A multi-system fusion management system for rail transit

Abstract

The present invention relates to the field of control management technology, and discloses a multi-system fusion management system for the field of rail transit; the method comprises fusing target subsystems into an integrated control system, calculating fusion coefficients of subchannels, identifying target channels from subchannels, collecting integrated fusion data of target channels in sub-periods, predicting the system fusion state of the next sub-period, marking optimized subsystems from the target system, and formulating optimization strategies for the optimized subsystems; compared with the prior art, the present invention can fuse multiple independent and parallel subsystems into a complete integrated control system, so that a high correlation can be maintained between the multiple subsystems, and early warning can be given before abnormal phenomena occur in the integrated control system, thereby avoiding the lag caused by real-time monitoring and analysis, and effectively avoiding possible failure phenomena of the integrated control system in the future.

Description

Technical Field

[0001] The present invention relates to the field of control management technology, and more specifically, to a multi-system fusion management system for rail transit. Background Art

[0002] With the rapid development of urban rail transit systems, the rail transit industry faces major challenges in efficiently and accurately managing multiple business subsystems. In order to achieve the goal of coordinated management among multiple systems, it is necessary to integrate multiple systems into one system and achieve efficient management of rail transit through an integrated management approach.

[0003] The patent application with reference publication number CN114611726A discloses a cloud platform-based urban rail transit data fusion control system, including a passenger service center communicating with a driving command center through an internal service network, a passenger service center communicating with a non-urban rail transit transportation system through an external service network, and an operation and maintenance management center communicating with a driving command center through a production safety network. The driving command center includes a line network center for globally monitoring rail transit passenger flow and generating intelligent dispatching strategies, a line control center for executing train dispatching according to the intelligent dispatching strategies, and a digital interlocking system for realizing train interlocking during train dispatching;

[0004] In the existing rail transit multi-system integration management, the actual status of the multi-system integrated operation is judged by making multiple subsystems interact collaboratively and analyzing and evaluating the integration management effect of the multiple systems in real time. For example, in the above-mentioned patent application, a cloud platform built by big data and artificial intelligence technology is used to achieve comprehensive collaborative interaction of the driving control system, and then the effect of multi-system integration is analyzed and evaluated in real time. However, this method has the following shortcomings: on the one hand, after the integration of multiple subsystems, there is a lack of data transfer and interaction measures, which makes it easy for data islands to exist between data of different dimensions. On the other hand, the real-time analysis and evaluation method cannot predict the integrated operation status at future moments in advance, which makes the real-time analysis and evaluation method have a lag, and cannot avoid the upcoming abnormal operation status phenomenon, which reduces the effect of multi-system integrated management.

[0005] In view of this, the present invention proposes a multi-system fusion management system for the rail transit field to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art and to achieve the above-mentioned purpose, the present invention provides the following technical solution: a multi-system fusion management system for the field of rail transit, applied to a fusion management platform, comprising:

[0007] The system fusion module is used to select the target subsystem from the original subsystem of rail transit and fuse the target subsystem into a comprehensive control system based on the system fusion criteria;

[0008] The channel identification module is used to collect the channel parameters of the sub-channels in the integrated control system one by one. The channel parameters include unit transmission volume and minimum transmission time efficiency, calculate the fusion coefficient of the sub-channel, and identify the target channel from the sub-channel. The target channel is the sub-channel corresponding to the maximum value of the fusion coefficient;

[0009] The data acquisition module is used to divide the management cycle into sub-periods and collect the comprehensive fusion data of the target channel in the sub-periods. The comprehensive fusion data includes the video card frame rate, communication delay value, security trigger value and power balance duration;

[0010] The abnormality determination module is used to input the collected comprehensive fusion data into the system fusion prediction model, predict the system fusion state of the next sub-period, the system fusion state includes a normal fusion state and an abnormal fusion state, and determine whether to issue an abnormal state prompt;

[0011] The fusion optimization module is used to mark the optimization subsystem from the target system and formulate the optimization strategy for the optimization subsystem.

[0012] Further, the target subsystems include a video subsystem, a communication subsystem, a security subsystem, and a power subsystem;

[0013] The screening methods for the video subsystem, communication subsystem, security subsystem and power subsystem are as follows:

[0014] The attribute management system is used to query the attribute notes of all original subsystems one by one, the attribute semantics of the attribute notes are identified one by one through natural language processing technology, and the attribute semantics are split into attribute text and attribute value through word segmentation technology;

[0015] The attribute semantics of the attribute texts video, communication, security and electricity are recorded as video semantics, communication semantics, security semantics and electricity semantics respectively, and the original subsystems corresponding to the video semantics, communication semantics, security semantics and electricity semantics are recorded as video subsystem, communication subsystem, security subsystem and electricity subsystem.

[0016] Furthermore, the system fusion criteria are: only one target subsystem exists in a node;

[0017] The fusion method of the integrated control system is:

[0018] Select the initial system from the system database, establish three upper and lower distributed system layers in the initial system, and annotate the three system layers as node layer, transfer layer and control layer in order from top to bottom;

[0019] Establish four independently distributed nodes in the node layer, and import the video subsystem, communication subsystem, security subsystem and power subsystem into the four nodes one by one to generate video nodes, communication nodes, security nodes and power nodes;

[0020] Taking the video node, communication node, security node and power node as the starting point and the transit layer as the end point, A bidirectional transmission sub-channels are established between the four starting points and the end point, and the A sub-channels corresponding to the video node, communication node, security node and power node are respectively aggregated to generate four sub-channel sets;

[0021] With the transfer layer as the transfer starting point and the control layer as the transfer end point, a two-way transmission transfer channel is established between the transfer starting point and the transfer end point to generate a comprehensive control system.

[0022] Furthermore, the method for collecting the unit transmission volume is:

[0023] Find out the establishment time of each sub-channel one by one, and number the A sub-channels in the set of four sub-channels in ascending order according to the chronological order;

[0024] At time T1, a certain amount of test data is synchronously sent to the A sub-channels of the four sub-channel sets in ascending order of numbering;

[0025] At time T2, the number of test data exported from the A sub-channels of the four sub-channel sets is counted one by one through the transit layer to obtain the transmission volume of the A channels;

[0026] Compare the transmission amounts of A channels with the duration between time T1 and time T2 in sequence to obtain A sub-transmission amounts;

[0027] The expression of sub-transmission volume is:

[0028] ;

[0029] In the formula, For the The first subchannel set The sub-transmission volume of the sub-channel, =1,2,3,4, =1,2...A, For the The first subchannel set The channel transmission volume of sub-channels, is the duration between time T1 and time T2;

[0030] The A sub-transmission amounts of the four sub-channel sets are compared one by one, and the maximum value of the sub-transmission amounts of the four sub-channel sets is recorded as the unit transmission amount.

[0031] Furthermore, the expression of the fusion coefficient is:

[0032] ;

[0033] In the formula, For the The first subchannel set The fusion coefficient of sub-channels, For the The first subchannel set The minimum transmission time of sub-channels, , is the weight factor, , All are greater than 0;

[0034] The target channels include video channel, communication channel, security channel and power channel. The target channels between the video node, communication node, security node and power node and the transit layer are recorded as video channel, communication channel, security channel and power channel respectively.

[0035] Furthermore, the video card frame rate acquisition method is:

[0036] Based on the preset playback duration, the B sub-periods are equally divided into C sampling intervals, and the video images within the sampling intervals are recorded as sampled videos;

[0037] The real-time playback durations of the C sampled videos in the video channel are queried one by one, and the sampled videos whose real-time playback durations are longer than the preset playback durations are recorded as stuck-frame videos;

[0038] The number of stuck frame videos in the B sub-periods is counted one by one to obtain B stuck values, and the B stuck values ​​are compared with the number of sampled videos to obtain B video stuck frame rates;

[0039] The expression for the video card frame rate is:

[0040] ;

[0041] In the formula, For the The video card frame rate for each sub-period, =1,2...B, For the The jam value of each sub-period, is the number of sampled videos.

[0042] Furthermore, the communication delay value is collected by:

[0043] In B sub-periods, mark out communication events and identify them one by one Communication data in a communication event;

[0044] Query one by one by timestamp The sending and receiving time of the communication data, and the time between the sending time and the receiving time is recorded as the real-time communication time, and the Real-time communication duration;

[0045] Will The real-time communication durations are subtracted from the calibrated communication durations one by one, and the differences are accumulated and averaged to obtain the communication delay value;

[0046] The expression of communication delay value is:

[0047] ;

[0048] In the formula, For the The communication delay value of each sub-period, For the The sub-period Real-time communication time, The nominal communication duration.

[0049] Furthermore, the method for collecting the power balance duration is:

[0050] In B sub-periods, the power supply voltage and the power supply current of the power supply state in the power channel are detected in real time by the voltage sensor and the current sensor to obtain the real-time voltage value and the real-time current value;

[0051] The power supply state in which the real-time voltage value is within the calibrated voltage range and the real-time current value is within the calibrated current range is recorded as a balanced state, and the moment when the balanced state occurs in the sub-period is marked as the balanced moment;

[0052] The durations of all balancing moments in the B sub-periods are counted one by one to obtain B power balancing durations.

[0053] Furthermore, the training method of the system fusion prediction model is:

[0054] Pre-collecting multiple groups of comprehensive fusion data and system fusion states corresponding to the comprehensive fusion data;

[0055] The comprehensive fusion data is converted into multiple feature vectors using the sliding window method. The system fusion state is converted into a label corresponding to the comprehensive fusion data according to the sliding step. The normal fusion state is marked as 0, and the abnormal fusion state is marked as 1. One feature vector corresponds to one label and constitutes a set of training data. Multiple sets of training data constitute a training set. The comprehensive fusion data are arranged in the order of acquisition time, and the prediction time step, sliding step and sliding window length are preset.

[0056] The feature vector is used as the input of the model, the system fusion state of the next sub-period after the prediction time step is used as the output of the model, the subsequent system fusion state of each training set is used as the prediction target, and the model is trained with the minimized sum of prediction errors as the training target to obtain a system fusion prediction model that predicts the system fusion state of the next sub-period;

[0057] The method for determining whether to issue a status abnormality prompt is as follows:

[0058] When the output of the system fusion prediction model is 0, the predicted system fusion state of the next sub-period is a normal fusion state, and it is determined that no abnormal state prompt will be issued;

[0059] When the output of the system fusion prediction model is 1, the predicted system fusion state of the next sub-period is an abnormal fusion state, and it is determined to issue an abnormal state prompt.

[0060] Furthermore, the marking method of the optimization subsystem is:

[0061] Compare the video card frame rate, communication delay value, security trigger value and power balance duration with the corresponding safety values;

[0062] When the video card frame rate is greater than the card pin safety value, the video subsystem is recorded as an optimized subsystem;

[0063] When the communication delay value is greater than the delay safety value, the communication subsystem is recorded as an optimized subsystem;

[0064] When the security trigger value is greater than the trigger safety value, the security subsystem is recorded as an optimized subsystem;

[0065] When the power balance duration is less than the balance safety value, the power subsystem is recorded as an optimized subsystem;

[0066] The optimization strategy is formulated as follows:

[0067] When the optimized subsystem is a video subsystem, a strategy for reducing the frame rate of the video card is developed;

[0068] When the optimized subsystem is a communication subsystem, a strategy for reducing the communication delay value is developed;

[0069] When the optimization subsystem is a security subsystem, a strategy for reducing the security trigger value is developed;

[0070] When the optimization subsystem is the power subsystem, a strategy for improving the power balance duration is developed.

[0071] The technical effects and advantages of the multi-system fusion management system used in the field of rail transportation of the present invention are as follows:

[0072] The present invention selects the target subsystem from the original subsystem of rail transit, and fuses the target subsystem into an integrated control system based on the system fusion criterion, so that multiple independent and parallel subsystems in rail transit can be integrated into a complete integrated control system, so that the multiple subsystems can maintain a high correlation, and avoid the phenomenon of data islands between the multiple subsystems. By collecting the channel parameters of the subchannels in the integrated control system one by one, calculating the fusion coefficient of the subchannel, and identifying the target channel from the subchannel, the object of the integrated control system operation status evaluation data collection can be clearly defined, ensuring that the collected data can maintain rationality, accuracy and pertinence, and by collecting the comprehensive fusion data of the target channel in the sub-time period, combining the system fusion prediction model to predict the system fusion state of the next sub-time period, the integrated operation state of the integrated control system at the future moment can be predicted in advance, and an early warning can be given before the abnormal phenomenon occurs in the integrated control system, thereby avoiding the hysteresis caused by the real-time monitoring and analysis method, and by formulating an optimization strategy for optimizing the subsystem, the possible failure phenomenon of the integrated control system in the future can be effectively avoided, ensuring that the integrated control system maintains a normal and stable operation state, and improving the multi-system fusion management and control effect of rail transit. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 A schematic diagram of a multi-system fusion management system for rail transit provided in the first embodiment of the present invention;

[0074] Figure 2 A flowchart of a multi-system integration management method for rail transit provided in the second embodiment of the present invention;

[0075] Figure 3 A schematic diagram of the structure of an electronic device provided in Embodiment 3 of the present invention;

[0076] Figure 4 A schematic diagram of the structure of a computer-readable storage medium provided in Embodiment 4 of the present invention. DETAILED DESCRIPTION

[0077] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0078] Example 1: Please refer to Figure 1 As shown, the multi-system fusion management system for rail transit described in this embodiment is applied to the fusion management platform, including:

[0079] The system fusion module selects the target subsystem from the original subsystems of rail transit and integrates the target subsystem into a comprehensive control system based on the system fusion criteria;

[0080] The original subsystem refers to the control system used in rail transit to collect and analyze multi-dimensional data such as video, images, and text, and to ensure that rail transit can operate normally, so that the original subsystem can fully represent all aspects of rail transit;

[0081] The target subsystem refers to the subsystem in the original subsystem that can affect the normal and safe operation of rail transit, and serves as the basis for the subsequent integration of the integrated control system. Since a certain number of target subsystems need to be included in the subsequent integrated control system, the number of target subsystems is usually greater than 1, and each target subsystem can separately represent a certain aspect of rail transit;

[0082] The target subsystems include video subsystem, communication subsystem, security subsystem and power subsystem; the video subsystem is a system that can monitor the real-time images inside and outside the rail transit train compartments; the communication subsystem is a system that can control the wireless communication between the rail transit train compartments and between the train and the platform; the security subsystem is a system that can monitor the status of various security monitoring equipment of the rail transit train; the power subsystem is a system that can monitor the power supply of the rail transit train;

[0083] The screening methods for the video subsystem, communication subsystem, security subsystem and power subsystem are as follows:

[0084] The attribute notes of all original subsystems are queried one by one through the attribute management system, the attribute semantics of the attribute notes are identified one by one through natural language processing technology, and the attribute semantics are split into attribute text and attribute value through word segmentation technology; the attribute note is the note text used to represent the attribute category of the original subsystem, and the attribute semantics is the specific text that can concisely and accurately represent the attribute note, so as to facilitate the subsequent accurate screening of the target subsystem;

[0085] The attribute semantics whose attribute text is video is recorded as video semantics, and the original subsystem corresponding to the video semantics is recorded as video subsystem;

[0086] The attribute semantics whose attribute literal is communication is recorded as communication semantics, and the original subsystem corresponding to the communication semantics is recorded as communication subsystem;

[0087] The attribute semantics whose attribute text is security is recorded as security semantics, and the original subsystem corresponding to the security semantics is recorded as security subsystem;

[0088] The attribute semantics whose attribute text is electricity is recorded as electricity semantics, and the original subsystem corresponding to the electricity semantics is recorded as electricity subsystem.

[0089] After the target subsystem is acquired, the target subsystem can be used as the fusion basis to integrate multiple target subsystems into an integrated comprehensive control system, so that the comprehensive control system can be used as a direct object for subsequent judgment of the operation status of rail transit;

[0090] When integrating the target subsystems into an integrated control system, it is necessary to ensure the integration orderliness of the target subsystems under the constraints of the system integration criteria and improve the integration quality and stability of the subsequent integrated control system;

[0091] The system fusion principle is: there is only one target subsystem in one node; this ensures the relative independence of each target subsystem in the integrated control system, avoids data crossover between multiple target subsystems, and improves the stability of the integrated control system;

[0092] The fusion method of the integrated control system is:

[0093] The initial system is selected from the system database, and three system layers distributed vertically are established in the initial system. The three system layers are annotated as node layer, transfer layer and control layer in a top-down manner. The initial system refers to the system framework without substantial content pre-stored in the coefficient database. The node layer, transfer layer and control layer are application layers in the initial system used for importing target subsystems, data transfer and external interaction, and are used to realize different functions and effects.

[0094] Establish four independently distributed nodes in the node layer, and import the video subsystem, communication subsystem, security subsystem and power subsystem into the four nodes one by one to generate video nodes, communication nodes, security nodes and power nodes;

[0095] Taking the video node, communication node, security node and power node as the starting point and the transit layer as the end point, A bidirectional transmission sub-channels are established between the four starting points and the end point, and the A sub-channels corresponding to the video node, communication node, security node and power node are respectively summarized to generate four sub-channel sets; the sub-channel is a channel used to transmit and interact with the data in the video node, communication node, security node and power node to the transit layer, ensuring the fusion and interaction effect of data of different dimensions between different target subsystems, and avoiding the phenomenon of data islands;

[0096] With the transfer layer as the transfer starting point and the control layer as the transfer end point, a two-way transfer channel is established between the transfer starting point and the transfer end point to generate a comprehensive control system. The transfer channel is used to interact with the outside world for the fused data of different dimensions, ensuring that the data in different target subsystems can interact efficiently and accurately.

[0097] The channel identification module collects the channel parameters of the sub-channels in the integrated control system one by one, calculates the fusion coefficients of the sub-channels, and identifies the target channel from the sub-channels;

[0098] Channel parameters refer to data that can represent the transmission stability and real-time performance of a sub-channel, and serve as the basis for subsequent evaluation of the security and reliability of the sub-channel;

[0099] Channel parameters include unit transmission volume and minimum transmission time;

[0100] The unit transmission volume refers to the amount of data that a subchannel can transmit in a unit time, which can be used to represent the performance of the subchannel's transmission data volume. The larger the unit transmission volume, the stronger the performance of the subchannel's transmission data volume, and the larger the fusion coefficient.

[0101] The method for collecting unit transmission volume is:

[0102] Find out the establishment time of each sub-channel one by one, and number the A sub-channels in the set of four sub-channels in ascending order according to the chronological order;

[0103] At time T1, a certain amount of test data is synchronously sent to the A sub-channels of the four sub-channel sets in ascending order of numbering;

[0104] At time T2, the number of test data exported from A sub-channels is counted one by one through the transit layer to obtain the transmission volume of A channels;

[0105] Compare the transmission amounts of A channels with the duration between time T1 and time T2 in sequence to obtain A sub-transmission amounts;

[0106] The expression of sub-transmission volume is:

[0107] ;

[0108] In the formula, For the The first subchannel set The sub-transmission volume of the sub-channel, =1,2,3,4, =1,2...A, For the The first subchannel set The channel transmission volume of sub-channels, is the duration between time T1 and time T2;

[0109] The A sub-transmission amounts of the four sub-channel sets are compared one by one, and the maximum value of the sub-transmission amounts of the four sub-channel sets is recorded as the unit transmission amount.

[0110] The minimum transmission time refers to the minimum value of the transmission time when the sub-channel transmits data, which can represent the transmission time performance of the data in the sub-channel. The larger the minimum transmission time, the worse the data transmission time performance of the sub-channel, and the smaller the fusion coefficient. The minimum transmission time is obtained by querying the transmission time of the test data of A sub-channels in the set of four sub-channels and taking the minimum value.

[0111] The fusion coefficient is used to represent the safety and reliability of the sub-channel and serves as a direct basis for subsequent identification of the target channel;

[0112] The expression of the fusion coefficient is:

[0113] ;

[0114] In the formula, For the The first subchannel set The fusion coefficient of sub-channels, For the The first subchannel set The minimum transmission time of sub-channels, , is the weight factor, , All are greater than 0;

[0115] in, , , The setting is to balance the proportion of unit transmission volume and minimum transmission time in the fusion coefficient, so as to ensure that changes in the unit transmission volume and minimum transmission time can cause corresponding changes in the fusion coefficient.

[0116] The target channel is the sub-channel corresponding to the maximum value of the fusion coefficient in the sub-channel set, so that the target channel can be used as the data acquisition channel for the subsequent evaluation of the fusion operation status of the integrated control system to ensure that the data collected for subsequent evaluation can maintain accuracy. Therefore, four target channels can be identified from the four sub-channel sets. The target channels include video channel, communication channel, security channel and power channel. The target channels between the video node, communication node, security node and power node and the transit layer are recorded as video channel, communication channel, security channel and power channel respectively.

[0117] The data acquisition module divides the management cycle into sub-periods and collects the comprehensive fusion data of the target channel in the sub-periods. The comprehensive fusion data includes the video card frame rate, communication delay value, security trigger value and power balance duration;

[0118] The management cycle refers to the time taken for a complete data collection, calculation, analysis, instruction formulation and instruction execution process of the integrated control system, which can be used as a time limit for the effective control of the integrated control system of rail transit. Therefore, the time corresponding to the management cycle is set according to the maximum value of the time taken for a complete data collection, calculation, analysis, instruction formulation and instruction execution process of the integrated control system of rail transit in history;

[0119] Sub-periods refer to the division of a management cycle with a larger time span into periods with smaller time spans. In this case, the time span corresponding to the sub-period is smaller. In order to ensure that the sub-periods can represent the management cycle in a full and continuous manner, it is necessary that the duration corresponding to each sub-period is consistent and that the two adjacent sub-periods are continuous on the timeline, which can ensure that the management cycle can be divided into B sub-periods of equal length and continuity.

[0120] Comprehensive fusion data refers to data that can be collected through the target channel and can represent the fusion operation status of the integrated control system in multiple dimensions, and serve as the basis for subsequent judgment on whether the fusion operation status of the integrated control system will be abnormal;

[0121] Comprehensive fusion data includes video card frame rate, communication delay value, security trigger value and power balance duration;

[0122] The video card frame rate refers to the number of freezes in the monitoring video image in the video channel within a sub-period per unit time, which can be used to indicate the smooth playback performance of the video image in the integrated control system. The larger the video card frame rate, the worse the smooth playback performance of the video image in the integrated control system.

[0123] The video card frame rate acquisition method is:

[0124] Based on the preset playback duration, the B sub-periods are equally divided into C sampling intervals, and the video images in the sampling intervals are recorded as sampled videos; the preset playback duration is used to specifically limit the duration span corresponding to the sampling intervals, and to limit the duration of subsequent sampled videos, so as to ensure that the duration corresponding to each sampled video is consistent. Specifically, in order to ensure that the sub-periods can be divided into a sufficient number of sampling intervals, the preset playback duration should be less than one twentieth of the duration of the sub-period;

[0125] The real-time playback durations of the C sampled videos in the video channel are queried one by one, and the sampled videos whose real-time playback durations are longer than the preset playback durations are recorded as stuck-frame videos;

[0126] The number of stuck frame videos in the B sub-periods is counted one by one to obtain B stuck values, and the B stuck values ​​are compared with the number of sampled videos to obtain B video stuck frame rates;

[0127] The expression for the video card frame rate is:

[0128] ;

[0129] In the formula, For the The video card frame rate for each sub-period, =1,2...B, For the The jam value of each sub-period, is the number of sampled videos.

[0130] The communication delay value refers to the length of time that the transmission time of the communication data in the communication channel in the sub-period exceeds the calibration time, which can be used to represent the real-time performance of communication in the integrated control system. The larger the communication delay value, the worse the real-time performance of communication in the integrated control system.

[0131] The communication delay value is collected by:

[0132] In B sub-periods, mark out communication events and identify them one by one The communication data in a communication event; the communication event is used to represent the overall communication interaction process in the communication subsystem within the communication channel, and the communication data refers to the specific data in the communication event that directly represents the communication interaction and is one of the parts that constitute the communication event;

[0133] Query one by one by timestamp The sending and receiving time of the communication data, and the time between the sending time and the receiving time is recorded as the real-time communication time, and the Real-time communication duration;

[0134] Will The real-time communication duration is subtracted from the calibrated communication duration one by one, and the differences are accumulated and averaged to obtain the communication delay value; the calibrated communication duration refers to the maximum value of the real-time communication duration when there is no communication delay, which can ensure the accuracy of the calculation of subsequent communication delay values;

[0135] The expression of communication delay value is:

[0136] ;

[0137] In the formula, For the The communication delay value of each sub-period, For the The sub-period Real-time communication time, The nominal communication duration.

[0138] The security trigger value refers to the number of security defense prompts triggered in the security channel within a sub-period, which can represent the security defense performance in the integrated control system. The larger the security trigger value, the worse the security defense performance in the integrated control system. The security trigger value is obtained by counting the total number of security defense prompts triggered in B sub-periods in the security channel.

[0139] The power balance duration refers to the total duration of the power supply in the power channel in the sub-period being in a balanced state, which can represent the stability performance of the power supply in the integrated control system. The longer the power balance duration is, the stronger the stability performance of the power supply in the integrated control system is.

[0140] The method for collecting power balance duration is:

[0141] In B sub-periods, the power supply voltage and the power supply current of the power supply state in the power channel are detected in real time by the voltage sensor and the current sensor to obtain the real-time voltage value and the real-time current value;

[0142] The power supply state in which the real-time voltage value is within the calibrated voltage range and the real-time current value is within the calibrated current range is recorded as a balanced state, and the moment when the balanced state occurs in the sub-period is marked and recorded as the balanced moment; the calibrated voltage range refers to the range between the minimum voltage and the maximum voltage when the power supply voltage of the power subsystem is not in an unbalanced state, and the calibrated current range refers to the range between the minimum current and the maximum current when the power supply current of the power subsystem is not in an unbalanced state, which can provide a direct basis for judging whether the power supply state is balanced;

[0143] The durations of all balancing moments in the B sub-periods are counted one by one to obtain B power balancing durations.

[0144] It should be noted that the video card frame rate, communication delay value, security trigger value and power balance duration are used to represent the data of the four dimensions of video, communication, security and power in the integrated control system respectively.

[0145] The abnormality determination module inputs the collected comprehensive fusion data into the system fusion prediction model, predicts the system fusion status of the next sub-period, and determines whether to issue a status abnormality prompt;

[0146] The system fusion prediction model refers to a machine learning model that can make specific and accurate predictions on the fusion operation status of the integrated control system at future moments, so that the system fusion prediction model uses the integrated fusion data as input data and the system fusion status of the next sub-period as output data, and is used to provide evaluation and analysis operations for the integrated control system of rail transit;

[0147] The system fusion status is a specific indication of the fusion operation status of the integrated control system, and serves as a basis for subsequent optimization and adjustment of the integrated system of rail transit. The system fusion status includes normal fusion status and abnormal fusion status. Normal fusion status means that the fusion operation status of the integrated control system is normal, and abnormal fusion status means that the fusion operation status of the integrated control system is abnormal. The system fusion status is obtained by collecting a large number of historical normal fusion status and abnormal fusion status corresponding to the video card frame rate, communication delay value, security trigger value and power balance duration.

[0148] The training method of the system fusion prediction model is:

[0149] Pre-collecting multiple groups of comprehensive fusion data and system fusion states corresponding to the comprehensive fusion data;

[0150] The comprehensive fusion data is converted into multiple feature vectors using a sliding window method, and the system fusion state is converted into a label corresponding to the comprehensive fusion data according to the sliding step, and the label is digitally labeled. For example, the normal fusion state is labeled as 0, and the abnormal fusion state is labeled as 1. One feature vector corresponds to one label and constitutes a set of training data. Multiple sets of training data constitute a training set. The comprehensive fusion data are arranged in the order of acquisition time, and the prediction time step, sliding step and sliding window length are preset;

[0151] The feature vector is taken as the input of the model, the system fusion state of the next sub-period after the prediction time step is taken as the output of the model, the subsequent system fusion state of each training set is taken as the prediction target, and the model is trained with the minimized sum of prediction errors as the training target to obtain a system fusion prediction model that predicts the system fusion state of the next sub-period.

[0152] Exemplarily, the system fusion prediction model adopts any one of CNN or AlexNet;

[0153] The calculation formula for the prediction error is:

[0154] ;

[0155] In the formula, is the prediction error, is the group number of the eigenvector; For the The predicted state value corresponding to the group feature vector, For the The actual state value corresponding to the group training data.

[0156] Input the collected comprehensive fusion data into the system fusion prediction model to predict the system fusion state of the next sub-period, and determine whether the integrated control system has abnormal fusion operation state based on the predicted system fusion state, and issue a state abnormality prompt in a timely manner when an abnormal phenomenon occurs;

[0157] The method for determining whether to issue a status abnormality prompt is as follows:

[0158] When the output of the system fusion prediction model is 0, it means that the predicted system fusion state in the next sub-period is a normal fusion state. At this time, the integrated control system will not have abnormal operation in the next sub-period, and it is determined that no abnormal state prompt will be issued;

[0159] When the output of the system fusion prediction model is 1, it means that the predicted system fusion state of the next sub-period is an abnormal fusion state. At this time, the integrated control system will have abnormal operation in the next sub-period, and it is determined to issue a state abnormality prompt.

[0160] Fusion optimization module, marking the optimization subsystem from the target system and developing the optimization strategy for the optimization subsystem;

[0161] The optimization subsystem refers to the target system corresponding to the specific integrated fusion data that causes the system fusion state of the integrated control system to be abnormal in the next sub-period, and serves as the object of subsequent multi-system fusion management of rail transit;

[0162] The marking method of the optimization subsystem is:

[0163] Compare the video card frame rate, communication delay value, security trigger value and power balance duration with the corresponding safety values;

[0164] When the video card frame rate is greater than the card pin safety value, it means that the video card frame rate will cause the system fusion state of the integrated control system to be abnormal in the next sub-period, and the video subsystem will be recorded as an optimized subsystem; the card pin safety value refers to the maximum value of the video card frame rate of the integrated control system when no abnormal state prompt is issued, which is the basis for judging whether the video subsystem needs to be optimized;

[0165] When the communication delay value is greater than the delay safety value, it means that the communication delay value will cause the system fusion state of the integrated control system to be abnormal in the next sub-period, and the communication subsystem will be recorded as an optimized subsystem; the delay safety value refers to the maximum value of the communication delay value of the integrated control system when no abnormal state prompt is issued, which is the basis for judging whether the communication subsystem needs to be optimized;

[0166] When the security trigger value is greater than the trigger safety value, it means that the security trigger value will cause the system fusion state of the integrated control system to be abnormal in the next sub-period, and the security subsystem will be recorded as an optimized subsystem; the trigger safety value refers to the maximum value of the security trigger value of the integrated control system when no abnormal state prompt is issued, which is the basis for judging whether the security subsystem needs to be optimized;

[0167] When the power balance duration is less than the balance safety value, it means that the power balance duration will cause the system fusion state of the integrated control system to be abnormal in the next sub-period, and the power subsystem will be recorded as an optimized subsystem. The balance safety value refers to the minimum value of the power balance duration of the integrated control system when no abnormal status prompt is issued, which is the basis for judging whether the power subsystem needs to be optimized.

[0168] When the optimization subsystem is marked, it is necessary to formulate an optimization strategy for optimizing the optimization subsystem according to the different types of the optimization subsystem, so that the optimization strategy can provide guidance for the optimization adjustment of the optimization subsystem, ensure the normal and safe operation of the integrated control system, and improve the effect of multi-system integration management;

[0169] The optimization strategy is formulated as follows:

[0170] When the optimized subsystem is a video subsystem, it is necessary to optimize and improve the phenomenon of video screen freeze in the video subsystem to improve the smoothness of the video screen in the video subsystem, and then formulate a strategy to reduce the video card frame rate;

[0171] When the optimized subsystem is a communication subsystem, it is necessary to optimize and improve the transmission time of communication data in the communication subsystem, increase the transmission rate of communication data in the communication subsystem, and formulate a strategy to reduce the communication delay value;

[0172] When the optimization subsystem is a security subsystem, it is necessary to optimize and improve the security defense measures in the security subsystem to improve the stability of the security protection measures of the security subsystem, and then formulate a strategy to reduce the security trigger value;

[0173] When the optimization subsystem is the power subsystem, it is necessary to optimize and improve the power supply balance state in the power subsystem, improve the stability of the power supply voltage and current, and formulate a strategy to increase the power balance time.

[0174] In this embodiment, by screening out the target subsystem from the original subsystem of rail transit and integrating the target subsystem into an integrated control system based on the system fusion criteria, multiple independent and parallel subsystems in rail transit can be integrated into a complete integrated control system, so that the multiple subsystems can maintain a high correlation and avoid the phenomenon of data islands between multiple subsystems. By collecting the channel parameters of the subchannels in the integrated control system one by one, calculating the fusion coefficient of the subchannel, and identifying the target channel from the subchannel, the object of the integrated control system operation status evaluation data collection can be clearly defined to ensure that the collected data can maintain rationality, accuracy and pertinence. By collecting the comprehensive fusion data of the target channel in the sub-time period and combining the system fusion prediction model to predict the system fusion state of the next sub-time period, the integrated operation state of the integrated control system at the future moment can be predicted in advance, and an early warning can be given before abnormal phenomena occur in the integrated control system, thereby avoiding the lag caused by real-time monitoring and analysis. By formulating an optimization strategy for optimizing the subsystem, the possible failure of the integrated control system in the future can be effectively avoided, ensuring that the integrated control system maintains a normal and stable operation state, and improving the multi-system fusion management and control effect of rail transit.

[0175] Example 2: Please refer to Figure 2 As shown, the part not described in detail in this embodiment is described in the first embodiment, which provides a multi-system integration management method for the rail transit field, which is applied to the integration management platform and is implemented based on a multi-system integration management system for the rail transit field, including:

[0176] S1: Select the target subsystem from the original subsystem of rail transit, and integrate the target subsystem into a comprehensive control system based on the system fusion criteria;

[0177] S2: Collect the channel parameters of the sub-channels in the integrated control system one by one, the channel parameters include unit transmission volume and minimum transmission time, calculate the fusion coefficient of the sub-channel, and identify the target channel from the sub-channel. The target channel is the sub-channel corresponding to the maximum value of the fusion coefficient;

[0178] S3: Divide the management cycle into sub-periods and collect the comprehensive fusion data of the target channel in the sub-periods. The comprehensive fusion data includes the video card frame rate, communication delay value, security trigger value and power balance duration;

[0179] S4: input the collected comprehensive fusion data into the system fusion prediction model, predict the system fusion state of the next sub-period, the system fusion state includes a normal fusion state and an abnormal fusion state, and determine whether to issue an abnormal state prompt;

[0180] S5: If a status abnormality prompt is issued, the optimization subsystem is marked from the target system, and an optimization strategy for the optimization subsystem is formulated.

[0181] Example 3: Please refer to Figure 3 As shown, this embodiment discloses an electronic device, including a processor and a memory;

[0182] Wherein, the memory stores a computer program that can be called by the processor;

[0183] The processor executes the multi-system integration management method for the rail transit field by calling the computer program stored in the memory.

[0184] Since the electronic device introduced in this embodiment is an electronic device used to implement a multi-system fusion management method for the rail transit field in Example 2 of this application, based on the multi-system fusion management method for the rail transit field introduced in the embodiment of this application, the technical personnel of this field can understand the specific implementation of the electronic device of this embodiment and its various variations, so how the electronic device implements the method in the embodiment of this application will not be introduced in detail here. As long as the technical personnel of this field implement the electronic device used in the multi-system fusion management method for the rail transit field in the embodiment of this application, it belongs to the scope of protection of this application.

[0185] Example 4: Please refer to Figure 4 As shown, this embodiment discloses a computer-readable storage medium on which a rewritable computer program is stored;

[0186] When the computer program is executed, the multi-system integration management method for the rail transit field is implemented.

[0187] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A multi-system fusion management system for rail transit, applied to a fusion management platform, characterized in that: include: The system fusion module is used to select the target subsystem from the original subsystem of rail transit and fuse the target subsystem into a comprehensive control system based on the system fusion criteria; The channel identification module is used to collect the channel parameters of the sub-channels in the integrated control system one by one. The channel parameters include unit transmission volume and minimum transmission time efficiency, calculate the fusion coefficient of the sub-channel, and identify the target channel from the sub-channel. The target channel is the sub-channel corresponding to the maximum value of the fusion coefficient; The data acquisition module is used to divide the management cycle into sub-periods and collect the comprehensive fusion data of the target channel in the sub-periods. The comprehensive fusion data includes the video card frame rate, communication delay value, security trigger value and power balance duration; The video card frame rate acquisition method is: Based on the preset playback duration, the B sub-periods are equally divided into C sampling intervals, and the video images within the sampling intervals are recorded as sampled videos; The real-time playback durations of the C sampled videos in the video channel are queried one by one, and the sampled videos whose real-time playback durations are longer than the preset playback durations are recorded as stuck-frame videos; The number of stuck frame videos in the B sub-periods is counted one by one to obtain B stuck values, and the B stuck values ​​are compared with the number of sampled videos to obtain B video stuck frame rates; The communication delay value is collected by: In B sub-periods, mark out communication events and identify them one by one Communication data in a communication event; Query one by one by timestamp The sending and receiving time of the communication data, and the time between the sending time and the receiving time is recorded as the real-time communication time, and the Real-time communication duration; Will The real-time communication durations are subtracted from the calibrated communication durations one by one, and the differences are accumulated and averaged to obtain the communication delay value; The method for collecting power balance duration is: In B sub-periods, the power supply voltage and the power supply current of the power supply state in the power channel are detected in real time by the voltage sensor and the current sensor to obtain the real-time voltage value and the real-time current value; The power supply state in which the real-time voltage value is within the calibrated voltage range and the real-time current value is within the calibrated current range is recorded as a balanced state, and the moment when the balanced state occurs in the sub-period is marked as the balanced moment; Count the durations of all balancing moments in B sub-periods one by one to obtain B power balancing durations; The abnormality determination module is used to input the collected comprehensive fusion data into the system fusion prediction model, predict the system fusion state of the next sub-period, the system fusion state includes a normal fusion state and an abnormal fusion state, and determine whether to issue an abnormal state prompt; The fusion optimization module is used to mark the optimization subsystem from the target system and formulate the optimization strategy for the optimization subsystem.

2. A multi-system fusion management system for rail transit according to claim 1, characterized in that: The target subsystems include video subsystem, communication subsystem, security subsystem and power subsystem; The screening methods for the video subsystem, communication subsystem, security subsystem and power subsystem are as follows: The attribute management system is used to query the attribute notes of all original subsystems one by one, the attribute semantics of the attribute notes are identified one by one through natural language processing technology, and the attribute semantics are split into attribute text and attribute value through word segmentation technology; The attribute semantics of the attribute texts video, communication, security and electricity are recorded as video semantics, communication semantics, security semantics and electricity semantics respectively, and the original subsystems corresponding to the video semantics, communication semantics, security semantics and electricity semantics are recorded as video subsystem, communication subsystem, security subsystem and electricity subsystem.

3. The multi-system fusion management system for rail transit according to claim 2 is characterized in that: The system fusion criteria are: there is only one target subsystem on a node; The fusion method of the integrated control system is: Select the initial system from the system database, establish three upper and lower distributed system layers in the initial system, and annotate the three system layers as node layer, transfer layer and control layer in order from top to bottom; Establish four independently distributed nodes in the node layer, and import the video subsystem, communication subsystem, security subsystem and power subsystem into the four nodes one by one to generate video nodes, communication nodes, security nodes and power nodes; Taking the video node, communication node, security node and power node as the starting point and the transit layer as the end point, A bidirectional transmission sub-channels are established between the four starting points and the end point, and the A sub-channels corresponding to the video node, communication node, security node and power node are respectively aggregated to generate four sub-channel sets; With the transfer layer as the transfer starting point and the control layer as the transfer end point, a two-way transmission transfer channel is established between the transfer starting point and the transfer end point to generate a comprehensive control system.

4. A multi-system fusion management system for rail transit according to claim 3, characterized in that: The method for collecting unit transmission volume is: Find out the establishment time of each sub-channel one by one, and number the A sub-channels in the set of four sub-channels in ascending order according to the chronological order; At time T1, a certain amount of test data is synchronously sent to A sub-channels of the four sub-channel sets in ascending order of numbering; At time T2, the number of test data exported from the A sub-channels of the four sub-channel sets is counted one by one through the transit layer to obtain the transmission volume of the A channels; Compare the transmission amounts of A channels with the duration between time T1 and time T2 in sequence to obtain A sub-transmission amounts; The expression of sub-transmission volume is: ； In the formula, For the The first subchannel set The sub-transmission volume of the sub-channel, =1,2,3,4, =1,2...A, For the The first subchannel set The channel transmission volume of sub-channels, is the duration between time T1 and time T2; The A sub-transmission amounts of the four sub-channel sets are compared one by one, and the maximum value of the sub-transmission amounts of the four sub-channel sets is recorded as the unit transmission amount.

5. The multi-system fusion management system for rail transit according to claim 4 is characterized in that: The expression of the fusion coefficient is: ； In the formula, For the The first subchannel set The fusion coefficient of sub-channels, For the The first subchannel set The minimum transmission time of sub-channels, , is the weight factor, , All are greater than 0; The target channels include video channel, communication channel, security channel and power channel. The target channels between the video node, communication node, security node and power node and the transit layer are recorded as video channel, communication channel, security channel and power channel respectively.

6. A multi-system fusion management system for rail transit according to claim 5, characterized in that: The expression for the video card frame rate is: ； In the formula, For the The video card frame rate for each sub-period, =1,2...B, For the The jam value of each sub-period, is the number of sampled videos.

7. A multi-system fusion management system for rail transit according to claim 6, characterized in that: The expression of communication delay value is: ； In the formula, For the The communication delay value of each sub-period, For the The sub-period Real-time communication time, The nominal communication duration.

8. The multi-system fusion management system for rail transit according to claim 7, characterized in that: The training method of the system fusion prediction model is: Pre-collecting multiple groups of comprehensive fusion data and system fusion states corresponding to the comprehensive fusion data; The comprehensive fusion data is converted into multiple feature vectors using the sliding window method. The system fusion state is converted into a label corresponding to the comprehensive fusion data according to the sliding step. The normal fusion state is marked as 0, and the abnormal fusion state is marked as 1. One feature vector corresponds to one label and constitutes a set of training data. Multiple sets of training data constitute a training set. The comprehensive fusion data are arranged in the order of acquisition time, and the prediction time step, sliding step and sliding window length are preset. The feature vector is used as the input of the model, the system fusion state of the next sub-period after the prediction time step is used as the output of the model, the subsequent system fusion state of each training set is used as the prediction target, and the model is trained with the minimized sum of prediction errors as the training target to obtain a system fusion prediction model that predicts the system fusion state of the next sub-period; The method for determining whether to issue a status abnormality prompt is as follows: When the output of the system fusion prediction model is 0, the predicted system fusion state of the next sub-period is a normal fusion state, and it is determined that no abnormal state prompt will be issued; When the output of the system fusion prediction model is 1, the predicted system fusion state of the next sub-period is an abnormal fusion state, and it is determined to issue an abnormal state prompt.

9. A multi-system fusion management system for rail transit according to claim 8, characterized in that: The marking method of the optimization subsystem is: Compare the video card frame rate, communication delay value, security trigger value and power balance duration with the corresponding safety values; When the video card frame rate is greater than the card pin safety value, the video subsystem is recorded as an optimized subsystem; When the communication delay value is greater than the delay safety value, the communication subsystem is recorded as an optimized subsystem; When the security trigger value is greater than the trigger safety value, the security subsystem is recorded as an optimized subsystem; When the power balance duration is less than the balance safety value, the power subsystem is recorded as an optimized subsystem; The optimization strategy is formulated as follows: When the optimized subsystem is a video subsystem, a strategy for reducing the frame rate of the video card is developed; When the optimized subsystem is a communication subsystem, a strategy for reducing the communication delay value is developed; When the optimization subsystem is a security subsystem, a strategy for reducing the security trigger value is developed; When the optimization subsystem is the power subsystem, a strategy for improving the power balance duration is developed.

Images