Switch power failure alarm method and system applied to railway video monitoring system

Abstract

The invention discloses a switch power-off alarm method and system applied to a railway video monitoring system, relates to the technical field of bandwidth management, and is used to improve the problem of low sensitivity of a switch power-off alarm threshold when the network load is low. The method comprises collecting historical data, dividing network load distribution applications according to the historical data, collecting application data, evaluating the network load of each application by using a hierarchical analysis method based on the historical data and the application data, analyzing the overall network load degree by setting different load weights according to the network load of each application, collecting application behavior data after determining the overall network load degree, calculating and analyzing the switch power-off situation by using a logistic regression method based on the overall network load degree and the application behavior data, adjusting the alarm threshold according to the switch power-off situation, collecting feedback data before and after adjusting the alarm threshold, and selecting to improve the alarm threshold processing or perform power-off optimization processing according to the feedback data.

Description

Technical Field

[0001] The present invention relates to the technical field of bandwidth management, and more specifically, to a switch power-off alarm method and system applied to a railway video monitoring system. Background Art

[0002] Bandwidth management technology is a technology that users use to allocate and distribute network bandwidth resources to meet the needs of different applications and users. Bandwidth management technology applied to the switch power failure alarm of the railway video surveillance system can make the power failure alarm more timely and convenient, thereby reducing railway safety risks.

[0003] The prior art has the following deficiencies:

[0004] In the past, the network in the railway video surveillance system had a large amount of traffic business, and a large bandwidth was usually set to accommodate the large network load. When the network load was low, problems such as switch power failure and surveillance video interruption would occur before the alarm threshold was reached, resulting in missing important monitoring information. Summary of the invention

[0005] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a switch power-off alarm method applied to a railway video surveillance system, which dynamically adjusts the alarm threshold and alarm threshold time by analyzing real-time data of drugs, inventory and pharmaceutical factories to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The switch power failure alarm method applied to the railway video monitoring system comprises the following steps:

[0008] Step S1: Collect historical data and divide network load distribution applications;

[0009] Step S2: collect application data, integrate historical data and application data and use the analytic hierarchy process to determine the overall network load level;

[0010] Step S3: collect application behavior data, analyze the switch power failure situation by using logistic regression method in combination with the overall network load level, and adjust the alarm threshold according to the switch power failure situation;

[0011] Step S4: Collect feedback data, and improve the alarm threshold or power-off optimization processing according to the feedback data;

[0012] In step S2, the application data includes the bandwidth utilization rate of each part of the application and the video frame rate of each part. The load estimation of each part of the application is calculated by using the hierarchical analysis method based on the historical data and the application data. The load weight is set according to the load estimation of each part of the application. The overall network load degree is determined by combining the load estimation of each part of the application and the load weight.

[0013] In step S3, the application behavior data is the change in overall power consumption. A period of time is selected in the historical time as the analysis time, and the overall power consumption within the analysis time is calculated. Then, the current overall power consumption is detected and calculated within the same time interval, and the overall power consumption change is obtained by subtracting the two.

[0014] After normalizing the change in overall power consumption, the overall network load factor is combined as the input of the logistic regression model. The calculated result is used as the power-off coefficient. The power-off situation of the switch is analyzed by comparing the power-off coefficient with the preset power-off threshold. If the power-off situation is serious, the alarm threshold is adjusted according to the power-off coefficient.

[0015] The feedback data is the continuous power failure increase rate. The overall power loss change before and after the alarm threshold is adjusted is calculated respectively, and the quotient of the overall power loss change after the alarm threshold is adjusted and the overall power loss change before the alarm threshold is adjusted is taken as the continuous power failure increase rate.

[0016] In step S4, the alarm threshold or power-off optimization is determined by comparing the continuous power-off increase rate with the preset increase threshold. If the continuous power-off increase rate exceeds the increase threshold, it is judged that the power-off amplitude of the switch has deepened, and power-off optimization is performed; if the continuous power-off increase rate is lower than the increase threshold, it is judged that the power-off amplitude of the switch has changed slightly, and the alarm threshold is refined by combining the continuous power-off increase rate and the power-off coefficient.

[0017] In a preferred embodiment, in step S1, the historical data is the power consumption of each application in the railway video surveillance system. Based on the historical data, multiple applications with large power consumption are selected as analysis samples, and the overall network load level is analyzed using the network load of each application in the analysis sample.

[0018] In a preferred embodiment, in step S2, when determining the overall network load level, the calculated load weights of each application and the bandwidth utilization of each application are weighted and summed to obtain the overall network load coefficient, and the overall network load coefficient is compared with a preset overall network load threshold to determine the overall network load level.

[0019] A switch power-off alarm system applied to a railway video surveillance system, used to implement the switch power-off alarm method applied to a railway video surveillance system, comprising a data acquisition module, a power-off analysis module, a confirmation processing module and a feedback adjustment module;

[0020] The data acquisition module is used to collect various types of data and transmit them to different subsequent modules for the normal operation of the subsequent modules;

[0021] The power failure analysis module analyzes the collected different data, determines the overall network load level of the railway video surveillance system, and sends the judgment result to the confirmation processing module;

[0022] The confirmation processing module receives the judgment result and performs different alarm threshold adjustment operations according to the corresponding judgment result;

[0023] The feedback adjustment module is used to receive feedback data and select an improved alarm threshold or power-off optimization according to the feedback data.

[0024] The technical effects and advantages of the switch power failure alarm method and system applied to the railway video monitoring system of the present invention are as follows:

[0025] The present invention collects historical data, divides network load distribution applications according to the historical data, collects application data, evaluates the network load of each application by integrating the historical data and the application data, analyzes the overall network load level according to the network load of each application, collects application behavior data, analyzes the power failure of the switch by integrating the overall network load level and the application behavior data, adjusts the alarm threshold according to the power failure of the switch, collects feedback data, and selects to improve the alarm threshold or perform power failure optimization according to the feedback data. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a schematic diagram of a switch power failure alarm method applied to a railway video monitoring system according to the present invention.

[0027] Figure 2 The present invention is a flow chart of a switch power failure alarm system applied to a railway video monitoring system. DETAILED DESCRIPTION

[0028] 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.

[0029] The present invention collects historical data, divides network load distribution applications according to the historical data, collects application data, evaluates the network load of each application by integrating the historical data and the application data, analyzes the overall network load level according to the network load of each application, collects application behavior data, analyzes the power failure of the switch by integrating the overall network load level and the application behavior data, adjusts the alarm threshold according to the power failure of the switch, collects feedback data, and selects to improve the alarm threshold or perform power failure optimization according to the feedback data.

[0030] Example 1, a switch power failure alarm method applied to a railway video monitoring system, such as Figure 1 As shown, the following steps are included:

[0031] Step S1: Collect historical data and divide network load distribution applications.

[0032] Step S2: Collect application data, integrate historical data and application data, and use the analytic hierarchy process to determine the overall network load level.

[0033] Step S3: collect application behavior data, analyze the switch power failure situation using logistic regression method in combination with the overall network load level, and adjust the alarm threshold according to the switch power failure situation.

[0034] Step S4: Collect feedback data, and perform refined alarm threshold or power-off optimization processing according to the feedback data.

[0035] The specific implementation is as follows:

[0036] In step S1, the historical data is the power consumption of each application in the railway video surveillance system. According to the historical data, N applications with large power consumption are used as analysis samples, and the overall network load is analyzed using the network load of each of the N applications in the analysis samples. In most railway video surveillance systems, video transmission, video storage and real-time monitoring have large power consumption. This example uses video transmission, video storage and real-time monitoring as analysis samples for example:

[0037] Application 1: Video transmission. The high-definition data collected by the video surveillance camera needs to be transmitted to the video processing device through the network. This part of the data traffic is large and occupies network bandwidth resources.

[0038] Application 2: Video storage. The railway video surveillance system needs to store video data on a server or storage device for subsequent playback and query. The storage and playback process also requires large data flow and bandwidth resources.

[0039] Application 3: Real-time monitoring. The railway video monitoring system needs to receive video images and alarm information from each monitoring point in real time. These real-time data will also require a large network load and bandwidth resources.

[0040] It should be noted that historical data can be obtained by accessing the historical database of the railway video surveillance system. The power consumption and network load of each application are sorted by size, and the top N applications are selected as analysis samples. In this example, N is 3, and the application is selected by default.

[0041] In step S2, the application data is used to assess the network load of each part of the application, and the overall network load of the system is judged based on the network load of each part of the comprehensive application. The application data includes the bandwidth utilization of each part of the application and the video frame rate of each part. The higher the bandwidth utilization or the higher the video frame rate, the better the corresponding network load and the higher the network load.

[0042] The bandwidth utilization calculation formula for each application is: Bandwidth utilization = actual bandwidth used / total available bandwidth, where the actual bandwidth used is the amount of data actually transmitted on the network within a certain period of time, which can be obtained through network monitoring equipment or software. The total available bandwidth refers to the maximum theoretical transmission capacity of the network device or link, which is usually determined by the physical bandwidth of the network interface.

[0043] The video frame rate calculation formula for each application is: Video frame rate = total number of frames / video duration, where total number of frames refers to the total number of frames contained in the video file, and video duration refers to the total duration of the video file, usually in seconds.

[0044] When evaluating the network load of each application, historical data and application data are combined to evaluate the network load of each application. The specific steps are as follows:

[0045] According to historical data, the power consumption of each application in the railway video surveillance system is divided by the overall power consumption of the railway video surveillance system to obtain the power consumption coefficient of each application, which is marked as a, and the corresponding bandwidth utilization and video frame rate are marked as b and c. The network load evaluation calculation formula for each application can be: F (i) = a + b + c, where F (i) is the estimated load of each application, and i represents the serial number of each application. It should be explained that the greater the power consumption of the application in the railway video surveillance system, the greater the power consumption coefficient, and the greater the network bandwidth resources it occupies, that is, the greater the estimated network load of the corresponding application, the higher the power loss, bandwidth utilization or video frame rate, and the greater the network load of the corresponding application.

[0046] It should be noted that the network load assessment of each application can also be analyzed using other relevant parameters, which will not be analyzed here. The overall network load level refers to the overall network load utilization of the railway video surveillance system.

[0047] Set the load weight according to the estimated load of each application, analyze the overall network load, sort the estimated load of each application from small to large, use the hierarchical analysis method to set the load weight, and determine the overall network load level based on the load weight. The specific steps are as follows:

[0048] Construct a judgment matrix: Construct a judgment matrix to compare the network load between each application. You can use a 3-point scale from 1 to 3, where 1 indicates the same network load, 2 indicates a larger network load, and 3 indicates a significant network load.

[0049]

[0050] Calculate the eigenvector: obtain the eigenvector of each factor by taking the weighted average of each column, normalize each column of the judgment matrix to ensure that the sum of each column is equal to 1, and calculate the weighted average value as the element value of the eigenvector for each column. The above judgment matrix can be normalized as follows:

[0051]

[0052] The weighted average of each factor is calculated as follows:

[0053] Part A application network load = (0.1667 + 0.1428 + 0.1818) / 3 = 0.1638;

[0054] Part B application network load = (0.3333 + 0.2857 + 0.2727) / 3 = 0.2972;

[0055] Part C application network load = (0.5000 + 0.5715 + 0.5455) / 3 = 0.5390;

[0056] The estimated load values ​​of video transmission, video storage and real-time monitoring calculations in this example are arranged from small to large and correspond to the network load values ​​of parts A, B and C respectively. The calculated load weights of each application and the weighted sum of the bandwidth utilization of each part are used to obtain the overall network load coefficient. The overall network load coefficient is compared with the preset overall network load threshold to determine the overall network load level.

[0057] When the overall network load coefficient exceeds a preset overall network load threshold, the overall network load level is determined to be high; otherwise, the overall network load level is determined to be low.

[0058] Through the hierarchical analysis method, the importance of different functions in the railway video surveillance system and their proportion in the overall network load are determined according to the network load required by their functions, so as to determine the overall network load level of the entire railway video surveillance system under a certain bandwidth.

[0059] It should be noted that the analytic hierarchy process is a multi-criteria decision-making method. It constructs a decision judgment matrix, compares the indicators at each level, and calculates the weight of each indicator to obtain the final decision result. The overall network load threshold involved in the above comparison can be set according to the actual situation. For example, the overall network load threshold is set to 80%, etc., which will not be described here.

[0060] In step S3, the application behavior data is the change in overall power consumption. A period of time is selected in the historical time as the analysis time to ensure the normal operation of the railway video surveillance system during the analysis time. The actual power of the railway video surveillance system is calculated using a single-chip integrated circuit. Two time points are randomly selected during the analysis time to calculate the time difference. The overall power consumption is calculated based on the time difference and the actual power. The calculation formula is: overall power consumption = actual power * time difference.

[0061] When the current railway video monitoring system is operating normally, the current time point is recorded and detected using a single-chip integrated circuit, and the overall power consumption is calculated in the same manner when the same time difference as the analysis time has passed.

[0062] The overall power loss change is obtained by subtracting the overall power loss calculated twice. It should be explained that the overall power loss change is obtained by subtracting the overall power loss calculated during the analysis time from the overall power loss calculated by the current railway video monitoring system. By setting N groups of time differences during the analysis time, that is, randomly selecting time points N times during the analysis time to calculate the time difference, the N overall power loss changes obtained by the above method are merged into a loss change data set, and the data in the loss change data set are normalized. The Min-Max normalization formula can be used for normalization: ,in is the data in the loss change dataset, is the normalized result of the corresponding data in the loss change dataset, is the maximum data in the loss change data set, is the minimum data in the loss change data set.

[0063] The overall network load factor is marked as z, the normalized median of N data in the loss change data set is selected as s, and the result is calculated by the logistic regression formula: ,in is the result of logistic regression calculation, e is the natural base, y is the power-off variable, and the power-off variable y can be set as: , the result of logistic regression calculation is used as the power-off coefficient, and the power-off situation of the switch is analyzed by comparing the power-off coefficient with the power-off threshold.

[0064] The power-off threshold can be set by accessing the railway video surveillance system database to obtain the system power, and the system power is used instead of the actual power to calculate the overall power loss change in the same way. The overall power loss change calculated by the system power is used to calculate the power-off coefficient using the logistic regression formula as the power-off threshold. When calculating the power-off threshold by using the system power, it is also necessary to ensure that the time difference used in calculating the overall power loss change is the same.

[0065] When the power-off coefficient is lower than the power-off threshold, the switch is judged to have a slow power-off rate; when the power-off coefficient exceeds the power-off threshold, the switch is judged to have a fast power-off rate and the switch alarm threshold needs to be adjusted.

[0066] When the power failure rate is fast, the alarm threshold needs to be adjusted to be more sensitive. The adjustment coefficient can be obtained by dividing the power failure coefficient by the power failure threshold and marked as and >1, adjust the alarm threshold according to the adjustment coefficient: ,in, is the adjusted alarm threshold, is the alarm threshold before adjustment. For example, if the alarm threshold before adjustment is 30%, and the calculated adjustment coefficient is 1.2, the alarm threshold after adjustment is 30%*1.2=36%. When the power failure rate is fast, increasing the alarm threshold can remind relevant personnel to make corresponding strategies more quickly.

[0067] It should be noted that the single-chip integrated circuit can use voltage and current sensors at the same time. When connected to the system to be detected, it can directly measure the voltage and current parameters, and then use these two parameters to calculate the actual power. The actual power is different from the system power. The system power is the predicted electrical power under normal operation of the system, and the actual power is the actual power detected under normal operation of the system.

[0068] In step S4, the feedback data is the continuous power failure rate. The continuous power failure rate needs to be detected before and after adjusting the alarm threshold. Before adjusting the alarm threshold, two time periods with a time interval of t are selected to calculate the overall power loss and the difference is obtained to obtain the overall power loss change, which is marked as Similarly, after adjusting the alarm threshold, the overall power consumption is calculated in the time interval t and the difference is obtained to obtain the overall power consumption change, which is marked as ,Will and The calculated result obtained by doing the quotient is used as the continuous power-off increase rate.

[0069] Compare the continuous power failure rate with the preset amplification threshold to select the refined alarm threshold or power failure optimization. If the continuous power failure rate exceeds the amplification threshold, it is judged that the power failure of the switch has deepened and power failure optimization is required; if the continuous power failure rate is lower than the amplification threshold, it is judged that the power failure degree of the switch has changed slightly and the alarm threshold is refined.

[0070] The continuous power-off increase rate refers to the power-off rate before and after adjusting the alarm threshold. If the continuous power-off increase rate is greater than 1, it means that the overall system power-off rate is getting faster and faster. It is not appropriate to simply adjust the increase threshold. The switch needs to be updated or repaired. When the continuous power-off threshold is lower than the increase threshold, the alarm threshold is refined. When the continuous power-off increase rate exceeds the increase threshold, power-off optimization is performed. This reduces the risk of problems such as surveillance video interruption or alarm being too late to be processed due to rapid power-off, such as the switch powering off before reaching the alarm threshold.

[0071] It should be noted that power-off optimization can view the current firmware version information in the web management interface provided by the switch, download the latest firmware according to the current firmware version information and upload it to the switch for updating. The alarm threshold can be improved by summing the continuous power-off increase rate and the adjustment coefficient calculated in step S3 and taking the average value as the new adjustment coefficient. For example, the alarm threshold before adjustment is 30%, and the calculated adjustment coefficient is 1.2. The continuous power-off increase rate is 1.4, then the new adjustment coefficient is (1.2+1.4) / 2=1.3, and the adjusted alarm threshold is 30%*1.3=39%. The increase threshold involved in the above steps is set by professionals in this field according to actual conditions. For example, the increase threshold is set to 1.5, etc., which will not be analyzed here.

[0072] Example 2, a switch power failure alarm system applied to a railway video monitoring system, such as Figure 2 As shown, the switch power failure alarm method used to implement the above-mentioned railway video monitoring system includes a data acquisition module, a power failure analysis module, a confirmation processing module and a feedback adjustment module;

[0073] The data acquisition module is used to collect various types of data and transmit them to different subsequent modules for the normal operation of the subsequent modules;

[0074] The power failure analysis module analyzes the collected different data, determines the overall network load level of the railway video surveillance system, and sends the judgment result to the confirmation processing module;

[0075] The confirmation processing module receives the judgment result and performs different alarm threshold adjustment operations according to the corresponding judgment result;

[0076] The feedback adjustment module is used to receive feedback data and select different response measures such as improving the alarm threshold or power-off optimization according to the feedback data.

[0077] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product.

[0078] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application of the technical solution and the invention constraints. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0079] In addition, each functional module in each embodiment of the present application may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module.

[0080] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0081] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A switch power failure alarm method applied to a railway video monitoring system, characterized in that: The following steps are included: Step S1: Collect historical data and divide network load distribution applications; Step S2: collect application data, integrate historical data and application data and use the analytic hierarchy process to determine the overall network load level; Step S3: collect application behavior data, analyze the switch power failure situation by using logistic regression method in combination with the overall network load level, and adjust the alarm threshold according to the switch power failure situation; Step S4: Collect feedback data, and improve the alarm threshold or power-off optimization processing according to the feedback data; In step S2, the application data includes the bandwidth utilization rate of each part of the application and the video frame rate of each part. The load estimation of each part of the application is calculated by using the hierarchical analysis method based on the historical data and the application data. The load weight is set according to the load estimation of each part of the application. The overall network load degree is determined by combining the load estimation of each part of the application and the load weight. In step S3, the application behavior data is the change in overall power consumption. A period of time is selected in the historical time as the analysis time, and the overall power consumption within the analysis time is calculated. Then, the current overall power consumption is detected and calculated within the same time interval, and the overall power consumption change is obtained by subtracting the two. After normalizing the change in overall power consumption, the overall network load factor is combined as the input of the logistic regression model. The calculated result is used as the power-off coefficient. The power-off situation of the switch is analyzed by comparing the power-off coefficient with the preset power-off threshold. If the power-off situation is serious, the alarm threshold is adjusted according to the power-off coefficient. The feedback data is the continuous power failure increase rate. The overall power loss change before and after the alarm threshold is adjusted is calculated respectively, and the quotient of the overall power loss change after the alarm threshold is adjusted and the overall power loss change before the alarm threshold is adjusted is taken as the continuous power failure increase rate. In step S4, the alarm threshold or power-off optimization is determined by comparing the continuous power-off increase rate with the preset increase threshold. If the continuous power-off increase rate exceeds the increase threshold, it is judged that the power-off amplitude of the switch has deepened, and power-off optimization is performed; if the continuous power-off increase rate is lower than the increase threshold, it is judged that the power-off amplitude of the switch has changed slightly, and the alarm threshold is refined by combining the continuous power-off increase rate and the power-off coefficient.

2. The switch power failure alarm method applied to a railway video monitoring system according to claim 1 is characterized in that: In step S1, the historical data is the power consumption of each application in the railway video monitoring system. According to the historical data, multiple applications with large power consumption are selected as analysis samples, and the overall network load level is analyzed using the network load of each application in the analysis sample.

3. The switch power failure alarm method applied to a railway video monitoring system according to claim 1 is characterized in that: In step S2, when determining the overall network load level, the calculated load weights of each application and the bandwidth utilization of each application are weighted and summed to obtain the overall network load coefficient, and the overall network load coefficient is compared with the preset overall network load threshold to determine the overall network load level.

4. A switch power failure alarm system applied to a railway video surveillance system, used to implement the switch power failure alarm method applied to a railway video surveillance system as described in any one of claims 1 to 3, characterized in that: It includes data acquisition module, power-off analysis module, confirmation processing module and feedback adjustment module; The data acquisition module is used to collect various types of data and transmit them to different subsequent modules for the normal operation of the subsequent modules; The power failure analysis module analyzes the collected different data, determines the overall network load level of the railway video surveillance system, and sends the judgment result to the confirmation processing module; The confirmation processing module receives the judgment result and performs different alarm threshold adjustment operations according to the corresponding judgment result; The feedback adjustment module is used to receive feedback data and select an improved alarm threshold or power-off optimization according to the feedback data.

Images