System and method for judging power failure area at low voltage based on mobile optical modem equipment

Abstract

The invention relates to a system and method for low-voltage judgment of a power failure area based on mobile optical modem equipment, and the method comprises the steps: collecting the static data and offline event data of each mobile optical modem in real time, and storing the data after the data cleaning operation; based on historical offline event data of each mobile optical modem, an initial weight is distributed for each optical modem offline event, and then the offline weight of each mobile optical modem is dynamically adjusted according to the spatial and temporal distribution characteristics of the real-time offline event data of each mobile optical modem; in combination with the offline weight of each mobile optical modem, determining the power failure judgment weight of the corresponding area through space-time scanning statistics; and comparing the power failure judgment weight of each area with a corresponding power failure threshold value to judge whether a power failure fault occurs in the area, and if the power failure fault occurs, outputting position information and power failure warning information corresponding to the area. Compared with the prior art, the method can timely and accurately identify the regional power failure event and determine the fault region.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of communication and power monitoring, in particular to a system and method for judging low-voltage power failure area based on mobile optical modem equipment. BACKGROUND

[0002] In the prior art, the fault detection of the power system mainly relies on distributed sensor networks and user reports, but this method has the disadvantages of response delay and incomplete coverage. Low-voltage faults are basically discovered by users and then reported for repair, and since low-voltage fault power failure event information cannot be automatically identified and judged, timely fault repair cannot be performed, which is not conducive to ensuring the power reliability of users. In addition, a power failure identification and analysis scheme is proposed in Chinese patent CN109472476A, which obtains real-time power failure alarm event information of the metering automation system, device information, and related power failure event information of the production system and the distribution system, and then collects them according to single households, meter boxes, branch lines, transformers, feeders and busbars. By comparing with existing power failure events, low-voltage fault events to be eliminated and low-voltage fault events to be confirmed are automatically generated. This scheme still has the problem of data information lag, and cannot accurately obtain power failure area information.

[0003] At present, the smart grid technology design integrates remote monitoring, but is usually limited to high-voltage transmission networks, and lacks an instant feedback mechanism for end low-voltage user power failure events, making it difficult to real-time grasp the power state of the low-voltage end. SUMMARY

[0004] The purpose of the present application is to overcome the defects of the prior art and provide a system and method for judging low-voltage power failure area based on mobile optical modem equipment, which can accurately identify regional power failure events and determine the fault area.

[0005] The purpose of the present application can be achieved by the following technical solution: a system for judging low-voltage power failure area based on mobile optical modem equipment, comprising a device management platform, a research and judgment center and an alarm platform connected in sequence, the device management platform is used for collecting static data and offline event data of each mobile optical modem and transmitting to the research and judgment center;

[0006] The research and judgment center analyzes and processes the data information uploaded by the device management platform, dynamically adjusts the weight of the offline event data, combines the time and space scanning statistical method, judges whether a power failure occurs and determines the power failure area, and outputs the corresponding alarm information to the alarm platform.

[0007] Further, the judgment center is provided with a data cleaning module, a database and a power failure detection module, the data cleaning module is used for abnormal value screening and invalid filtering processing on data information uploaded by the equipment management platform, and the processed data information is stored in the database;

[0008] The power failure detection module obtains historical data information and current real-time drop event data of each mobile optical CATV modem from the database, determines whether a power failure fault occurs currently and determines a fault area by dynamically adjusting a weight and combining a time-space scanning statistical method.

[0009] A method for judging a power failure area based on a mobile optical CATV modem equipment, comprising the following steps:

[0010] S1, collecting static data and drop event data of each mobile optical CATV modem in real time, and storing after data cleaning operation;

[0011] S2, assigning an initial weight to each optical CATV modem drop event based on historical drop event data of each mobile optical CATV modem, and then dynamically adjusting the drop weight of each mobile optical CATV modem according to the time-space distribution characteristics of real-time drop event data of each mobile optical CATV modem;

[0012] S3, combining the drop weight of each mobile optical CATV modem, determining the power failure judgment weight of the corresponding area through time-space scanning statistics;

[0013] S4, comparing the power failure judgment weight of each area with the corresponding power failure threshold to determine whether a power failure fault occurs in the area, and if a power failure fault occurs, outputting the position information and power failure alarm information corresponding to the area.

[0014] Further, the static data of the mobile optical CATV modem in the step S1 includes geographical position and equipment ID information of the optical CATV modem, and the drop event data includes equipment online time and equipment offline time.

[0015] Further, the data cleaning operation in the step S1 includes cleaning the data to remove invalid, duplicate or abnormal data.

[0016] Further, the step S2 comprises the following steps:

[0017] S21, assigning an initial weight to each optical CATV modem drop event based on historical drop event data of each mobile optical CATV modem;

[0018] S22, dynamically adjusting the drop weight of each mobile optical CATV modem according to the geographical position and drop time corresponding to the real-time drop event data of each mobile optical CATV modem, combining historical drop time period, drop time number and network state data.

[0019] Further, the drop weight of the mobile optical CATV modem in the step S22 is specifically:

[0020] W=(0.7x1 / density+0.3x(1-history drop rate))x(1+time factor+chain factor)x network factor

[0021] Wherein, time factor: evening peak = 0.5, early morning = 0.1;

[0022] Chain factor = 0.2xN, N is the number of chain dropouts within a set radius within a set time;

[0023] Network factor: failure = 0.2, normal = 1.0.

[0024] Further, the step S3 is specifically that, within a set space-time window, the sum of the drop weights corresponding to all dropped optical modem devices in the space-time window is calculated as the power failure judgment weight of the space-time window.

[0025] Further, in the step S4, if the power failure judgment weight of the space-time window is greater than or equal to the power failure threshold, it is judged that a power failure occurs, and the coordinate information of the space-time window, including the grid ID and the time interval of the window, is output as the location information corresponding to the power failure area.

[0026] Further, the power failure threshold is specifically:

[0027] Q=kxln(M)x(alpha+beta)

[0028] Wherein, k is a sensitivity calibration coefficient, N is the total number of optical modems in the target grid, alpha is a time correction factor, and beta is a risk area correction factor.

[0029] Compared with the prior art, the present application has the following advantages:

[0030] The present application uses the device management platform to collect the static data and drop event data of each mobile optical modem and transmits them to the research and judgment center; the research and judgment center analyzes and processes the data information uploaded by the device management platform, dynamically adjusts the weight of the drop event data, combines the space-time scanning statistical method, and judges whether a power failure occurs and determines the power failure area. Thus, on the one hand, the optical modem devices of the home broadband users are used as power failure sensing nodes, which can realize real-time monitoring of wide-area coverage, thereby providing reliable data support for subsequent power failure detection; on the other hand, the weight of the drop event data is dynamically adjusted, and the space-time scanning statistics are performed by combining the dynamic weight, which can quickly cluster and accurately judge the power failure event, thereby identifying the power failure and locating the failure area.

[0031] This invention collects static data and disconnection event data from each mobile optical modem, cleans and stores the data, then obtains the initial weight of the optical modem based on historical disconnection event data, and dynamically adjusts the weight by combining the spatiotemporal distribution characteristics of the current real-time disconnection event data of the optical modem. This weight adjustment strategy considers factors such as the geographical location and occurrence time of disconnection events, and is based on strategies such as time period, number of events, or changes in network status. This dynamic weight adjustment method can more accurately reflect the current network status and improve the accuracy and timeliness of power outage judgment in the community.

[0032] This invention combines the dynamic weights of optical modems (ONTs) for spatiotemporal scanning statistics and power outage anomaly detection. Within a spatiotemporal analysis window, the dynamic weight values ​​of all offline ONT devices within the window are statistically analyzed. The statistical values ​​of the window are compared with a power outage threshold to determine whether a power outage fault has occurred within that window. This spatiotemporal scanning statistical method, which combines variable weights, not only improves the accuracy of power outage anomaly detection but also enhances the timeliness of alarm information. The statistical values ​​of the window can quantify the confidence level of a regional power outage; that is, the more high-weight ONT devices there are, the higher the corresponding confidence level. Furthermore, the statistical values ​​of the window automatically decay during network failures, thereby improving the detection's anti-interference capability and ensuring the accuracy of power outage detection. Attached Figure Description

[0033] Figure 1 This is a schematic diagram of the system structure of the present invention;

[0034] Figure 2 This is a schematic diagram of the application architecture of Example 1;

[0035] Figure 3 This is a schematic diagram of the method flow of the present invention;

[0036] The markings in the diagram are as follows: 1. Equipment Management Platform, 2. Analysis Center, 3. Alarm Platform, 201. Data Cleaning Module, 202. Database, 203. Power Outage Detection Module. Detailed Implementation

[0037] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0038] Example 1

[0039] like Figure 1 As shown, a system for determining power outage areas based on mobile optical modem devices includes a device management platform 1, an analysis center 2, and an alarm platform 3 connected in sequence. The device management platform 1 is used to collect static data and disconnection event data of each mobile optical modem and transmit them to the analysis center 2.

[0040] The research and judgment center 2 analyzes and processes the data information uploaded by the equipment management platform 1, judges whether a power failure fault occurs and determines a power failure area by dynamically adjusting the weight of the offline event data and combining a time-space scanning statistical method, and outputs corresponding alarm information to the alarm platform 3.

[0041] The research and judgment center 2 is provided with a data cleaning module 201, a database 202 and a power failure detection module 203, the data cleaning module 201 is used for performing outlier screening and invalid filtering processing on the data information uploaded by the equipment management platform 1, and storing the processed data information into the database 202;

[0042] The power failure detection module 203 obtains the historical data information of each mobile optical CATV modem and the current real-time offline event data from the database 202, determines whether a power failure fault occurs at present and determines a fault area by dynamically adjusting the weight and combining a time-space scanning statistical method.

[0043] The embodiment applies the above scheme, and builds an application architecture as shown in Figure 2 The research and judgment center is established at the mobile operator side, the equipment management platform regularly collects and summarizes the location information (provincial, municipal, district and cell level), quantity and real-time online state of the optical CATV modem equipment, so as to realize data collection and aggregation;

[0044] When the optical CATV modem equipment is powered off due to power line problems, the equipment management platform automatically pushes the power-off event information to the research and judgment center, and records the power-off time and the optical CATV modem equipment ID;

[0045] The research and judgment center uses an algorithm to evaluate whether a preset threshold is reached according to the number and geographical position distribution of the power-off event information received in a short time, determines whether a regional power failure event occurs, and immediately triggers an alarm to the corresponding power supplier, so as to realize power failure judgment and alarm;

[0046] Once the power is restored, the optical CATV modem equipment is online again, the equipment management platform pushes the power restoration information again, the research and judgment center records and confirms that the area has been restored to power supply, realizes a closed loop, and completes power restoration confirmation.

[0047] Embodiment two

[0048] The embodiment is based on the system scheme proposed in embodiment one, and realizes a method for judging a low-voltage power failure area based on a mobile optical CATV modem equipment, as shown in Figure 3 The method comprises the following steps:

[0049] S1, real-time collection of static data (including geographical position and equipment ID information of the optical CATV modem) and offline event data (including equipment online time and equipment offline time) of each mobile optical CATV modem, and storage after data cleaning operation, wherein the data cleaning operation comprises cleaning the data and removing invalid, duplicate or abnormal data;

[0050] S2, assign an initial weight to each mobile optical cat drop event based on the historical drop event data of each mobile optical cat, and then dynamically adjust the drop weight of each mobile optical cat according to the spatio-temporal distribution characteristics of the real-time drop event data of each mobile optical cat;

[0051] Specifically, first, an initial weight is assigned to each optical cat drop event based on the historical drop event data of each mobile optical cat;

[0052] Then, according to the geographical position and drop time corresponding to the real-time drop event data of each mobile optical cat, combined with the historical drop time period, the number of drop times and the network state data, the drop weight of each mobile optical cat is dynamically adjusted:

[0053] W=(0.7x1 / density+0.3x(1-historical drop rate))x(1+time factor+chain factor)x network factor

[0054] Chain factor=0.2xN, N is the number of chain dropouts within a set radius within a set time;

[0055] Network factor: failure=0.2, normal=1.0;

[0056] S3, combine the drop weight of each mobile optical cat, and determine the power failure judgment weight of the corresponding area through spatio-temporal scanning statistics. Specifically, in a set spatio-temporal window, the sum of the drop weights corresponding to all drop optical cat devices in the spatio-temporal window is calculated as the power failure judgment weight of the spatio-temporal window;

[0057] S4, compare the power failure judgment weight of each area with the corresponding power failure threshold to determine whether a power failure fault occurs in the area. If it is determined that a power failure fault occurs, the position information and power failure alarm information corresponding to the area are output;

[0058] If the power failure judgment weight of the spatio-temporal window is greater than or equal to the power failure threshold, it is determined that a power failure fault occurs, and the coordinate information of the spatio-temporal window is output as the position information corresponding to the power failure area. The coordinate information of the spatio-temporal window includes the grid ID and the time interval of the window, and the power failure threshold is specifically:

[0059] Q=kxln(M)x(alpha+beta)

[0060] Where k is a sensitivity calibration coefficient, N is the total number of optical cats in the target grid, alpha is a time correction factor, and beta is a risk area correction factor.

[0061] The embodiment applies the above scheme, and the main process includes:

[0062] I. Collecting data. Collecting offline event data from the home broadband user optical cat equipment of the communication operator, including offline time, geographical location, device ID and other information.

[0063] II. Data preprocessing. Clean the collected data to remove invalid, duplicate or abnormal data, and ensure the accuracy and integrity of the data.

[0064] III. Variable weight setting and dynamic adjustment.

[0065] The present scheme innovatively proposes a dynamic weight adjustment method based on historical offline event data and network status. The method first assigns an initial weight to each optical cat offline event, and then dynamically adjusts the weight according to the spatio-temporal distribution characteristics of real-time offline event data. The weight adjustment strategy takes into account factors such as the geographical location and occurrence time of offline events, and is based on strategies such as time period, event quantity or network status change. This dynamic weight adjustment method can more accurately reflect the current network status, improving the accuracy and timeliness of cell power outage judgment.

[0066] The weight formula is: W = (0.7 x 1 / density + 0.3 x (1-historical offline rate)) x (1+time factor+chain factor) x network factor

[0067] Wherein, time factor: evening peak = 0.5, early morning = 0.1;

[0068] Chain factor = 0.2 x N, (N is the number of chain outages within 10 minutes within a 1km radius);

[0069] Network factor: fault = 0.2, normal = 1.0.

[0070] The final output is a mapping table of device ID and weight value {id:W}, which is used for subsequent regional aggregation analysis.

[0071] Due to the uneven spatio-temporal distribution characteristics of offline event data in the scene, existing technologies cannot well utilize these data for accurate power outage judgment. Therefore, the present scheme innovatively proposes a dynamic weight adjustment algorithm based on historical offline event data and network status. This algorithm can more accurately reflect the current network status by dynamically adjusting the weight, improving the accuracy and timeliness of cell power outage judgment. This dynamic weight adjustment method not only considers factors such as the geographical location and occurrence time of offline events, but also is based on strategies such as time period, event quantity or network status change, so as to more comprehensively utilize offline event data for power outage judgment.

[0072] Four, spatio-temporal scanning statistics combined with abnormal detection. The scheme also innovatively proposes a spatio-temporal scanning algorithm integrating dynamic weight. The method performs the following operations within the set spatio-temporal analysis window (in this embodiment, the spatio-temporal analysis window is designed as a 1km x 1km grid / 10 minutes):

[0073] 1), Calculate the core index:

[0074] Sum of weights (ΣW): The total sum of dynamic weight values of all offline optical CATV devices in the window;

[0075] This embodiment also calculates the offline event density, i.e., the number of offline optical CATV devices per unit area, for auxiliary reference, delay alarm, and block size division;

[0076] 2), Abnormal detection:

[0077] Power outage threshold Q:

[0078] Q=k x ln(M) x (a+β), where k is a sensitivity calibration coefficient (0.8 in this embodiment), M is the total number of optical CATV devices in the target grid, a is a time correction factor (0.8 for late peak, 1.2 for early morning), and β is a risk area correction factor (0.7 for hospital, 1.1 for factory).

[0079] When the total sum of weights ΣW of offline devices in the grid is greater than or equal to Q, a power outage alarm is triggered.

[0080] 3), Fault area positioning:

[0081] Mark all spatio-temporal window coordinates (including grid ID + time interval) where ΣW≥Q, and output to the power system for fault area prompt.

[0082] Considering that existing spatio-temporal scanning algorithms only rely on device quantity or density (such as "50 offline devices in 10 minutes"), without distinguishing the actual importance of different device outages, leading to missed reports in sparse areas and false alarms in dense areas. This scheme introduces the sum of weights (ΣW) as the core criterion, which on the one hand quantifies the confidence of regional power outage (the more high-weight optical CATV devices, the higher the confidence); on the other hand, it improves the detection anti-interference: when the network fails, the weight automatically decays (ΣW drops sharply), thereby avoiding false triggering; in addition, in this scheme, the power outage threshold Q can be dynamically adjusted (for example, Q=3.0 around hospitals, Q=8.0 in industrial areas), thereby significantly improving detection accuracy (rural / urban scene adaptation) and timeliness (no need for manual review of sporadic events).

[0083] In summary, the present scheme ingeniously utilizes the optical modem equipment of the home broadband user as a power outage sensing node. These devices are distributed throughout the city and can achieve wide-area coverage and real-time monitoring of power outage events. By collecting offline event data from optical modem equipment, power outage information can be obtained in a timely manner, providing reliable data support for subsequent judgment and alarm. Using optical modem equipment as a sensing node, offline event data is collected to achieve real-time monitoring of power outage events. This technology has the advantages of wide coverage, high data accuracy, and strong real-time performance, and can provide reliable power outage information for power suppliers.

[0084] After obtaining the offline event data, the present scheme dynamically adjusts the weight, combines temporal and spatial scanning statistics and anomaly detection to quickly cluster and accurately judge the power-off event, effectively identify the fault area, and determine whether it is a power outage event. Not only does it improve the accuracy of the judgment, but it also greatly shortens the response time, providing timely and reliable alarm information for power suppliers.

[0085] The application of the present scheme in practice will significantly shorten the fault discovery and response time, improve the emergency response speed, and based on the data analysis of the dynamic weight adjustment model of geographical distribution, the positioning of the power outage area is more accurate. In addition, it can also reduce manual intervention and improve processing efficiency and accuracy. The present scheme can significantly improve the operation and maintenance efficiency of the power system and user experience, reduce economic losses caused by power outages, and has broad market application prospects.

Claims

1. A system for determining power outage areas based on a mobile optical modem, characterized in that, The device management platform (1), the research and judgment center (2) and the alarm platform (3) are sequentially connected, the device management platform (1) is used for collecting static data and offline event data of each mobile optical modem, and transmitting to the research and judgment center (2); The research and judgment center (2) analyzes and processes the data information uploaded by the device management platform (1), dynamically adjusts the weight of the offline event data, combines the space-time scanning statistical method, judges whether a power failure fault occurs and determines the power failure area, and outputs corresponding alarm information to the alarm platform (3). 2.The system for determining a power outage area based on a mobile optical modem device according to claim 1, wherein, The research and judgment center (2) is provided with a data cleaning module (201), a database (202) and a power failure detection module (203), the data cleaning module (201) is used for performing outlier screening and invalid filtering processing on the data information uploaded by the device management platform (1), and storing the processed data information into the database (202); The power failure detection module (203) obtains historical data information and current real-time offline event data of each mobile optical modem from the database (202), dynamically adjusts the weight, combines the space-time scanning statistical method, determines whether a power failure fault occurs at present and determines the fault area.

3. A method for judging a power failure area based on a mobile optical cat device, characterized by, The method comprises the following steps: S1, real-time collection of static data and offline event data of each mobile optical modem, and storage after data cleaning operation; S2, based on the historical offline event data of each mobile optical modem, an initial weight is allocated to each optical modem offline event, and then the offline weight of each mobile optical modem is dynamically adjusted according to the space-time distribution characteristics of the real-time offline event data of each mobile optical modem; S3, combining the offline weight of each mobile optical modem, the power failure judgment weight of the corresponding area is determined through space-time scanning statistics; S4, comparing the power failure judgment weight of each area with the corresponding power failure threshold to judge whether a power failure fault occurs in the area, if a power failure fault occurs, outputting the position information and power failure alarm information corresponding to the area.

4. The method for determining a power outage area based on a mobile optical modem device according to claim 3, wherein, The static data of the mobile optical modem in the step S1 includes the geographical position and device ID information of the optical modem, and the offline event data includes the device online time and device offline time.

5. The method for determining a power outage area based on a mobile optical modem device according to claim 3, wherein, The data cleaning operation in the step S1 includes cleaning the data to remove invalid, duplicate or abnormal data.

6. The method for determining a power outage area based on a mobile optical cat device according to claim 4, wherein, The step S2 comprises the following steps: S21, based on the historical offline event data of each mobile optical modem, an initial weight is allocated to each optical modem offline event; S22, according to the geographical position and offline time corresponding to the real-time offline event data of each mobile optical modem, combining the historical offline time period, the number of offline times and the network state data, the offline weight of each mobile optical modem is dynamically adjusted.

7. The method for determining a power outage area based on a mobile optical cat device according to claim 6, wherein, The offline weight of the mobile optical modem in the step S22 is specifically: W=(0.7×1 / density+0.3×(1-historical offline rate))×(1+time factor+chain factor)×network factor wherein, time factor: evening peak=0.5, early morning=0.1; Chain factor=0.2×N, N is the number of chain offline within a set time within a set radius; Network factor: fault=0.2, normal=1.

0. 8.The method of claim 6, wherein the method further comprises: determining whether the mobile optical modem device is located in the low-voltage area based on the low-voltage signal. The step S3 is specifically calculating the sum of the offline weights corresponding to all offline optical modem devices in the space-time window as the power failure judgment weight of the space-time window.

9. The method for determining a power outage area based on a mobile optical cat device according to claim 8, wherein, In the step S4, if the power failure judgment weight of the space-time window is greater than or equal to the power failure threshold, it is judged that a power failure fault occurs, and the coordinate information of the space-time window is output as the position information corresponding to the power failure area, wherein the coordinate information of the space-time window includes the grid ID and the time interval of the window.

10. The method for determining a power outage area based on a mobile optical cat device according to claim 9, wherein, The power failure threshold is specifically: Q=k*ln(M)*(α+β) wherein k is a sensitivity calibration coefficient, N is the total number of optical modems in the target grid, α is a time correction factor, and β is a risk area correction factor.

Images