Online distribution and push system based on emergency scene

Abstract

The invention discloses an online distribution and push system based on an emergency scene, which comprises a data acquisition layer, a data processing layer and an information distribution and push layer which are connected in sequence, and is characterized in that a scene recognition module is introduced, and system parameters are automatically adjusted to adapt to specific scene requirements by analyzing feature data of a current emergency scene; the method can identify the scene information, can perform targeted processing on the identified scene information, realizes accurate matching of distribution objects according to a content distribution network established by a hierarchical chain, a professional chain and a role chain, provides support for real-time study and judgment, disposal, real-time monitoring and quick response of disaster situations, and can perform real-time monitoring and quick response after receiving intelligently pushed key attention information. The command center carries out emergency command decision making with each specialty, through hierarchical, specialty and role accurate pushing, command instruction rapid issuing, rapid response, execution progress real-time monitoring, accurate feedback, timely processing and key information pushing are realized, and the emergency processing efficiency is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power emergency scenario distribution, and in particular to an online distribution and push system based on emergency scenarios. Background Art

[0002] The power industry faces a wide variety of emergency scenarios, each requiring distinct response measures and priorities. For example, in earthquake rescue operations, distress signals must be prioritized and quickly distributed to the nearest rescue team. In chemical plant explosions, on the other hand, monitoring for hazardous material leaks and promptly notifying nearby residents to evacuate are crucial. In other words, each emergency scenario requires tailored response measures and targeted distribution of emergency information to the appropriate recipients for targeted action. Therefore, developing intelligent emergency scenario content distribution strategies based on disaster monitoring and early warning data and the fundamentals of safe production is crucial, as is developing a system that precisely pushes key information.

[0003] However, due to the large number of elements in emergency scenarios, simple scenario information summary cannot highlight the information that needs to be focused on at different levels, professions, and roles under different disaster changes and time sections. Emergency repair operation command decisions cannot be carried out quickly. In traditional emergency scenario distribution systems, emergency scenario information cannot be processed in a targeted manner. When a disaster occurs, it is impossible to effectively identify emergency scenario information and distribute it to the matching people. The emergency information is not strongly correlated with the distribution object. Therefore, there is an urgent need for an emergency scenario distribution system that can identify specific emergency scenarios and distribute them to matching objects based on the current emergency scenario information to minimize the extent of damage. Summary of the Invention

[0004] The present invention provides an online distribution and push system based on emergency scenarios, which is used to identify specific emergency scenarios and distribute them to matching objects according to current emergency scenario information to minimize the degree of damage.

[0005] In view of this, the present invention provides an online distribution and push system based on emergency scenarios, the system comprising: a data acquisition layer, a data processing layer and an information distribution and push layer connected in sequence;

[0006] The data collection layer is used to collect environmental data of the current emergency scenario and data that needs to be fed back by various departments through the sensor network, and transmit the data to the data processing layer;

[0007] The data processing layer receives data transmitted by the data acquisition layer, analyzes and processes the data using machine learning algorithms, identifies the current emergency scenario type, and combines the power system's own operating data to determine the specific characteristics of the emergency scenario. Based on the identification results, the system parameters are automatically adjusted to obtain system parameters that match the current emergency scenario.

[0008] The information distribution and push layer is used to classify the collected and processed data according to the needs of different types of emergency scenarios, and use different distribution and push algorithms to push the current emergency scenario information for different types of information.

[0009] Optionally, the data processing layer includes a scene recognition module and a parameter adjustment module, wherein:

[0010] The scenario recognition module is used to receive environmental data from the data acquisition layer, analyze and process the data using a decision tree algorithm, identify the current emergency scenario type, and at the same time, determine the specific characteristics of the emergency scenario based on the power system's own operating data;

[0011] The parameter adjustment module is used to automatically adjust system parameters according to the recognition results of the scene recognition module to obtain system parameters that match the current emergency scene.

[0012] Optionally, the execution process of the scene recognition module includes:

[0013] Label the collected environmental data of emergency scenarios, collect various operating data of the power system, as well as the operating status and fault records of the equipment, and pre-process the collected power system operating data;

[0014] Build a decision tree and train it using a labeled dataset;

[0015] Input the collected environmental data of the current emergency scene into the scene recognition module to identify the type of emergency scene currently in place;

[0016] Correlate the identified emergency scenario types with power system operation data;

[0017] Establish a mapping relationship between emergency scenarios and changes in power system operating parameters;

[0018] Determine the specific location of the power fault based on fault records and real-time monitoring data in the power system operation data. At the same time, analyze the topology and power flow distribution of the power network to assess the scope of the fault's impact on the power system.

[0019] Assess the severity of power failures in emergency scenarios by considering the degree of deviation from power system operating parameters, equipment damage, and the impact on power supply;

[0020] Combine the real-time operating status of the power system and the development trend of emergency scenarios to predict potential risks that may arise.

[0021] Optionally, the specific process of identifying the current emergency scene type in the scene recognition module is:

[0022] Collect emergency scene environmental data measured by various sensors;

[0023] Extract key features from sensor data;

[0024] Label the collected data and mark the emergency scenario type corresponding to each data sample;

[0025] Use the labeled data set to train the constructed decision tree;

[0026] Prune the trained decision tree;

[0027] The collected data of the current emergency scene is input into the scene recognition module to identify the current emergency scene type.

[0028] Optionally, the process of the scene recognition module identifying the current emergency scene type includes:

[0029] At each node of the decision tree, an optimal attribute is selected from all features as the partitioning attribute based on the information gain. According to the selected partitioning attribute, the data set is divided into different subsets. For each divided subset, the above steps of selecting the partitioning attribute and partitioning the data set are repeated, and the subtrees of the decision tree are recursively constructed until a certain stopping condition is met.

[0030] Optionally, in the parameter adjustment module, the execution process is as follows:

[0031] Analyze the needs of different emergency scenarios, formulate system parameter adjustment strategies based on the needs of different scenarios, organize the correspondence between emergency scenarios and parameter adjustment strategies into a parameter adjustment rule library and store it. The system parameter adjustment strategy includes data processing priority parameters, distribution algorithm parameters, push frequency and range parameters;

[0032] Analyze the recognition results and extract key information;

[0033] According to the emergency scenario identification results after analysis, the matching rules are searched in the parameter adjustment rule library, and the system parameters are adjusted according to the matched rules.

[0034] Optionally, the information distribution and push layer includes an information classification submodule and a distribution and push algorithm module, wherein:

[0035] The information classification submodule is used to classify the collected and processed data according to the needs of different emergency scenarios;

[0036] The distribution and push algorithm module is used to push different types of information using different distribution and push algorithms.

[0037] Optionally, in the information classification submodule, information is divided into power facility information, emergency scenario feature information, personnel information, surrounding environment information and notification and warning information.

[0038] Optionally, the classification process in the information classification submodule includes:

[0039] Classification rules are formulated based on the characteristics and definitions of different categories of information. The information classification submodule reads the collected and processed data and matches them according to the pre-established rules;

[0040] Collect historical data with labeled categories as training sets and train machine learning algorithms;

[0041] Associating emergency scenario types with scenario feature information of corresponding categories;

[0042] The information classification submodule obtains the output results of the scene recognition module and classifies the corresponding data into specific categories according to the current emergency scene type and the pre-set association relationship.

[0043] Optionally, the information distribution push layer also includes a multi-channel push module for pushing information through multiple channels.

[0044] It can be seen from the above technical solutions that the present invention has the following advantages:

[0045] The present invention provides an online distribution and push system based on emergency scenarios, which introduces a scenario recognition module. By analyzing the characteristic data of the current emergency scenario, the system parameters are automatically adjusted to adapt to the needs of specific scenarios. The identified scenario information can be processed in a targeted manner. According to the content distribution network established by the hierarchical chain, professional chain, and role chain, accurate matching of distribution objects is achieved, providing support for real-time assessment and judgment of the disaster situation, real-time monitoring of the disposal, and rapid response. After receiving the key information pushed by the intelligent push, the command center and each professional make emergency command decisions, and push the information accurately by hierarchical, professional, and role, so as to achieve rapid issuance of command instructions, rapid response, real-time monitoring of execution progress, accurate feedback, timely processing and push of key information, and greatly improve the efficiency of emergency handling. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0047] Figure 1 A system architecture diagram of an online distribution and push system based on emergency scenarios provided by an embodiment of the present invention;

[0048] Figure 2 A flowchart of a scene recognition module provided by an embodiment of the present invention;

[0049] Figure 3 A flowchart for identifying the current emergency scenario type provided by an embodiment of the present invention;

[0050] Figure 4 A flow chart of a parameter adjustment module provided in an embodiment of the present invention;

[0051] Figure 5 This is a flowchart of classification in the information classification submodule provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, features, and advantages of the present invention more obvious and easy to understand, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described below 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 making creative work are within the scope of protection of the present invention.

[0053] See also Figure 1 , an online distribution and push system based on emergency scenarios provided in an embodiment of the present invention includes a data acquisition layer, a data processing layer and an information distribution and push layer connected in sequence;

[0054] The data collection layer is used to collect environmental data of the current emergency scenario through the sensor network, as well as data that needs to be fed back by various departments (for example, data fed back by disaster surveyors based on the damage situation on site), and transmit the data to the data processing layer;

[0055] The data processing layer receives data transmitted by the data acquisition layer, analyzes and processes the data using machine learning algorithms, identifies the current emergency scenario type, and combines the power system's own operating data to determine the specific characteristics of the emergency scenario. Based on the identification results, the system parameters are automatically adjusted to obtain system parameters that match the current emergency scenario.

[0056] The information distribution and push layer is used to classify the collected and processed data according to the needs of different types of emergency scenarios, and adopt different distribution and push algorithms for different types of information to push the current emergency scenario information.

[0057] In one embodiment, the data processing layer includes a scene recognition module and a parameter adjustment module;

[0058] The scene recognition module is used to receive environmental data from the data acquisition layer, analyze and process the data using a decision tree algorithm, identify the current emergency scene type, and at the same time, combine the power system's own operating data to determine the specific characteristics of the emergency scene;

[0059] The parameter adjustment module is used to automatically adjust the system parameters according to the recognition results of the scene recognition module to obtain system parameters that match the current emergency scene.

[0060] Specifically, such as Figure 2 As shown in Figure 2, the execution process of the scene recognition module is as follows:

[0061] First, the collected environmental data of emergency scenarios are labeled. At the same time, various operating data of the power system (including but not limited to real-time data such as the voltage, current, power, and frequency of the power grid), as well as the operating status of the equipment (such as the opening and closing status of the switch, the oil temperature of the transformer, etc.), and fault records (such as the time of fault occurrence, fault type, etc.) are collected. The collected power system operating data is preprocessed to remove outliers, fill in missing values, and organize the data according to a certain format and time series for subsequent analysis.

[0062] Next, a decision tree is constructed and trained using the labeled dataset.

[0063] Then, the collected environmental data of the current emergency scene is input into the scene recognition module to identify the current emergency scene type.

[0064] The identified emergency scenario type is then correlated with power system operating data. For example, if the identified scenario type is a typhoon, the system searches for operating data for power equipment near the typhoon area and analyzes the impact of the typhoon on the power system, such as whether any lines were outages or which equipment was damaged.

[0065] Then, a mapping relationship is established between emergency scenarios and changes in power system operating parameters. It should be noted that through historical data and empirical analysis, typical patterns of changes in power system parameters under different emergency scenarios are summarized. For example, earthquake scenarios may cause transient fluctuations in grid voltage and sudden changes in current on some lines. Establishing a mapping relationship between these parameter changes and earthquake scenarios allows for further confirmation of scenario characteristics through parameter changes in new emergency scenarios.

[0066] Next, based on fault records and real-time monitoring data from the power system's operating data, the specific location of the power fault is determined. Simultaneously, by analyzing the power network's topology and power flow distribution, the scope of the fault's impact on the power system is assessed, specifically including the area affected by the power outage and the number of affected users. For example, in a fire scenario, by analyzing abnormal changes in current and voltage, as well as equipment status information, the point of the power line short circuit caused by the fire can be determined, and the scope of the area affected by the power outage can be calculated. The topology of the power network refers to the connection relationship between the various components of the power system, such as a circuit diagram. Power flow distribution refers to the flow of power (active power and reactive power) on each line in the system under given operating conditions, as well as the distribution of voltage amplitude and phase angle at each node. Simply put, it describes the flow path of electrical energy in the power network and the voltage status of each point.

[0067] The severity of the power failure in an emergency scenario is then assessed by comprehensively considering factors such as the degree of deviation in power system operating parameters, equipment damage, and the impact on power supply. For example, the severity of the failure is classified into minor, moderate, and severe levels based on the magnitude of the voltage drop, the magnitude of the frequency deviation, and the amount and significance of the equipment damage.

[0068] Finally, by combining the real-time operating status of the power system with the development trends of emergency scenarios, potential risks can be predicted. For example, during a rainstorm, if the water level in a certain area is monitored to be rising continuously and close to power equipment, the risk of damage from moisture can be predicted based on the waterproof performance of the equipment and historical experience, as well as the potential for cascading failures such as short circuits and extended power outages.

[0069] Furthermore, in one embodiment, Figure 3 As shown in FIG, the specific process of identifying the current emergency scene type in the scene recognition module is as follows:

[0070] First, collect emergency scenario environmental data measured by various sensors, such as meteorological data such as temperature, humidity, wind speed, rainfall, and equipment status data such as transformer oil temperature, line current, voltage fluctuation, etc.

[0071] Next, key features are extracted from the sensor data, such as geographic location coordinates, rate of change of specific gas concentrations, abnormal temperature fluctuations, etc.

[0072] Next, the collected data is labeled to mark the emergency scenario type corresponding to each data sample.

[0073] The constructed decision tree is then trained using the labeled dataset. Specifically, by continuously adjusting the decision tree's structure and parameters, the decision tree is able to classify the samples in the training dataset as accurately as possible, that is, correctly identify the emergency scenario type corresponding to each sample.

[0074] Next, to prevent overfitting of the decision tree, the trained decision tree is pruned. It should be noted that pruning is divided into pre-pruning and post-pruning. Pre-pruning stops the node splitting during the decision tree construction process when the splitting of a node does not improve classification performance. Post-pruning, after the decision tree is built, starts with the leaf nodes and recursively examines non-leaf nodes. If converting the node to a leaf node improves the generalization performance of the decision tree, pruning is performed.

[0075] Finally, the collected data about the current emergency scenario is input into the scenario recognition module. It should be noted that data flows along the trained decision tree, starting from the root node. The data is judged based on the current node's partitioning attributes, determining which child node the data flows to until it reaches a leaf node. The category marked on the leaf node is the identified type of emergency scenario.

[0076] The scene recognition module identifies the current emergency scenario type by selecting the optimal attribute from all features at each node in the decision tree based on criteria such as information gain, gain ratio, and Gini index. For example, in power emergency scenario identification, if the feature "abnormally increased wind speed" has a high information gain for distinguishing typhoon scenarios from other scenarios, this feature is selected as the partitioning attribute for the current node. Based on the selected partitioning attribute, the dataset is divided into subsets. For example, for "abnormally increased wind speed," the dataset can be divided into two subsets: abnormally increased wind speed and normal wind speed. For each subset, the above steps of selecting partitioning attributes and partitioning the dataset are repeated, recursively constructing subtrees of the decision tree until a stopping condition is met, such as when all samples in the subset belong to the same category (i.e., purity reaches 100%), or when no more attributes can be found to effectively partition the dataset.

[0077] In one embodiment, Figure 4 As shown, in the parameter adjustment module, the execution process is as follows:

[0078] First, the system analyzes the requirements of different emergency scenarios and develops parameter adjustment strategies based on these requirements. The corresponding relationships between emergency scenarios and parameter adjustment strategies are organized into a parameter adjustment rule library and stored. The system's parameter adjustment strategies include parameters for data processing priority, distribution algorithm parameters, and push notification frequency and range. For example, in a fire scenario, fire-related sensor data (such as abnormally elevated temperature and smoke concentration) is prioritized for analysis and processing. For distress signals in earthquake scenarios, the distribution algorithm parameters are adjusted to prioritize distance, quickly distributing distress signals to the nearest rescue team. In a chemical plant explosion scenario, to ensure that nearby residents receive timely information about hazardous material leaks, the frequency of push notifications to residents within a certain radius is increased, while also expanding the reach of the notifications.

[0079] Next, the recognition results are analyzed to extract key information, such as the type of emergency scenario (such as earthquake, fire, chemical plant explosion, typhoon, etc.), as well as possible scene feature information (such as the scale of the fire, the magnitude of the earthquake, the degree of typhoon damage, etc.).

[0080] Finally, based on the emergency scenario identification results after analysis, the matching rules are searched in the parameter adjustment rule library, and the system parameters are adjusted according to the matched rules. It should be noted that after the system parameters are adjusted, the operating status and related indicators of the system are monitored in real time. For example, the speed and accuracy of data processing, the timeliness and success rate of information distribution, the feedback received on push information, etc. are monitored to evaluate whether the parameter adjustment has achieved the expected effect. If the monitoring finds that the effect after the parameter adjustment is not ideal, the system will pass this feedback information to the parameter adjustment module. The parameter adjustment module will optimize the parameter adjustment rules based on the feedback information, combined with the scene characteristics and the actual system operation situation, such as further increasing the weight of distance in the distribution algorithm, and then perform the parameter adjustment operation again to form a closed-loop automatic optimization mechanism to ensure that the system can maintain the best operating state in different emergency scenarios.

[0081] In one embodiment, the information distribution and push layer includes an information classification submodule and a distribution and push algorithm module, wherein:

[0082] The information classification submodule is used to classify the collected and processed data according to the needs of different emergency scenarios;

[0083] The distribution push algorithm module is used to push different types of information using different distribution push algorithms.

[0084] It should be noted that when classifying information, it is necessary to consider the division of responsibilities among personnel at different levels and with different professions (in the case of rescue teams, the number, skill level, and material reserves of the rescue team must be considered so that the system can match the rescue team with appropriate capabilities based on the emergency situation at the scene). This allows the system to push relevant content based on the content distribution network established by the hierarchical chain, professional chain, and role chain, achieving precise matching of distribution targets. After receiving the key information pushed intelligently by the system, the command center and various professionals make emergency command decisions. Through precise push based on hierarchical, professional, and role, command instructions can be quickly issued, quickly responded to, and the execution progress can be monitored in real time with accurate feedback. After determining the distribution target based on the classified content, an appropriate distribution algorithm is used to push the content.

[0085] Furthermore, in one embodiment, in the information classification submodule, information is mainly divided into several categories: power facility information, emergency scenario feature information, personnel information, surrounding environment information, and notification and warning information.

[0086] Specifically, the first category is information related to power facilities, including fault information, such as the type of power system fault, fault location, and fault time, as well as operating status information, such as the real-time operating parameters of the power system, voltage, current, power and other data. For example, this type of information can be sent to relevant units or personnel with power knowledge reserves and who can carry out emergency repairs on power equipment.

[0087] The second category is emergency scenario characteristic information, including disaster type information and disaster severity information, which is distributed to corresponding personnel for processing according to the disaster situation at the emergency site and their responsibilities.

[0088] The third category is personnel-related information, including distress signal information from disaster victims, such as distress location, distress content, distress time, and other data. For example, this type of information can be distributed to rescue teams with rescue capabilities.

[0089] The fourth category is surrounding environment information, including geographic information, such as the geographical location, topography, traffic conditions, and other data of power facilities and surrounding areas, as well as meteorological information, such as real-time temperature, humidity, wind speed, wind direction, and other data. For example, this type of information can be distributed to rescue teams or repair personnel.

[0090] The fifth category is notification and warning information, including internal notification information, which is mainly notifications to internal personnel of the power company, such as fault notifications, emergency repair task assignments, emergency command and dispatch instructions, etc., as well as external notification information, which is warning information issued to surrounding residents or relevant units, such as hazardous material leakage warnings, power outage notifications, safety tips, etc.

[0091] Furthermore, in one embodiment, Figure 5As shown in Figure 2, the classification process in the information classification submodule is as follows:

[0092] First, classification rules are formulated based on the characteristics and definitions of different categories of information. The information classification submodule reads the collected and processed data and matches them according to the pre-established rules.

[0093] Next, a large amount of historical data with labeled categories is collected as a training set, and training is performed based on machine learning algorithms (such as support vector machines (SVMs), naive Bayesian algorithms, etc.). The trained model is deployed in the information classification submodule. When new collected and processed data is input, the model classifies and predicts the data based on the learned feature patterns.

[0094] Then, the emergency scene type is associated with the scene feature information of the corresponding category; for example, smoke concentration data related to fire should be classified as fire scene information.

[0095] Next, the information classification submodule obtains the output results of the scene recognition module and classifies the corresponding data into specific categories according to the current emergency scene type and the pre-set association relationship.

[0096] The information distribution and push layer also includes a multi-channel push module, which distributes information via various channels, including SMS platforms, mobile applications (APPs), and broadcast systems. In emergencies, the broadcast system is prioritized for widespread notification, while detailed information is simultaneously delivered via SMS and the app, ensuring timely and accurate information delivery to relevant personnel.

[0097] In some embodiments:

[0098] In the data collection layer, the sensor network includes geographic information sensors, disaster type detection sensors, power system status sensors and environmental meteorological sensors.

[0099] Specifically, geographic information sensors are used to obtain the location information of the current emergency scene, as well as the surrounding geographical environment and route information. When an emergency occurs, the system can respond quickly to help dispatchers choose the best route. High-resolution optical camera sensors or lidar sensors can be used for positioning.

[0100] Disaster detection sensors monitor environmental parameters. For example, seismic sensors monitor ground vibrations caused by earthquake activity and determine the severity of an earthquake based on parameters such as intensity, frequency, and duration. Smoke sensors detect signs of fire by measuring smoke concentration in the air. Temperature sensors detect changes in ambient temperature; abnormally high temperatures may indicate a potential fire hazard or overheating of electrical equipment. Gas sensors detect the concentration of harmful or flammable gases. If abnormal gas concentrations are detected, the system can promptly notify nearby residents to evacuate and alert repair personnel to take protective measures.

[0101] Power system status sensors are used to monitor current, voltage, and frequency in the power system. For example, current sensors measure the current in power lines to determine whether the system is experiencing overloads, circuit breaker failures, or other faults. Voltage sensors monitor the voltage of power lines to ensure the normal operation of power equipment. Frequency sensors monitor the stability of the power system's frequency.

[0102] Environmental meteorological sensors are used to measure environmental meteorological data. For example, wind speed sensors are used to measure wind speed and direction, rain sensors are used to measure rainfall, and humidity sensors are used to measure ambient air humidity, which helps to assess the safety of power facilities in severe weather conditions.

[0103] The data collection layer includes a communication module, which uses satellite phones, cluster communications, 5G communication technology, and Mesh self-organizing networks to maintain communication with the outside world, obtain remote data, and push information to the outside world.

[0104] Traditional emergency scene distribution systems communicate through public mobile communication networks, shortwave communications, and emergency broadcasts. Since public mobile communication networks rely on base stations for data transmission, the communication quality of shortwave communications is greatly affected by ionospheric changes and the signal is unstable. In emergency broadcast communication methods, the public's initiative to receive information is relatively weak. Therefore, it is necessary to adjust the communication means of emergency scenes and improve the quality of communication to ensure the reliability and effectiveness of communication.

[0105] In the present invention, a satellite phone can be used to communicate via a satellite link without being restricted by the geographical environment. Alternatively, cluster communication, a professional wireless communication system, can be used, which has functions such as group calling, quick calling, and priority scheduling, and can achieve one call and a hundred responses, facilitating rapid communication and scheduling in emergency command. Alternatively, 5G communication technology can be used, which has the characteristics of high speed, low latency, and large capacity, and can support high-definition video monitoring and real-time transmission, so that the emergency command center can clearly see the on-site situation and facilitate remote command and decision-making; it can also assist in the application of intelligent devices such as drones and robots in emergency rescue, and achieve precise control and real-time data transmission. Alternatively, a Mesh self-organizing network is a multi-hop wireless network in which nodes can automatically form a network and automatically route without relying on fixed infrastructure. It has strong flexibility, scalability, and anti-destruction capabilities, and can quickly build a communication network in complex terrain and areas without network coverage.

[0106] Assuming an emergency scenario in the power industry, take "distress signal information" as an example to illustrate, as follows:

[0107] In power industry emergency scenarios, distress signals must be sent to rescue teams to ensure personnel safety. Distance is the primary consideration, so distance is prioritized. Distance is used to leverage the geographic location information contained in distress signals, along with the real-time location data of repair teams and rescue agencies. The team's expertise, equipment, and other information are recorded to ensure a consistent response between the need for assistance and the capabilities of the rescue team. For example, distress signals involving high-voltage equipment are prioritized for delivery to teams with high-voltage operating qualifications and specialized equipment.

[0108] Based on the location coordinates in the distress signal and the location coordinates of all deployable rescue forces, use distance calculation formulas, such as the Euclidean distance formula or the actual distance algorithm based on map projection, to calculate the distance between the distress location and each rescue force.

[0109] Based on the specific needs in the distress signal, such as whether it involves electrical equipment rescue, whether medical assistance is needed, etc., the collection of rescue teams with corresponding capabilities is screened out.

[0110] Prioritize the selected rescue teams based on distance and rescue capability. Teams with closer distances and higher rescue capability match have higher priority. You can set distance weights and capability match weights to calculate a comprehensive score for ranking. The comprehensive score is calculated using the following formula: s=d*w1+a*w2, where s represents the comprehensive score, d represents the distance score, w1 represents the distance weight, a represents the capability match score, and w2 represents the capability match weight. The distance score is normalized based on distance, and the capability match score is based on the matching degree of skills and equipment. For example, you can set the distance weight w1 = 0.6 and the capability match weight w2 = 0.4 to calculate the comprehensive score.

[0111] Distress signals are pushed to the top rescue teams in order of priority. This can be done via text message or a dedicated emergency communication app, ensuring that rescuers receive the distress message, along with detailed location and needs, in a timely manner.

[0112] The present invention can form a real-time distribution strategy for emergency measures for wind and flood prevention by level, region, profession and time based on weather disaster warning data, combined with power grid, equipment, operation risk warning and preventive measures data; and form a real-time distribution strategy for emergency repair measures by level, region, profession and role based on real-time monitoring and abnormal alarm data of power grid, equipment and user power outages, and according to changes in the disaster situation and the progress of repair tasks.

[0113] The present invention can realize the intelligent distribution of content related to different types of emergency scenarios, mainly targeting the distribution of content related to typhoon emergency scenarios. The following describes the distribution of relevant information for typhoon emergency scenarios using a content distribution network based on hierarchical chains, professional chains, and role chains.

[0114] 1. Hierarchical chain:

[0115] Senior decision-makers: responsible for making decisions, such as senior leaders of power companies and heads of government energy management departments. The information that can be distributed includes: typhoon overall situation analysis report, including typhoon path forecast, intensity change trend, expected landfall time and location, etc.; overall disaster assessment of the power system, covering the number and distribution areas of substations and transmission lines that may be affected, as well as an assessment of the potential impact on power supply; preliminary suggestions for emergency resource reserves and deployment plans, such as the types and approximate quantities of required power generation vehicles and emergency repair materials.

[0116] Middle-level management personnel: Responsible for resource allocation, task supervision and coordination based on information, such as department managers and regional operations managers of power companies. The information that can be distributed is real-time reports on the damage to power facilities in specific areas, with detailed lists of the names and locations of confirmed damaged substations, towers, and lines; task allocation and progress tracking for each repair team, clarifying the repair area each team is responsible for, task content (such as repairing a certain line, replacing a certain tower), and current completion progress; updating material demand and supply status, informing the actual consumption and inventory status of various types of repair materials, as well as the estimated time and quantity of subsequent material deployment.

[0117] Grassroots executive personnel, such as front-line power repair workers and on-site operation supervisors, can distribute information such as detailed on-site repair work instructions, which provide detailed repair steps, safety precautions, and technical requirements for specific damaged facilities; real-time weather warning information, especially wind and rainfall changes closely related to the current repair site; surrounding traffic conditions and road access information to help repair personnel plan the best travel routes and avoid delays in repair progress due to road waterlogging, landslides, etc.

[0118] 2. Professional chain:

[0119] Power operation and maintenance professionals, such as power line repairers, substation operation and maintenance personnel, and other professional and technical personnel, can distribute the following information: detailed technical parameters and historical fault records of power equipment in the typhoon-affected area, so as to quickly determine the cause of the fault and formulate targeted repair plans; the latest technical guidelines and operating specifications for power equipment emergency repairs to ensure that the repair work meets the standard requirements; case analysis and handling experience summary of power facility failures in similar typhoon disasters to provide reference for current emergency repair work.

[0120] Meteorological professionals: Provide customized meteorological services for the power system and assist in formulating response strategies. For example, the power company's internal meteorological monitoring team and meteorological professionals working with the power company can distribute micro-meteorological monitoring data around power facilities, such as real-time wind speed, wind direction, and humidity near towers, to help them more accurately assess the impact of meteorological conditions on power facilities; the power system's special needs and sensitive parameters for meteorological conditions, such as the wind speed tolerance limit of certain high-voltage lines, so that meteorological personnel can provide more targeted meteorological warnings and recommendations.

[0121] Geographic information professionals, such as geographic information system (GIS) engineers and professionals engaged in power geographic information analysis, can distribute detailed geographic distribution data of the power network in the typhoon-affected area, including line directions, substation locations, etc.; information on topographic changes in the affected area, such as changes in the terrain around power facilities caused by landslides and mudslides caused by typhoons; and geographic information-based power facility damage risk assessment models and related parameters for more in-depth analysis.

[0122] 3. Role chain:

[0123] Emergency repair command roles: such as the commander-in-chief of the power emergency repair site, leaders of each emergency repair team, etc., can distribute the following information: a comprehensive power emergency repair situation map, which integrates the real-time progress, personnel distribution, equipment status and other information of all emergency repair sites, and displays it in an intuitive graphical way to facilitate a comprehensive grasp of the overall situation; coordination and communication records and to-do items between departments and groups to ensure smooth information flow and timely handling of cross-departmental collaboration issues; important instructions and decision-making deployments from superiors to ensure that the emergency repair work is consistent with the overall strategic direction.

[0124] Safety supervision role: Conduct safety supervision on the emergency repair site, promptly discover and correct unsafe behaviors, and prevent the occurrence of safety accidents. For example, safety supervisors and relevant personnel of the safety management department can distribute information such as on-site safety inspection standards and specification lists to clarify the various safety requirements for power emergency repair operations in typhoon environments; real-time safety risk warning information, such as leakage risks in specific areas, risks of working at heights, etc.; cases and analysis of safety accidents that have occurred, especially accidents similar to the current emergency repair scenario, to serve as a warning.

[0125] External communication role: Responsible for conveying accurate power repair information to the public and the media, maintaining the company's image, and collecting public feedback to provide reference for company decision-making. For example, power company spokespersons, public relations specialists, etc. The information that can be distributed is the reviewed news releases on the power disaster situation, including power outage areas, expected power restoration time, repair progress and other information of public concern; communication script templates for different media and public platforms to ensure the consistency and accuracy of information conveyed; summary of public feedback and frequently asked questions to respond to social concerns in a timely manner.

[0126] In summary, the embodiment of the present invention provides an online distribution and push system based on emergency scenarios, including a data acquisition layer, a data processing layer and an information distribution and push layer connected in sequence, and introduces a scene recognition module. By analyzing the characteristic data of the current emergency scenario, the system parameters are automatically adjusted to adapt to the specific scenario requirements, and the identified scene information can be targetedly processed. The content distribution network established according to the hierarchical chain, professional chain and role chain can achieve accurate matching of distribution objects, and provide support for real-time assessment of disaster situations, real-time monitoring of disposal and rapid response. After receiving the key information pushed by the intelligent push, the command center and various professions make emergency command decisions, and through accurate push by hierarchical, professional and role levels, the command instructions can be quickly issued, responded to and executed in real time, and the progress can be monitored and accurately fed back. Key information can be processed and pushed in time, which greatly improves the efficiency of emergency handling.

[0127] In the several embodiments provided herein, it should be understood that the disclosed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative. For example, the division of units is merely a logical functional division. In actual implementation, other division methods may be used, such as combining or integrating multiple units or components into another system, or some features may be ignored or not implemented. In addition, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, which may be electrical, mechanical, or other forms.

[0128] Units described as separate components may or may not be physically separate, and components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0129] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0130] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the various embodiments of the method of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0131] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. An online distribution and push system based on emergency scenarios, characterized by: It includes the data collection layer, data processing layer and information distribution and push layer connected in sequence; The data collection layer is used to collect environmental data of the current emergency scenario and data that needs to be fed back by various departments through the sensor network, and transmit the data to the data processing layer; The data processing layer receives data transmitted by the data acquisition layer, analyzes and processes the data using machine learning algorithms, identifies the current emergency scenario type, and combines the power system's own operating data to determine the specific characteristics of the emergency scenario. Based on the identification results, the system parameters are automatically adjusted to obtain system parameters that match the current emergency scenario. The information distribution and push layer is used to classify the collected and processed data according to the needs of different types of emergency scenarios, and use different distribution and push algorithms to push the current emergency scenario information for different types of information.

2. The online distribution and push system based on emergency scenarios according to claim 1 is characterized in that: The data processing layer includes a scene recognition module and a parameter adjustment module, wherein: The scenario recognition module is used to receive environmental data from the data acquisition layer, analyze and process the data using a decision tree algorithm, identify the current emergency scenario type, and at the same time, determine the specific characteristics of the emergency scenario based on the power system's own operating data; The parameter adjustment module is used to automatically adjust system parameters according to the recognition results of the scene recognition module to obtain system parameters that match the current emergency scene.

3. The online distribution and push system based on emergency scenarios according to claim 2 is characterized in that: The execution process of the scene recognition module includes: Label the collected environmental data of emergency scenarios, collect various operating data of the power system, as well as the operating status and fault records of the equipment, and pre-process the collected power system operating data; Build a decision tree and train it using a labeled dataset; Input the collected environmental data of the current emergency scene into the scene recognition module to identify the type of emergency scene currently in place; Correlate the identified emergency scenario types with power system operation data; Establish a mapping relationship between emergency scenarios and changes in power system operating parameters; Determine the specific location of the power fault based on fault records and real-time monitoring data in the power system operation data. At the same time, analyze the topology and power flow distribution of the power network to assess the scope of the fault's impact on the power system. Assess the severity of power failures in emergency scenarios by considering the degree of deviation from power system operating parameters, equipment damage, and the impact on power supply; Combine the real-time operating status of the power system and the development trend of emergency scenarios to predict potential risks that may arise.

4. The online distribution and push system based on emergency scenarios according to claim 3 is characterized in that: The specific process of identifying the current emergency scene type in the scene recognition module is as follows: Collect emergency scene environmental data measured by various sensors; Extract key features from sensor data; Label the collected data and mark the emergency scenario type corresponding to each data sample; Use the labeled data set to train the constructed decision tree; Prune the trained decision tree; The collected data of the current emergency scene is input into the scene recognition module to identify the current emergency scene type.

5. The online distribution and push system based on emergency scenarios according to claim 4 is characterized in that: The process of the scene recognition module identifying the current emergency scene type includes: At each node of the decision tree, an optimal attribute is selected from all features as the partitioning attribute based on the information gain. According to the selected partitioning attribute, the data set is divided into different subsets. For each divided subset, the above steps of selecting the partitioning attribute and partitioning the data set are repeated, and the subtrees of the decision tree are recursively constructed until a certain stopping condition is met.

6. The online distribution and push system based on emergency scenarios according to claim 2 is characterized in that: In the parameter adjustment module, the execution process is as follows: Analyze the needs of different emergency scenarios, formulate system parameter adjustment strategies based on the needs of different scenarios, organize the correspondence between emergency scenarios and parameter adjustment strategies into a parameter adjustment rule library and store it. The system parameter adjustment strategy includes data processing priority parameters, distribution algorithm parameters, push frequency and range parameters; Analyze the recognition results and extract key information; According to the emergency scenario identification results after analysis, the matching rules are searched in the parameter adjustment rule library, and the system parameters are adjusted according to the matched rules.

7. The online distribution and push system based on emergency scenarios according to claim 1 is characterized in that: The information distribution and push layer includes an information classification submodule and a distribution and push algorithm module, wherein: The information classification submodule is used to classify the collected and processed data according to the needs of different emergency scenarios; The distribution and push algorithm module is used to push different types of information using different distribution and push algorithms.

8. The online distribution and push system based on emergency scenarios according to claim 7 is characterized in that: In the information classification submodule, information is divided into power facility information, emergency scenario feature information, personnel information, surrounding environment information and notification and warning information.

9. The online distribution and push system based on emergency scenarios according to claim 8 is characterized in that: The classification process in the information classification submodule includes: Classification rules are formulated based on the characteristics and definitions of different categories of information. The information classification submodule reads the collected and processed data and matches them according to the pre-established rules; Collect historical data with labeled categories as training sets and train machine learning algorithms; Associating emergency scenario types with scenario feature information of corresponding categories; The information classification submodule obtains the output results of the scene recognition module and classifies the corresponding data into specific categories according to the current emergency scene type and the pre-set association relationship.

10. The online distribution and push system based on emergency scenarios according to claim 7 is characterized in that: The information distribution and push layer also includes a multi-channel push module for pushing information through multiple channels.

Images