the fault type, fault location, impact scope and urgency, use the K-Means++ algorithm to cluster the historical fault handling data, classify similar historical fault cases in the historical fault handling data into one category, and according to the fault type in each category of historical fault cases, use a simulation method to execute the corresponding fault handling plan, and record the success rate, average resolution time and resource consumption of each fault handling plan as evaluation indicators.

[0081] Step 2032, based on the success rate, average resolution time and resource consumption of each fault handling plan, and through expert scoring, the weight of each evaluation indicator is determined, and then the comprehensive score of each fault handling plan is calculated in combination with the weight, and the comprehensive score of each fault handling plan in the multiple fault handling plans corresponding to each role is ranked to obtain a ranking result.

[0082] Step 2033, according to the ranking results, each fault handling plan is assigned a corresponding level, including "optimal solution", "relatively good solution" and "available solution", and each level is assigned a corresponding recommendation strength, including "strongly recommended", "generally recommended" and "optional recommendation".

[0083] Step 2034, determining the fault handling measures corresponding to each role; the fault handling measures corresponding to each role are determined based on the level, recommendation strength and actual situation of each fault handling plan.

[0084] In order to achieve the evaluation and ranking of the above fault handling plans, the specific calculation process is as follows:

[0085] Construct a decision matrix A, where the rows represent different fault handling plans and the columns represent different evaluation indicators. ij Contains the score of a fault handling plan under a certain evaluation indicator.

[0086] Since the dimensions of each evaluation indicator are different, the maximum and minimum standardization method is used for standardization. The formula is as follows:

[0087]

[0088] Among them, x ij is the original data of row i and column j, r ij is the standardized value, max(x j ) and min(x j ) are the maximum and minimum values ​​of the j-th index respectively.

[0089] Determine the weights of success rate, average solution time and resource consumption through expert scoring j , the sum of the weights is 1. Multiply the standardized values ​​by the corresponding weights to obtain the weighted standardized decision matrix V:

[0090] v ij =w j ·r ij ;

[0091] For each fault handling plan, calculate the sum of its weighted scores under all indicators to obtain the comprehensive score S i :

[0092]

[0093] According to the comprehensive score S i All fault handling plans are ranked, divided into different levels according to the ranking results, and given corresponding recommendation strengths. The staff will select the plan based on the actual situation.

[0094] Step 204, sending the fault handling measures corresponding to the station attendant to the mobile terminal subsystem, sending the fault handling measures corresponding to the station integrated controller to the station subsystem, and displaying the fault handling measures corresponding to the train dispatcher.

[0095] Step 205, according to the operation command input by the train dispatcher, output the dispatch control instruction, introduce the intelligent control method, and automatically link other systems to execute the operation command input by the train dispatcher; the operation command input by the train dispatcher is obtained based on the fault handling measures corresponding to the train dispatcher.

[0096] The system first analyzes the operation command issued by the traffic dispatcher, verifies the legality and integrity of the command, and ensures that it complies with the operation specifications and safety requirements. After that, the system determines which systems need to respond to the control command based on the preset logical relationship and operation dependency.

[0097] The dispatching control instructions are sent to the train automatic monitoring system. The train automatic monitoring system is used to execute the car detention operation, the car detention release operation and the operation diagram adjustment operation according to the dispatching control instructions. The commonly used dispatching control instructions and the timing of issuing the commands are shown in Table 5.

[0098] Table 5 Scheduling control instructions and timing of issuing commands

[0099] Order opportunity Seize the car Confirm the failure of the shield door Unlock the car The shield door fault has been resolved Adjust the operation diagram During troubleshooting, Troubleshooting completed

[0100] After the train dispatcher confirms the execution of the car detention operation, the system will immediately link with the train automatic monitoring system to automatically evaluate whether the train meets the car detention standards and implement the car detention operation in sequence. At the same time, the system cooperates with the public address (PA) and passenger information system (PIS) to automatically release the corresponding broadcast information according to the dispatch control instructions to ensure that passengers are notified in time;

[0101] When the dispatcher issues a command to release the train detention, after the system verifies that the station subsystem or mobile terminal subsystem has reported that the train at the faulty station is ready for departure, the system will automatically link with the ATS and gradually release the detention of the train in the established order to ensure smooth operation recovery;

[0102] After the dispatcher confirms the execution of the operation diagram adjustment plan, the system sends the operation diagram to the ATS, and the ATS system executes the adjusted operation diagram to adapt to the current operation needs;

[0103] In addition, when the station attendant fails to solve the problem through emergency measures, the station subsystem will be informed, and the station subsystem will automatically report the fault to the equipment and facility maintenance system and contact the maintenance personnel to go to the site for repair. In addition, the system records the detailed information of all linkage operations, including operation time, operator, operation results, etc., for post-audit and analysis.

[0104] The present application also provides an application scenario, which applies the above-mentioned auxiliary decision-making method for platform screen door fault handling. Specifically: the auxiliary decision-making method for platform screen door fault handling provided in this embodiment can be applied in the platform screen door fault handling scenario. The platform screen door fault handling scenario includes a fault handling measure generation link and a fault handling measure sending and display link; the fault handling measure generation link is used to obtain the fault handling measures corresponding to each role based on the screen door fault information and the screen door fault handling knowledge graph; the fault handling measure sending and display link is used to send the fault handling measures corresponding to the station attendant to the mobile terminal subsystem, send the fault handling measures corresponding to the station integrated controller to the station subsystem, and display the fault handling measures corresponding to the train dispatcher. The auxiliary decision-making method for platform screen door fault handling provided in this embodiment belongs to the fault handling measure generation link and the fault handling measure sending and display link.

[0105] Based on the same inventive concept, the embodiment of the present application also provides an auxiliary decision system for implementing the auxiliary decision method for handling platform screen door faults involved above. The implementation scheme for solving the problem provided by the auxiliary decision system for handling platform screen door faults is similar to the implementation scheme recorded in the above method, so the specific limitations in the one or more auxiliary decision system embodiments provided below can refer to the limitations on the auxiliary decision method above, and will not be repeated here.

[0106] In an exemplary embodiment, Figure 5 As shown, a platform screen door fault processing auxiliary decision system is provided, comprising:

[0107] The mobile terminal subsystem is used to collect the platform shield door fault information input by the station attendant and send it to the station subsystem; the platform shield door fault information includes the platform shield door fault type, fault location, fault time and fault phenomenon.

[0108] The station subsystem is used to: send platform screen door fault information to the dispatching center subsystem; monitor the platform screen door status of the station where the station controller is located, and collect the station operation log input by the station controller.

[0109] The dispatching center subsystem is used to: construct a knowledge graph of platform screen door fault processing based on the acquired platform screen door fault processing related data, and obtain the fault handling measures corresponding to each role based on the platform screen door fault information and the platform screen door fault processing knowledge graph; the roles include station attendants, station integrated controllers and train dispatchers; send the fault handling measures corresponding to the station attendants to the mobile terminal subsystem, send the fault handling measures corresponding to the station integrated controllers to the station subsystem, and display the fault handling measures corresponding to the train dispatchers; dispatch and monitor the status of the platform screen doors of the entire urban rail transit line.

[0110] In an exemplary embodiment, a computer device is provided. The computer device may be a server or a terminal. The internal structure diagram thereof may be as follows: Figure 6 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store platform screen door fault processing data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, an auxiliary decision-making method for platform screen door fault processing is implemented.

[0111] Those skilled in the art will understand that Figure 6 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0112] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above-mentioned method embodiments when executing the computer program.

[0113] In an exemplary embodiment, a computer-readable storage medium is provided, storing a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0114] In an exemplary embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0115] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM may be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0116] The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. The non-relational database may include a distributed database based on blockchain, etc., but is not limited thereto. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but is not limited thereto.

[0117] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0118] This article uses specific examples to illustrate the principles and implementation methods of this application. The description of the above embodiments is only used to help understand the method and core ideas of this application. At the same time, for those skilled in the art, according to the ideas of this application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.

Claims

1. An auxiliary decision-making method for platform screen door fault handling, characterized in that: The auxiliary decision-making method for platform screen door fault processing includes: Acquire the platform screen door fault information sent by the station subsystem; the platform screen door fault information is input into the mobile terminal subsystem by the station attendant and sent to the station subsystem, and the platform screen door fault information includes the platform screen door fault type, platform screen door fault location, platform screen door fault time and platform screen door fault phenomenon; Based on the platform screen door fault information and the platform screen door fault processing knowledge graph, the fault processing measures corresponding to each role are obtained respectively; the platform screen door fault processing knowledge graph is constructed according to the obtained platform screen door fault processing related data, and the roles include the station attendant, the station integrated controller and the train dispatcher; The fault handling measures corresponding to the station attendant are sent to the mobile terminal subsystem, the fault handling measures corresponding to the station integrated controller are sent to the station subsystem, and the fault handling measures corresponding to the train dispatcher are displayed.

2. The auxiliary decision-making method for platform screen door fault processing according to claim 1, characterized in that: The platform shielding door fault processing knowledge graph is constructed based on the acquired platform shielding door fault processing related data, specifically including: Acquire data related to the processing of shielding door faults, and classify the data related to the processing of shielding door faults according to data formats to obtain structured data, semi-structured data, and unstructured data; the structured data includes fault records, maintenance logs, and historical cases, the semi-structured data includes an encyclopedia of knowledge related to shielding door faults, and the unstructured data includes operating procedures, interview audio files, fault handling manuals, and authority and responsibility regulations for different roles; Using D2R technology to perform knowledge extraction operation on the structured data to obtain a first data result; Using a wrapper method to perform a knowledge extraction operation on the semi-structured data to obtain a second data result; Using natural language processing technology to perform knowledge extraction operations on the unstructured data to obtain a third data result; A BERT model is used to perform a knowledge fusion operation on the first data result, the second data result and the third data result to obtain the knowledge graph of the shielding door fault processing; the first data result, the second data result and the third data result are all RDF format data.

3. The auxiliary decision-making method for platform screen door fault processing according to claim 2, characterized in that: The third data result includes entities, role annotations and relationships; the third data result is obtained by performing a knowledge extraction operation on the unstructured data using natural language processing technology, specifically including: The Skip-gram model in Word2Vec is used to vectorize the words in the unstructured data to obtain the word vectors of entities, the word vectors of role labels, and the word vectors of authority and responsibility; The word vector of the entity, the word vector of the role label, and the word vector of the authority and responsibility are merged by using a weighted average method to obtain a unified first vector; Normalizing the first vector to obtain a second vector; The second vector is input into an entity recognition model to obtain entities and corresponding role labels, and the second vector is input into a relationship extraction model to obtain relationships between entities.

4. The auxiliary decision-making method for platform screen door fault processing according to claim 3 is characterized in that: The entity recognition model includes a bidirectional long short-term memory network, a conditional random field and a fully connected layer connected in sequence, wherein the bidirectional long short-term memory network is used to process the second vector to obtain a label probability matrix; the conditional random field is used to process the label probability matrix to obtain an optimal entity sequence; the fully connected layer is used to process the optimal entity sequence to obtain a role label corresponding to each entity; The relationship extraction model is a text convolutional neural network.

5. The auxiliary decision-making method for platform screen door fault processing according to claim 1, characterized in that: The fault handling measures corresponding to each role are obtained based on the screen door fault information and the screen door fault handling knowledge graph, specifically including: According to the shielding door fault information, a plurality of fault handling plans corresponding to each role are automatically retrieved and inferred from the shielding door fault handling knowledge graph using rules and an inference engine; According to the evaluation indicators of each fault handling plan, a comprehensive score of each fault handling plan is calculated, and the comprehensive scores of each fault handling plan in the multiple fault handling plans corresponding to each role are ranked to obtain a ranking result; the evaluation indicators include success rate, average resolution time and resource consumption; According to the ranking results, each fault handling plan is assigned a corresponding grade and a corresponding recommendation strength; Determine the fault handling measures corresponding to each role; the fault handling measures corresponding to each role are determined based on the level of each fault handling plan, the recommended strength and the actual situation.

6. The platform screen door fault handling auxiliary decision-making method according to claim 5, characterized in that: According to the evaluation indicators of each fault handling plan, the comprehensive score of each fault handling plan is calculated, including: Obtain historical fault handling data; Based on the fault type, fault location, impact scope and urgency, the K-Means++ algorithm is used to cluster the historical fault processing data, and similar historical fault cases in the historical fault processing data are classified into one category. According to the fault type in each category of historical fault cases, the corresponding fault processing plan is executed by simulation method, and the success rate, average resolution time and resource consumption of each fault processing plan are recorded; The comprehensive score of each troubleshooting plan is calculated based on the success rate, average resolution time and resource consumption of each troubleshooting plan.

7. The auxiliary decision-making method for platform screen door fault processing according to claim 1, characterized in that: The auxiliary decision-making method for platform screen door fault processing also includes: According to the operation command input by the train dispatcher, a dispatching control instruction is output, and the dispatching control instruction is sent to the train automatic monitoring system, and the broadcasting system and the passenger information system are linked at the same time; the operation command input by the train dispatcher is obtained based on the corresponding fault handling measures of the train dispatcher, and the train automatic monitoring system is used to execute the car detention operation, release the car detention operation and adjust the operation diagram operation according to the dispatching control instruction, and the broadcasting system and the passenger information system are used to publish corresponding broadcast information according to the dispatching control instruction.

8. An auxiliary decision-making method for handling platform screen door faults, characterized in that: The auxiliary decision-making method for platform screen door fault processing includes: The mobile terminal subsystem collects the platform screen door fault information input by the station attendant and sends it to the station subsystem; the platform screen door fault information includes the platform screen door fault type, platform screen door fault location, platform screen door fault time and platform screen door fault phenomenon; The station subsystem sends the platform screen door fault information to the dispatching center subsystem; The dispatch center subsystem obtains the fault handling measures corresponding to each role based on the platform screen door fault information and the platform screen door fault handling knowledge graph; the platform screen door fault handling knowledge graph is constructed according to the acquired platform screen door fault handling related data, and the roles include the station attendant, station integrated controller and train dispatcher; The dispatch center subsystem sends the fault handling measures corresponding to the station attendant to the mobile terminal subsystem, sends the fault handling measures corresponding to the station controller to the station subsystem, and displays the fault handling measures corresponding to the train dispatcher.

9. An auxiliary decision-making system for platform screen door fault handling, characterized in that: The auxiliary decision-making system for platform screen door fault processing includes: The mobile terminal subsystem is used to collect the platform screen door fault information input by the station attendant and send it to the station subsystem; the platform screen door fault information includes the platform screen door fault type, platform screen door fault location, platform screen door fault time and platform screen door fault phenomenon; The station subsystem is used to send the platform screen door fault information to the dispatching center subsystem; The dispatch center subsystem is used to: Based on the platform screen door fault information and the platform screen door fault processing knowledge graph, the fault processing measures corresponding to each role are obtained respectively; the platform screen door fault processing knowledge graph is constructed according to the obtained platform screen door fault processing related data, and the roles include the station attendant, the station integrated controller and the train dispatcher; The fault handling measures corresponding to the station attendant are sent to the mobile terminal subsystem, the fault handling measures corresponding to the station integrated controller are sent to the station subsystem, and the fault handling measures corresponding to the train dispatcher are displayed.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the auxiliary decision-making method for platform screen door fault processing described in any one of claims 1 to 7 is implemented.

Images