A crane risk event handling method, a terminal device, and a storage medium

Abstract

The application relates to the technical field of engineering machinery safety monitoring, and discloses a crane risk event disposal method, a terminal device and a storage medium, which comprises the following steps: performing alarm identification on original operation data of each crane to generate a crane risk event; calling an instruction template according to a risk feature of the crane risk event and filling parameters matched with template content placeholders in the corresponding placeholders to generate a business disposal instruction; distributing the instruction to a corresponding disposal object; if instruction confirmation information of the disposal object is received, searching for a disposal plan in a risk case database based on the risk feature and the identification of the business disposal instruction to create a risk event disposal task and send the risk event disposal task to the disposal object for execution. The application realizes the automation of risk disposal, improves response accuracy and execution consistency, ensures that the state of each link is traceable, the behavior is traceable, and the responsibility is definable, and significantly enhances the intelligence, robustness and continuous evolution capability of the crane operation safety risk control system.

Description

Technical Field

[0001] This application relates to the field of engineering machinery safety monitoring technology, and in particular to a method for handling crane risk events, terminal equipment and storage medium. Background Technology

[0002] Currently, sensor-based data acquisition systems are widely deployed in the field of crane safety monitoring. These systems can acquire operating parameters such as wind speed, load, tilt angle, and GPS location in real time and upload them to the monitoring platform via wired or wireless means. Most mainstream platforms adopt a B / S or C / S architecture and have basic data storage, threshold over-limit alarm, equipment positioning, and historical curve playback functions. Some advanced systems have introduced message queues (such as RocketMQ) to achieve data decoupling or use Flink / Kafka for streaming computing, initially supporting alarm aggregation and simple rule judgment. On the terminal side, mobile apps and web dashboards have begun to be integrated, supporting alarm push and status viewing, providing a certain level of information support for on-site management.

[0003] Existing technologies still suffer from three prominent shortcomings: First, data processing is crude, leading to inaccurate risk identification—raw data lacks standardized cleaning (such as deduplication, anomaly detection, field completion, and quality marking) and geographic semantic enhancement, resulting in low data credibility input into risk control models and difficulty in supporting the accurate quantification of complex risks; Second, instruction generation is disconnected, leading to inefficient handling and execution—alarm information cannot be automatically converted into role-oriented, context-rich, and optional structured instructions, relying on manual translation which is prone to errors, delayed responses, and unclear responsibilities; Third, a closed-loop mechanism is lacking, resulting in weak control capabilities—there is no PDCA state machine-driven task flow, no mandatory binding of execution evidence and on-site verification, and no operation feedback-driven risk parameter and instruction strategy self-optimization mechanism, causing risk handling to become a mere formality and failing to form a continuous improvement security governance closed loop. Summary of the Invention

[0004] In view of this, the embodiments of this application provide a crane risk event handling method, terminal equipment and storage medium, which can effectively solve the problems in the prior art such as the inability to identify compound risks caused by multiple alarm coupling, the disconnect between instruction content and real-time risk status, and the lack of end-to-end closed-loop tracking and feedback optimization capabilities in the handling process.

[0005] In a first aspect, embodiments of this application provide a method for handling crane risk events, including: The system acquires raw operating data from multiple cranes, performs alarm identification processing on the raw operating data of each crane, and generates corresponding crane risk events. Based on the risk characteristics of the crane risk event, a preset instruction template is invoked, and the parameters in the crane risk event that match the content placeholder of the instruction template are filled into the content placeholder of the instruction template to generate a business processing instruction carrying a unique instruction identifier. The business processing instructions are distributed to the corresponding target objects through preset delivery channels; If the target object receives an instruction confirmation message for the business handling instruction, then based on the risk characteristics of the crane risk event and the unique instruction identifier of the business handling instruction, a risk handling plan is retrieved from the preset risk case database to create a risk event handling task and send it to the target object for execution.

[0006] In some embodiments, acquiring raw operating data from multiple cranes and performing alarm identification processing on the raw operating data of each crane to generate corresponding crane risk events includes: The system acquires raw operating data reported by each crane device through a preset protocol, and performs protocol parsing, external geographic information enhancement, and multi-dimensional data cleaning on the raw operating data to obtain standardized data. Based on preset alarm rules, the standardized data is subjected to time-series pattern matching to identify multiple discrete alarm events that satisfy preset correlation relationships; Based on the risk characteristics of the multiple discrete alarm events, the severity scores of each alarm event are extracted and accumulated. The accumulated results are then mapped according to a preset grading rule to determine the corresponding risk level. The multiple discrete alarm events and their corresponding risk levels are aggregated to form a corresponding structured crane risk event.

[0007] In some embodiments, the step of invoking a preset instruction template based on the risk characteristics of the crane risk event, and filling the content placeholders of the instruction template with parameters from the crane risk event that match the content placeholders of the instruction template to generate a business processing instruction carrying a unique instruction identifier, includes: Based on the equipment type and event type of the crane risk event, a target instruction template is matched from a preset instruction template library; The parameters in the crane risk event that match the content placeholder of the target instruction template are filled into the content placeholder of the target instruction template to generate the instruction content; Based on the risk level of the crane risk event, the final recommended action is determined from among the multiple recommended actions associated with the target instruction template; Based on the equipment type and event type of the crane risk event, a list of target handling roles is determined, and the instruction content, the final recommended action, the list of target handling roles, and the instruction validity period are encapsulated to generate a business handling instruction carrying a unique instruction identifier.

[0008] In some embodiments, distributing the business processing instruction to the corresponding target processing object through a preset delivery channel includes: Based on the list of handling roles with crane management authority specified in the business handling instruction and the equipment identifier of the crane risk event, query the target handling object; Based on the real-time online status of the target object, the optimal delivery channel is selected from the preset multi-level communication channels; The business processing instruction is distributed to the target processing object through the optimal delivery channel, and a timed task associated with the business processing instruction is triggered. If no instruction confirmation information carrying the unique instruction identifier and operator identity identifier of the business processing instruction is received within the preset time, the backup delivery channel is triggered to resend the instruction. Receive instruction confirmation information returned by the target object, which carries a unique instruction identifier and an operator identity identifier for the business processing instruction.

[0009] In some embodiments, the step of retrieving a risk handling plan from a preset risk case database based on the risk characteristics of the crane risk event and the unique instruction identifier of the business handling instruction to create a risk event handling task includes: In response to the instruction confirmation information, update the status of the service processing instruction according to the received confirmation status; Extract the risk characteristics of the crane risk event and the unique instruction identifier of the business handling instruction; Based on the extracted risk characteristics, a matching risk response plan is retrieved from a pre-set risk case database; Based on the historical success rate of each of the risk response plans, the risk response plans are sorted. Obtain the contents of the risk response plan that is ranked first; Based on the obtained contingency plan content, the crane risk event, and the unique instruction identifier of the business handling instruction, a risk event handling task is constructed.

[0010] In some embodiments, after creating the risk event handling task, the method further includes: The legal states and state transition rules of the risk event handling task are recorded using a state machine model. The types of states include pending dispatch, assigned, accepted, processing, pending verification, completed, upgraded, and canceled. When the risk event handling task is created, its status is initialized to pending dispatch; When the risk event handling task is assigned to the target object, the status of the risk event handling task is updated to "assigned". Upon receiving confirmation information from the target object regarding the business handling instruction, the status of the risk event handling task is updated to "accepted". Upon receiving a step execution request initiated by the target object, the status of the risk event handling task is updated to "processing". Upon receiving the task execution result information submitted by the target object, the status of the risk event handling task is updated to pending verification; When verifying the authenticity of the on-site handling by using evidence links based on the task execution result information, the status of the risk event handling task is updated to completed; Upon receiving an upgrade request initiated by the target object, or when the risk event handling task is not completed within a preset timeout period, the status of the risk event handling task is updated to upgraded. Upon receiving a cancellation request initiated by the target object, or when the business processing instruction is revoked, the status of the risk event processing task is updated to "cancelled".

[0011] In some embodiments, the method further includes: In response to the selection operation of the monitoring terminal user on the crane risk event, based on the crane equipment identifier associated with the crane risk event, query the standardized operation data generated by data cleaning and standardization within a preset time window; Based on the load rate and wind speed values ​​contained in the standardized operating data, a visual display data carrying load rate curves, wind speed curves and corresponding reference lines is generated. The visualized data is pushed to the monitoring terminal through a real-time communication channel.

[0012] In some embodiments, the method further includes: Receive risk level correction feedback information or false alarm labeling feedback information submitted by the target handling object in response to the crane risk event; wherein, the risk level correction feedback information includes at least the original risk level and the corrected risk level, and the false alarm labeling feedback information includes at least the alarm event identifier or alarm code marked as a false alarm and the reason for the false alarm; Based on the risk level correction feedback information, update the mapping relationship between the total severity score range and the risk level in the preset grading rules; Based on the false alarm label feedback information, update the alarm triggering conditions and / or timing association conditions in the preset alarm rules corresponding to the alarm events marked as false alarms.

[0013] Secondly, embodiments of this application provide a crane risk event handling device, comprising: The alarm identification and processing module is used to acquire the original operating data of multiple cranes, and to perform alarm identification processing on the original operating data of each crane to generate corresponding crane risk events. The business processing module is used to fill the content placeholder of the instruction template with the parameters in the crane risk event that match the content placeholder of the instruction template according to the risk characteristics of the crane risk event, so as to generate a business processing instruction carrying a unique instruction identifier. The instruction distribution module is used to distribute the business processing instructions to the corresponding target processing objects through preset delivery channels; The contingency plan retrieval module is used to, if it receives instruction confirmation information from the target disposal object regarding the business disposal instruction, retrieve risk disposal contingency plans from a preset risk case database based on the risk characteristics of the crane risk event and the unique instruction identifier of the business disposal instruction, so as to create a risk event disposal task and send it to the target disposal object for execution.

[0014] Thirdly, embodiments of this application provide a terminal device, the terminal device including a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement the crane risk event handling method of the first aspect described above.

[0015] Fourthly, embodiments of this application provide a computer-readable storage medium, wherein when the computer program is executed on a processor, it implements the crane risk event handling method of the first aspect described above.

[0016] The embodiments of this application have the following beneficial effects: By acquiring the original operating data of multiple cranes and performing alarm identification processing, structured crane risk events are generated, overcoming the shortcomings of traditional single-point alarm systems that only output discrete alarms and cannot identify composite risks, thus elevating risk assessment from "equipment anomaly" to "operational scenario risk"; by calling a preset instruction template according to the risk characteristics of the risk event and filling the content placeholders of the instruction template with the parameters in the crane risk event that match the content placeholders of the instruction template, a business handling instruction carrying a unique instruction identifier is generated, achieving strong coupling between the content of the handling instruction and the current risk situation, avoiding response delays and semantic deviations caused by manually writing instructions; the business handling instruction is distributed to the target through delivery channels. The system dynamically selects the optimal communication channel based on the real-time online status of the target, and automatically triggers a resend through a backup channel if confirmation is not received in a timely manner, ensuring the delivery and timeliness of instructions. Upon receiving instruction confirmation information carrying a unique instruction identifier and operator identification, the system searches for matching risk handling plans in a pre-set risk case database based on risk characteristics and instruction identifiers, and selects the optimal plan according to its historical execution success rate to construct risk event handling tasks. This shifts the handling action from experience-driven to data-driven, improving the scientific nature and reproducibility of handling measures. The entire method uses a unique instruction identifier to connect the entire process of data perception, instruction generation, delivery, confirmation, and handling, forming an end-to-end traceability capability and providing a complete audit chain and traceability support for the risk handling process. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart of a crane risk event handling method according to an embodiment of this application is shown; Figure 2 This paper illustrates yet another flowchart of a crane risk event handling method according to an embodiment of this application; Figure 3 Another flowchart of the crane risk event handling method according to an embodiment of this application is shown; Figure 4 A schematic diagram of a crane risk event handling device according to an embodiment of this application is shown. Detailed Implementation

[0019] The technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments.

[0020] The components of the embodiments of this application described and illustrated in the accompanying drawings can be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of this application provided in the drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0021] In the following text, the terms "comprising," "having," and their cognates, which may be used in various embodiments of this application, are intended only to indicate a particular feature, number, step, operation, element, component, or combination thereof, and should not be construed as primarily excluding the presence of one or more other features, numbers, steps, operations, elements, components, or combinations thereof, or adding the possibility of one or more combinations thereof. Furthermore, the terms "first," "second," "third," etc., are used only for distinguishing descriptions and should not be construed as indicating or implying relative importance.

[0022] Unless otherwise specified, all terms used herein (including technical and scientific terms) shall have the same meaning as commonly understood by one of ordinary skill in the art to which the various embodiments of this application pertain. Terms (such as those defined in commonly used dictionaries) shall be interpreted as having the same meaning as in their contextual meaning in the relevant technical field and shall not be construed as having an idealized or overly formal meaning, unless clearly defined in the various embodiments of this application.

[0023] The following detailed description of some embodiments of this application is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0024] Considering the shortcomings of existing technologies, such as the inability to identify complex risks caused by multiple coupled alarms, the disconnect between instruction content and real-time risk status, and the lack of end-to-end closed-loop tracking and feedback optimization capabilities in the handling process, a crane risk event handling method is proposed. This method involves acquiring raw operating data from multiple cranes and performing alarm identification processing on the raw operating data of each crane to generate corresponding crane risk events. Based on the risk characteristics of the crane risk events, a preset instruction template is invoked, and parameters from the crane risk events that match the content placeholders of the instruction template are filled into the content placeholders of the instruction template to generate a business handling instruction carrying a unique instruction identifier. The business handling instruction is distributed to the corresponding target handling object through a preset delivery channel. If an instruction confirmation message for the business handling instruction is received from the target handling object, a risk handling plan is retrieved from a preset risk case database based on the risk characteristics of the crane risk event and the unique instruction identifier of the business handling instruction to create a risk event handling task and send it to the target handling object for execution.

[0025] The following examples illustrate the method for handling risk events related to this crane.

[0026] Figure 1 A flowchart illustrating a crane risk event handling method according to an embodiment of this application is shown. Exemplarily, the crane risk event handling method includes the following steps: Step S100: Obtain the original operating data of multiple cranes, and perform alarm identification processing on the original operating data of each crane to generate corresponding crane risk events.

[0027] Here, "multiple cranes" refers to several crane devices connected to the system via onboard terminals; "raw operating data" refers to JSON format data packets reported by the crane onboard terminals via message queue transmission protocol, containing device identifiers, timestamps, sensor data, GPS coordinates, and alarm information. "Alarm identification and processing" refers to using a complex event processing engine to perform time-series pattern matching on standardized data according to alarm rules, identifying multiple discrete alarm events, and aggregating them to generate crane risk events; "crane risk events" refers to structured data objects generated after aggregation, containing globally unique event identifiers, device identifiers, event types, risk levels, lists of triggered alarm events, and timestamps. Further optionally, crane risk events may also include features such as device type, wind speed, load rate, boom length, and risk score.

[0028] For example, the system receives raw operating data reported by multiple cranes; it performs protocol parsing on each data entry to extract equipment identifiers, timestamps, sensor data, GPS coordinates, and alarm information; based on preset alarm rules, when a wind speed exceeding limit alarm and a load rate exceeding limit alarm occur sequentially within a five-minute time window under the same equipment identifier, these two alarm events are used as matching results; a total score is obtained by summing the severity scores of the two alarm events, and the risk level is determined based on the total score; then, using a globally unique string as the event identifier, the equipment identifier as the equipment identifier, composite risk as the event type, and the determined risk level as the risk level, a list of triggering alarm events is formed using the two alarm events, and the timestamp of the starting event in the matching result is used as the timestamp to encapsulate and generate a crane risk event.

[0029] In an optional embodiment, the wind speed and load rate values ​​from the two alarm events can also be extracted and substituted into the formula: Risk Score = 0.4 × Load Rate + 0.3 × (Wind Speed ​​Value ÷ 12.0) to calculate the risk score, and the risk score can be used as a quantitative representation of the crane risk event.

[0030] For example, all steps of this method are executed by a stream processing job. A stream processing job is a distributed real-time data processing program instance built on the Flink streaming computing framework, deployed on a standard Linux server cluster, and horizontally scaled by configuring task parallelism. The job includes a data source component, a transformation component, and a data receiver component. The data source component is used to access data from the message queue transmission protocol middleware, the transformation component is used to perform protocol parsing, geographic information augmentation, data cleaning, time-series pattern matching, and risk level determination, and the data receiver component is used to distribute the results to the downstream message queue system.

[0031] In an optional embodiment, step S100 includes the following sub-steps: S101: Obtain the raw operating data reported by each crane device through a preset protocol, and perform protocol parsing, external geographic information enhancement, and multi-dimensional data cleaning on the raw operating data to obtain standardized data.

[0032] Protocol parsing refers to decoding the message payload of the message queue transmission protocol into a string according to character encoding and parsing it into a JSON object, from which device identifiers, timestamps, sensor data, GPS coordinate information, and alarm information are extracted. Alarm information refers to the structured data contained in the alarm field of the JSON object, which must contain the fields code (alarm code), level (alarm level), and timestamp (alarm occurrence time). The alarm code code can take values ​​such as wind_speed_exceed (wind speed exceeding the limit alarm), load_ratio_high (load ratio exceeding the limit alarm), or boom_angle_exceed (boom angle exceeding the limit alarm). The alarm level can take values ​​such as LOW (low risk level), MEDIUM (medium risk level), WARNING (high risk level), or CRITICAL (critical risk level). External geographic information enhancement refers to calling a geographic information service system to obtain inverse geocoding, real-time weather, and traffic restriction information based on GPS coordinates, and caching the query results through a distributed caching system cluster. Multi-dimensional data cleaning includes: constructing deduplication keys based on device identifiers and timestamps, performing write operations on the distributed caching system cluster only if the key does not exist; discarding the data if the operation fails, and setting a 60-second expiration time if successful; extracting wind speed values ​​from sensor data, querying the distributed caching system cluster for statistical keys, obtaining historical mean and standard deviation; if the absolute value of the difference between the wind speed value and the historical mean is greater than three times the historical standard deviation, setting the quality flag as "suspect"; if the sensor data lacks a load rate field but has actual and rated loads, then... The load rate is calculated as the actual load divided by the rated load, and written into the sensor data. The millisecond-level timestamp is truncated to the second-level precision. Standardized data refers to data objects that include device identifiers, aligned timestamps, completed sensor data, GPS coordinates, geographic information objects, quality labels, and alarm information retained after cleaning. Quality labels are graded markers for the reliability of standardized data, with values ​​including valid (data is complete and without anomalies), suspect (data has suspected anomalies and requires manual verification), and invalid (data is severely distorted or missing key fields and will not participate in subsequent alarm identification and processing). Before performing time-series pattern matching in S102, the system automatically filters out records with the quality label of invalid to ensure that alarm events are identified only based on reliable data.

[0033] As an example, the device identifier, timestamp, sensor data, GPS coordinates, and alarm information are extracted from the JSON payload. Cache keys are constructed according to their intended use: the cache key format for deduplication is `dedup:{device identifier}:{timestamp}`, and the cache key format for geographic information queries is `stats:{device identifier}:wind speed`. Geographic information is queried from the distributed cache system cluster. If no match is found, the geographic information service system interface is called to obtain the response and write it to the distributed cache system cluster, setting an expiration time of 3600 seconds. Deduplication is performed. Historical statistics are queried, and wind speed is assessed for anomalies. If the sensor data does not contain a load rate but contains actual and rated loads, the load rate is calculated and written. The timestamp is truncated to the second level. Standardized data is then generated through encapsulation.

[0034] S102, perform time-series pattern matching on standardized data according to preset alarm rules, and identify multiple discrete alarm events that meet preset correlation relationships.

[0035] Among them, preset alarm rules refer to the judgment conditions used to trigger alarm identification and processing, as well as the composite rules formed by their combination; for example, wind speed exceeding the standard alarm, load rate exceeding the standard alarm, etc.; time sequence pattern matching refers to the pattern of a starting event followed by a subsequent event defined by the Complex Event Processing Engine (CEP), and the two occurring within a specified time window; preset association means that two alarm events must have the same device identifier, and the difference between the occurrence time of the subsequent event and the occurrence time of the starting event does not exceed 300 seconds; multiple discrete alarm events refer to records in standardized data where the alarm information is not empty and the alarm code matches the preset type.

[0036] For example, a record with a non-empty alarm message and an alarm code indicating excessive wind speed is mapped as the starting event, and a record with a non-empty alarm message and an alarm code indicating excessive load rate is mapped as the subsequent event; when the starting event and the subsequent event are detected to have the same device identifier, and the difference between the occurrence time of the subsequent event and the occurrence time of the starting event does not exceed 300 seconds, the two alarm events are used as the matching result.

[0037] For example: the starting event corresponds to the alarm code wind_speed_exceed (wind speed exceeds the standard), and the subsequent event corresponds to the alarm code load_ratio_high (load ratio is too high). The time window is within(Time.minutes(5)) (within a five-minute time window).

[0038] S103: Based on the risk characteristics of multiple discrete alarm events, extract the severity score of each alarm event and accumulate it. Map the accumulated result according to the preset grading rules to determine the corresponding risk level.

[0039] The risk features used to determine the risk level include the severity score of each alarm extracted from the matching results. The severity score of each alarm depends on the value of the `level` field in the alarm information: 20 points for a value of LOW (low risk), 40 points for a value of MEDIUM (medium risk), 60 points for a value of WARNING (high risk), and 100 points for a value of CRITICAL (critical risk). The preset grading rule refers to the mapping rule for determining the risk level based on the total accumulated severity score. The risk level refers to the qualitative classification obtained according to the mapping rule, and the risk level categories include critical risk, high risk, medium risk, and low risk. For example, in one implementation, the correspondence between each risk level and the total severity score is as follows: a total score ≥ 80 indicates critical risk, 80 > total score ≥ 60 indicates high risk, 60 > total score ≥ 30 indicates medium risk, and a total score < 30 indicates low risk.

[0040] As an example, the severity scores corresponding to multiple discrete alarm events are extracted from the matching results; the extracted severity scores are summed to obtain a total score; and the corresponding risk level is determined according to the interval in which the total score falls, based on the preset grading rules.

[0041] S104 aggregates multiple discrete alarm events and their corresponding risk levels to form a corresponding structured crane risk event.

[0042] Aggregation refers to encapsulating multiple discrete alarm events, determined risk levels, device identifiers, globally unique event identifiers, and timestamps from the matching results into a data object according to a fixed field structure.

[0043] As an example, a structured crane risk event is generated by encapsulating the following: a globally unique string is used as the event identifier, the device identifier in the matching result is used as the device identifier, the composite risk is used as the event type, the determined risk level is used as the risk level, multiple discrete alarm events in the matching result are used to form a list of triggering alarm events, and the timestamp of the starting event in the matching result is used as the timestamp.

[0044] For example, the event identifier is e99a8c7ecc724d8a9e5a2d4f7d7f8f7c, and the event type is COMPLEX_RISK.

[0045] In one optional embodiment, after identifying each discrete alarm event, relevant parameters are extracted from the risk characteristics of these alarm events and a risk score is calculated according to a preset calculation formula.

[0046] The risk score refers to the quantitative result obtained by extracting wind speed and load factor values ​​from multiple discrete alarm events according to a preset calculation formula. In one implementation, the preset calculation formula is: Risk Score R = 0.4 × Load Factor + 0.3 × (Wind Speed ​​Value ÷ 12.0); where 0.4 and 0.3 are preset coefficients, and 12.0 is the wind speed baseline value.

[0047] As an example, wind speed and load rate values ​​are extracted from multiple discrete alarm events. The extracted wind speed and load rate values ​​are substituted into a preset calculation formula to obtain the corresponding risk score. The risk score is then stored or displayed as a quantitative representation of crane risk events.

[0048] Further optionally, in response to situations where devices continuously trigger the same risk in real-world scenarios, this embodiment designs a dual deduplication mechanism to avoid overwhelming downstream systems with repeated or similar alarms. Specifically, it uses a configured quiet period (e.g., 30 seconds) to filter out instantaneous repeated risk events to achieve short-term suppression; at the same time, it also uses a caching mechanism to merge similar risk events with parameter fluctuations less than a preset value (e.g., 5%) over a longer period (e.g., 1 hour) to achieve long-term similarity detection.

[0049] In one implementation, the aforementioned risk score is used to calculate the similarity between two adjacent risk events. Specifically, the difference in risk scores is used to determine whether the current risk event is similar to the previous risk event, and then the current risk event is processed accordingly. For example, if the current risk event is similar to the previous risk event and is within a quiet period, the similarity event counter is only incremented (no new event is generated). In other words, any recurring identical risk event will be discarded without subsequent instruction generation or other operations. Conversely, if two adjacent risk events are not similar, the current risk event is reported as a new risk event and cached.

[0050] For example, if device CRANE001 triggers two COMPLEX_RISK events (both with a risk score of 0.75) within 10 seconds, the first risk event will be checked during a silent period (e.g., 30 seconds), reported normally, and a silent period will be set. The second risk event will be discarded during the silent period.

[0051] For example, if device CRANE001 triggers risk event 1 with a risk score of 0.74, and then triggers risk event 2 10 minutes later with a risk score of 0.77, meaning the risk score difference is 0.03 < the preset value (e.g., 0.05), then risk event 1 is reported and cached as a new risk event; while risk event 2 is judged as a similar event, so only the event counter is incremented, and no new risk event is generated.

[0052] It is understandable that the quiet period control is mainly used to prevent repeated alarms in a short period of time; similar event merging is mainly used to reduce invalid alarms caused by parameter fluctuations. By making dual judgments based on the quiet period and the difference in risk scores, invalid alarm transmission can be greatly reduced (the measured data shows that it is more than 50%), thereby reducing the system load. In addition, by merging similar events, the downstream system can analyze the persistence of risk based on counters (such as counter count>3 indicating continuous deterioration).

[0053] Step S200: Based on the risk characteristics of the crane risk event, a preset instruction template is invoked, and the parameters in the crane risk event that match the content placeholders of the instruction template are filled into the content placeholders of the instruction template to generate a business processing instruction carrying a unique instruction identifier.

[0054] Among them, the risk characteristics of a crane risk event refer to the equipment type, event type, risk level, wind speed value, load rate value, and boom length value of the event; the preset instruction template refers to the structured text skeleton pre-configured in the database, which includes business semantic placeholders and a set of recommended actions matching the risk level; the instruction template library refers to the document-oriented database that stores all instruction templates; the content placeholder refers to the variable marker enclosed in curly braces in the template; the unique instruction identifier refers to the string generated by the globally unique identifier generator, which is used to accurately track the entire lifecycle of a single instruction in the entire system; the business processing instruction refers to the structured data object that includes the instruction title, the filled instruction content, the target processing role list, the instruction validity period, the recommended actions, and the unique instruction identifier.

[0055] For example, the following steps are taken to extract the equipment type (tower crane), event type (high wind and high load operation risk), risk level (high risk), wind speed (15.2 m / s), load rate (0.85), and boom length (45 m) from a crane risk event. Based on the equipment type and event type, an instruction template with the template identifier WIND_HIGH_LOAD (high wind and high load) is matched from the instruction template library. The parameters in the crane risk event that match the content placeholders of the instruction template are filled into the content placeholders of the instruction template to generate the instruction content. Based on the high risk level, the final recommended action is PAUSE_OPERATION (pause operation) from the multiple recommended actions associated with the template. Based on the equipment type and event type, the target handling roles are determined as driver, site manager, and safety supervisor. The instruction content, the final recommended action, the target handling role list, the 60-second validity period, and the globally unique string are encapsulated to generate a business handling instruction.

[0056] For example, a stream processing job calls the backend instruction generation service interface, passing in the template identifier WIND_HIGH_LOAD (high wind and high load), context mapping table, risk event identifier, and device identifier, and receives the returned business processing instruction object.

[0057] In one alternative embodiment, such as Figure 2 As shown, step S200 includes the following sub-steps: S201, based on the equipment type and event type of the crane risk event, match the target instruction template from the preset instruction template library.

[0058] Among them, equipment type refers to the mechanical structure classification of cranes; event type refers to the business semantic classification identifier of crane risk events; target instruction template refers to the pre-set text template selected from the instruction template library that matches the current equipment type and event type.

[0059] For example, extract the equipment type and event type of crane risk events; query the instruction template library and filter out templates whose equipment type field matches the extracted equipment type and whose event type field matches the extracted event type; if multiple matching templates exist, select the one with the highest historical success rate; if no matching template exists, use a general template.

[0060] For example, when the equipment type of a crane risk event is TOWER_CRANE (tower crane) and the event type is WIND_HIGH_LOAD (high wind and high load), the system matches the instruction template with the template identifier WIND_HIGH_LOAD; if multiple matching templates exist, the PLAN_WIND_01 (high wind response plan No. 1) with the highest historical success rate is selected; if no match is found, GENERIC_PLAN (general template) is selected.

[0061] S202: Fill the content placeholder of the target instruction template with the parameters in the crane risk event that match the content placeholder of the target instruction template, and generate the instruction content.

[0062] Among them, the parameter that matches the content placeholder refers to the field value carried in the crane risk event whose data meaning is consistent with the business semantics identified by a content placeholder in the target instruction template; the content placeholder refers to the variable marker enclosed in curly braces in the template.

[0063] For example, extract the wind speed value of 15.2, load ratio value of 0.85, boom length value of 45, and risk level label of HIGH from the crane risk event; fill 15.2 into the {v} position, 0.85 into the {η} position, and 45 into the {L} position to get "Current wind speed 15.2m / s, load ratio 0.85, boom length 45m, there is a risk of overturning. {action}"; keep the {action} placeholder for later filling.

[0064] S203, based on the risk level of the crane risk event, determine the final recommended action from among the multiple recommended actions associated with the target instruction template.

[0065] Among them, the recommended action refers to the predefined operational suggestions in the template for a specific risk situation; the final recommended action refers to the handling suggestion that is most suitable for the current risk situation, determined from multiple recommended actions based on the matching relationship between the risk level and the recommended action.

[0066] In one implementation, the matching relationship between recommended actions and risk levels is as follows: Serious risk: Matching EMERGENCY_STOP (emergency braking); High risk: Matches PAUSE_OPERATION (pause job); Medium risk: Matching REDUCE_BOOM (shortening the boom); Low risk: Match ENHANCE_MONITOR (Enhanced Monitoring).

[0067] S204: Based on the equipment type and event type of the crane risk event, determine the target handling role list, and encapsulate the instruction content, final recommended action, target handling role list and instruction validity period to generate a business handling instruction carrying a unique instruction identifier.

[0068] Among them, the target handling role list refers to the set of operator roles that have been certified by the permission system and have corresponding management responsibilities; the instruction validity period refers to the time threshold for a business handling instruction to remain valid from the time it is generated; and the unique instruction identifier refers to a string generated by the globally unique identifier generator.

[0069] For example, based on the equipment type and event type of a crane risk event, the role configuration rules are queried to determine the corresponding list of target handling roles. The instruction content, the final recommended action, the list of target handling roles, the instruction validity period, and the globally unique string are encapsulated to generate a business handling instruction object. The role configuration rules refer to the role mapping relationship built into the system with equipment type and event type as the joint key. It records the list of target handling roles corresponding to each combination of equipment type and event type. When the equipment type is TOWER_CRANE (tower crane) and the event type is WIND_HIGH_LOAD (high wind and high load), the list of handling roles corresponding to this rule is ["driver","site_manager","safety_supervisor"].

[0070] For example: The instruction is valid for 60 seconds, and the target handling roles are driver, site manager, and safety supervisor.

[0071] For example, when the equipment type of a crane risk event is TOWER_CRANE and the event type is WIND_HIGH_LOAD, the system determines the target handling role list as driver, site manager, and safety supervisor; the instruction validity period is 60 seconds.

[0072] Step S300: Distribute the business processing instructions to the corresponding target processing objects through the preset delivery channels.

[0073] The delivery channel refers to a multi-level communication path dynamically selected based on the real-time status of the target object, including a first-level real-time communication channel, a second-level instant messaging channel, and a third-level backup communication channel; the target object refers to the operator who has been certified by the authorization system and has the corresponding crane management responsibilities, and their identity information and role tags are stored in a relational database; the business processing instruction refers to a structured data object carrying a unique instruction identifier, instruction content, a list of target processing roles, instruction validity period, and recommended actions; the instruction confirmation information refers to a deterministic response returned by the target object through any delivery channel, carrying a unique instruction identifier and operator identity identifier.

[0074] For example, based on the list of target handling roles specified in the business handling instruction and the equipment identifier of the crane risk event, a relational database is queried to obtain target handling objects that meet the permission conditions; based on the real-time online status of the target handling objects, the optimal delivery channel is selected from the preset multi-level communication channels; the business handling instruction is distributed to the target handling object through the selected channel, and a timed task associated with the instruction is triggered; if no instruction confirmation information carrying the unique instruction identifier and operator identification is received within a preset time, the backup delivery channel is triggered for retransmission; the instruction confirmation information carrying the unique instruction identifier and operator identification is received from the target handling object.

[0075] In one alternative embodiment, such as Figure 3 As shown, step S300 includes the following sub-steps: S301, based on the list of handling roles with crane management authority specified in the business handling instruction and the equipment identifier of the crane risk event, query the target handling object.

[0076] Among them, the list of handling roles with crane management authority refers to the set of operator roles preset in the business handling instructions and verified by the authorization system; the equipment identifier refers to the unique identifier of the equipment carried in the crane risk event; As an example, based on the list of target handling roles specified in the business handling instruction and the equipment identifier of the crane risk event, a query is initiated to the relational database to obtain the information of the operators in the current shift who have the corresponding roles and are in a valid position, including the personnel's unique identifier, DingTalk user identifier, mobile phone number, and real-time online status identifier.

[0077] S302, based on the real-time online status of the target object, selects the optimal delivery channel from the preset multi-level communication channels.

[0078] Among them, real-time online status refers to whether the target object is currently in a connected state that can receive real-time messages, which is obtained by querying the online status identifier in the distributed cache system with personnel:online (personnel online status) and web_online (webpage online) as keys; multi-level communication channels refer to communication paths classified according to real-time performance and reliability, including the first-level real-time communication channel (WebSocket), the second-level instant messaging channel (WeChat or DingTalk work notifications), and the third-level backup communication channel (SMS); the optimal delivery channel refers to prioritizing the available channel with the highest real-time performance while meeting the accessibility requirements.

[0079] For example, query the online status identifier corresponding to the target object in the distributed caching system; if the status identifier is WEB_ONLINE (webpage online), select the first-level real-time communication channel; if the status identifier is empty but a DingTalk user identifier exists, select the second-level instant messaging channel; if none of the above conditions are met, force the third-level backup communication channel to be enabled.

[0080] S303 distributes business processing instructions to the target processing objects through the optimal delivery channel and triggers the timed tasks associated with the business processing instructions.

[0081] Among them, the timed task refers to the countdown monitoring mechanism set for business processing instructions, which is used to trigger the backup plan if no confirmation is received within the preset time; the preset time refers to the confirmation waiting threshold set separately outside the validity period of the instruction.

[0082] As an example, the business processing instruction is serialized into a JSON string to construct the message body; the message target topic is set to RISK_COMMAND (risk instruction topic) and the tag is DELIVERY (delivery task); the message queue system producer interface is called to send the message; at the same time, a countdown task is registered in the distributed task scheduling system, and its trigger time is the time when the business processing instruction is generated plus thirty seconds.

[0083] S304: If no instruction confirmation information carrying a unique instruction identifier and operator identification identifier is received within a preset time, the backup delivery channel is triggered to resend the instruction.

[0084] Among them, the backup delivery channel refers to the next-level communication channel that is switched to according to a predetermined priority when the primary delivery channel fails; the instruction confirmation information refers to the response data returned by the target object through any delivery channel, which carries the unique instruction identifier of the business processing instruction and the operator's identity identifier.

[0085] For example, when a countdown task is triggered, the system queries the distributed cache system for confirmation records with the key command:confirmed. If no confirmation record matching the unique instruction identifier is found, the system attempts to send a work notification to the target's DingTalk account and an SMS to their mobile phone number in a preset priority order. If all of the above operations fail, the system calls the voice service to initiate an outbound phone call.

[0086] S305: Receive instruction confirmation information returned by the target object, which carries a unique instruction identifier for business processing instructions and an operator's identity identifier.

[0087] Among them, the operator identity identifier refers to the unique identity credential registered by the target object in the system, including the unique personnel identifier or DingTalk user identifier; the instruction confirmation information is the response returned by the target object through any channel such as WebSocket, DingTalk, or SMS, which contains the unique instruction identifier and the operator identity identifier.

[0088] For example, receive a JSON message returned by the target disposal object through a WebSocket channel, which contains the fields commandId (a unique instruction identifier corresponding to the business disposal instruction) and personId (a person identifier corresponding to the operator); write the message to a distributed cache system with the key command:receipt:{commandId} and set an expiration time of one hour; at the same time, record the instruction identifier and confirmation timestamp in the key command:confirmed.

[0089] Step S400: If an instruction confirmation message for a business handling instruction is received from the target object, a risk handling plan is retrieved from the preset risk case database based on the risk characteristics of the crane risk event and the unique instruction identifier of the business handling instruction, so as to create a risk event handling task and send it to the target object for execution.

[0090] Among them, instruction confirmation information refers to a deterministic response returned by the target disposal object through any delivery channel, carrying a unique instruction identifier for the business disposal instruction and an operator's identity identifier; the risk characteristics of a crane risk event refer to the equipment type, event type, risk level, wind speed value, load rate value, and boom length value carried by the event; the unique instruction identifier for the business disposal instruction refers to a string generated by a globally unique identifier generator; the preset risk case database refers to a document-based database used to store historical disposal plans; the risk disposal plan refers to a standardized disposal process formulated for a specific risk scenario, the content of which includes applicable conditions, a list of execution steps, and historical execution success rate indicators; and the risk event disposal task refers to a structured task object with state machine-driven capabilities constructed based on the content of the plan and the current event context.

[0091] As an example, in response to the instruction confirmation information returned by the target disposal object, the status of the business disposal instruction is updated; the risk characteristics of the crane risk event and the unique instruction identifier of the business disposal instruction are extracted; based on the extracted risk characteristics, a matching risk disposal plan is retrieved from the preset risk case database; the matching plans are sorted according to their historical execution success rates; the content of the plan contained in the top-ranked plan is obtained; based on the obtained plan content, the crane risk event, and the unique instruction identifier of the business disposal instruction, a risk event disposal task is constructed; the task is serialized into a JSON string, the message target subject is set to DISPOSAL_TASK (disposal task subject), and the message queue system producer interface is called to send it downstream.

[0092] In an optional embodiment, step S400 includes the following sub-steps: S401, respond to instruction confirmation information, update the status of the service processing instruction according to the received confirmation status.

[0093] Among them, the confirmation status refers to the system record formed after the target object accepts, rejects, or fails to respond within the timeout period to the business processing instruction; the status of the business processing instruction refers to the deterministic stage of the instruction in its life cycle, including pending, sent, confirmed, and executed.

[0094] For example, upon receiving a confirmation message, the system queries the distributed cache system for confirmation records with the key command:confirmed. If a record matching the unique instruction identifier is found, the status of the business processing instruction is updated to confirmed. If no record is found, the original status is maintained.

[0095] S402 is a unique instruction identifier for extracting risk characteristics and business handling instructions for crane risk events.

[0096] Among them, risk characteristics refer to the equipment type, event type, risk level, wind speed value, load rate value, and boom length value extracted from crane risk events; unique instruction identifier refers to the string field generated by the globally unique identifier generator in the business processing instruction object.

[0097] For example, extract the equipment type, event type, risk level, wind speed value, load rate value, and boom length value from the crane risk event object; and extract the value of the unique instruction identifier field from the business processing instruction object.

[0098] S403, based on the extracted risk characteristics, retrieve matching risk handling plans from the preset risk case database.

[0099] Among them, the preset risk case database refers to a document-based database that stores all risk response plans; the matched risk response plan refers to a plan whose applicable conditions match the extracted risk characteristics. The applicable conditions include equipment type, event type, load rate range, and wind speed range.

[0100] As an example, construct multi-dimensional query conditions, including the device type field being equal to the extracted device type, the event type field being equal to the extracted event type, and the minimum load rate being less than or equal to the extracted load rate and the maximum load rate being greater than or equal to the extracted load rate; initiate a query to the preset risk case database to obtain all risk handling plans that meet the conditions.

[0101] S404: Based on the historical success rate of each matched risk response plan, sort the risk response plans.

[0102] The historical execution success rate refers to the percentage of times the plan has been successfully executed and verified within a fixed time period in the past, out of the total number of executions. This metric is regularly updated by the distributed analysis engine from the time-series database.

[0103] As an example, for all the retrieved matching plans, read the value of their historical execution success rate field; sort them in descending order of this value; if multiple plans have the same success rate, sort them in ascending order of their creation time as a secondary sorting criterion.

[0104] S405: Obtain the content of the risk response plan ranked first. Based on the obtained plan content, the crane risk event, and the unique instruction identifier of the business response instruction, construct the risk event response task.

[0105] The content of the contingency plan refers to the set of standardized execution steps defined in the risk handling contingency plan. Each step includes a step number, step description, expected results, and required evidence type. The risk event handling task refers to a structured data object that includes a task identifier, risk event identifier, instruction identifier, equipment identifier, contingency plan identifier, initial state, and a list of contingency plan steps.

[0106] For example, the plan content field of the first-ranked plan is read. This field is an array containing multiple step objects. Each step object contains a step number, a step description, and an expected result field. A globally unique string is used as the task identifier, the event identifier of the crane risk event is used as the risk event identifier, the unique instruction identifier of the business handling instruction is used as the instruction identifier, the equipment identifier of the crane risk event is used as the equipment identifier, and the plan identifier of this plan is used as the plan identifier. With the initial state of pending dispatch, the above fields and the plan content array are encapsulated to generate a risk event handling task object.

[0107] For example, retrieve the top-ranked risk handling plan from the preset risk case database. This plan includes three handling steps: 1. Shorten the boom to within 30 meters, with the expected result being that the anemometer reading does not exceed 6 meters per second; 2. Suspend the current lifting operation, with the expected result being that the hook stops moving; 3. Upload on-site operation photos, with the expected result being that the evidence link is valid. Encapsulate the plan content, the crane risk event EVENT-e99a8c7e, and the unique instruction identifier CMD-7b3f7d7f of the business handling instruction to generate a risk event handling task. Its task identifier is TASK-8a9b0c1d, and its initial status is pending dispatch.

[0108] In one optional embodiment, after creating the risk event handling task, the method further includes: The state machine model is used to record the legal states of risk event handling tasks and the rules for transitions between states.

[0109] The state machine model refers to the formal description of a finite set of states and the rules for legal transitions between states for risk event handling tasks. A legal state is a state node that is allowed to exist throughout the entire lifecycle of the task and has a clear business meaning. The rules for transitions between states are the deterministic conditions that trigger state changes, including user operations, system timed tasks, and external event feedback. The types of states include pending assignment, assigned, accepted, processing, pending verification, completed, escalated, and canceled. Pending assignment means that the task has been created but not yet assigned to a specific person; assigned means that the task has been assigned to a specific person; accepted means that the person in charge has confirmed acceptance of the task; processing means that the person in charge has begun to execute the handling action; pending verification means that the handling action has been completed and is waiting for the on-site personnel to verify the effect; completed means that the verification has been passed and the risk has been confirmed to be resolved; escalated means that the handling has encountered obstacles and needs to be transferred to a higher level for handling; and canceled means that the risk has been misjudged or has been resolved through other means.

[0110] For example, when a risk event handling task is constructed, its status is initialized to "Pending Assignment." This status set is used for visualization on the front-end monitoring terminal, allowing operators to intuitively identify the task's stage. For instance, when a risk event handling task is created, its status is initialized to "Pending Assignment." When the risk event handling task is assigned to a target object, its status is updated to "Assigned." When the target object confirms the business handling instruction, its status is updated to "Accepted." When the target object initiates a step execution request, its status is updated to "Processing." When the target object submits the task execution result information, its status is updated to "Pending Verification." When verifying the authenticity of on-site handling based on evidence links in the task execution result information, its status is updated to "Completed." When the target object initiates an escalation request, or when the risk event handling task is not completed within a preset timeout period, its status is updated to "Escalated." Upon receiving a cancellation request from the target object, or when a business handling instruction is revoked, update the status of the risk event handling task to "cancelled".

[0111] In an optional embodiment, the crane risk event handling method further includes the following steps: In response to the user's selection of crane risk events on the monitoring terminal, the system queries standardized operating data generated after data cleaning and standardization within a preset time window, based on the crane equipment identifier associated with the risk event. Based on the load rate and wind speed values ​​included in the standardized operating data, the system generates visual data displaying load rate curves, wind speed curves, and corresponding reference lines. This visual data is then pushed to the monitoring terminal via a real-time communication channel.

[0112] Among them, the monitoring terminal user refers to the operator who monitors the crane's operation status through the web-based visualization system; the crane equipment identifier associated with a crane risk event refers to the unique identifier of the equipment carried in the event object; the time window refers to the time span used to query historical operation data, and its length is preset according to business needs; standardized operation data refers to structured data obtained after protocol parsing, geographic information enhancement, and multi-dimensional cleaning, including fields such as timestamp, load rate, wind speed, and GPS coordinates; the load rate curve refers to a trend line graph plotted with time as the horizontal axis and load rate value as the vertical axis; the wind speed curve refers to a trend line graph plotted with time as the horizontal axis and wind speed value as the vertical axis; the corresponding reference line refers to the horizontal marker line set according to the engineering safety threshold, including the load ratio overload line and the wind speed warning line; and the real-time communication channel refers to the WebSocket long connection established between the front end and the back end.

[0113] For example, when a monitoring terminal user clicks on a crane risk event with ID EVENT-e99a8c7e on the web interface, the system extracts the device identifier CRANE001 from the event and sends an HTTP request to the backend service at the path / api / data / history with parameters craneId=CRANE001&minutes=5. The backend queries the time-series database for standardized operating data of the device over the past five minutes, returning 300 records, each containing a timestamp, load rate, and wind speed. The frontend uses the ECharts chart library to load this data and configures a dual Y-axis chart: the left Y-axis displays the load rate (0.0–1.0), and the right Y-axis displays the wind speed (0–20 m / s). A red dashed line with y=0.9 is drawn on the left Y-axis as the load rate overload line, and an orange dashed line with y=6.0 is drawn on the right Y-axis as the wind speed warning line. The frontend pushes the rendered chart to the monitoring terminal user's browser interface in real time via an established WebSocket connection for display.

[0114] For example: EVENT-e99a8c7e (Risk Event Identifier): Used to uniquely identify a crane risk event; CRANE001 (Crane Equipment Identifier): Used to locate the specific monitored crane equipment; / api / data / history (Historical Data Query Interface Path): Used to request standardized operational data from the front end to the back end; craneId=CRANE001&minutes=5 (Query Parameters): Specifies the query equipment number and time window length; y=0.9 (Load Ratio Overload Line): Indicates that an overload warning is triggered when the load ratio reaches 0.9; y=6.0 (Wind Speed ​​Warning Line): Indicates that a wind speed warning is triggered when the wind speed reaches 6.0 meters per second; WebSocket (Real-time Communication Channel): Used to establish a long connection between the front end and the back end to ensure low-latency message push.

[0115] In an optional embodiment, the crane risk event handling method further includes the following steps: Receive risk level correction feedback or false alarm labeling feedback from the target entity regarding crane risk events. Based on the risk level correction feedback, update the mapping relationship between the total severity score range and the risk level in the preset grading rules. Based on the false alarm labeling feedback, update the alarm triggering conditions and / or timing association conditions in the preset alarm rules corresponding to the alarm events marked as false alarms.

[0116] Among them, risk level correction feedback information refers to the correction information submitted by the target handling object in response to the risk level output by the system, which includes at least the risk event identifier, the original risk level, the corrected risk level, and the reason for the correction; false alarm marking feedback information refers to the false alarm marking information submitted by the target handling object for the generated crane risk event, which includes at least the risk event identifier, the alarm event identifier or alarm code marked as a false alarm, and the reason for the false alarm; the difference between risk level correction feedback information and false alarm marking feedback information is that risk level correction feedback information is used to correct the determined risk level result, while false alarm marking feedback information is used to mark whether the alarm event involved in alarm identification and processing is a false alarm. The mapping relationship between the total severity score range and the risk level in the grading rules refers to the correspondence used to map the cumulative severity score of multiple discrete alarm events to severe risk, high risk, medium risk, or low risk.

[0117] Specifically, updating the mapping relationship between the total severity score range and the risk level in the grading rules refers to adjusting the preset score range divisions based on risk level correction feedback information. For example, if the preset grading rules are "total score ≥ 80 is severe risk, 80 > total score ≥ 60 is high risk, 60 > total score ≥ 30 is medium risk, and total score < 30 is low risk," and the target individual submits risk level correction feedback information multiple times with an original risk level of HIGH and a corrected risk level of MEDIUM, then the interval divisions of "80 > total score ≥ 60 is high risk" and "60 > total score ≥ 30 is medium risk" can be adjusted so that some total scores that originally fell into the high-risk range are adjusted to fall into the medium-risk range.

[0118] Alarm triggering conditions refer to the parameter conditions in the preset alarm rules used to determine whether an alarm is established; timing correlation conditions refer to the time window conditions and sequential occurrence conditions in the preset alarm rules used to determine the correlation between multiple alarm events. The timing correlation conditions correspond to the timing pattern matching in step S102. In step S102, timing pattern matching refers to the pattern defined by the complex event processing engine, where a starting event is followed by a subsequent event, and both occur within a specified time window; the preset correlation relationship means that two alarm events must have the same device identifier, and the difference between the occurrence time of the subsequent event and the occurrence time of the starting event does not exceed 300 seconds. Therefore, in this embodiment, updating the timing correlation conditions refers to adjusting the aforementioned time window conditions and sequential occurrence conditions used for timing pattern matching in step S102.

[0119] For example, if the crane risk event corresponding to the false alarm feedback information is matched within a five-minute time window with "excessive wind speed alarm" as the starting event and "overload rate alarm" as the subsequent event, and the target object marks the crane risk event as a false alarm, then the condition in step S102, "the difference between the occurrence time of the subsequent event and the occurrence time of the starting event does not exceed three hundred seconds", can be adjusted to a time range of less than three hundred seconds; or the time difference condition can be kept unchanged, but adjusted to match only when the same device identifier is used and the starting event is followed by a subsequent event.

[0120] For example, if the false alarm flag feedback information indicates that the "wind speed exceeding the standard alarm" itself is falsely triggered, the alarm triggering conditions corresponding to the alarm event can be adjusted; if the false alarm flag feedback information indicates that the false alarm is caused by the combination of "wind speed exceeding the standard alarm" and "overload rate alarm", the timing association conditions in step S102 can be adjusted.

[0121] For example, when a target individual clicks the "Level Correction" button on a risk event card through a web-based visualization system, the front-end system sends a request to the back-end service. This request includes the risk event identifier, the original risk level, the corrected risk level, and the reason for the correction. Upon receiving the risk level correction feedback, the back-end service writes this feedback into a pre-defined database and updates the mapping between the total severity score range and the risk level in the grading rules based on multiple accumulated risk level correction feedback messages within a pre-defined statistical period. Specifically, when accumulated risk level correction feedback indicates that a crane risk event with a total score between 60 and 79 has been repeatedly corrected to medium risk, the system can adjust the score range division of "80 > total score ≥ 60 is high risk".

[0122] For example, when the target user clicks the "False Alarm Mark" button through the web-based visualization system, the front-end system sends a request to the back-end service. This request includes a risk event identifier, the alarm event identifier or alarm code marked as a false alarm, and the reason for the false alarm. After receiving the false alarm marking feedback, the back-end service determines the preset alarm rule corresponding to the marked false alarm event and updates the alarm triggering conditions and / or timing correlation conditions within that preset alarm rule. Specifically, when the false alarm marking feedback indicates that the alarm code marked as a false alarm is wind_speed_exceed, and this false alarm corresponds to the matching result of "wind speed exceeding the standard alarm" and "overload rate alarm" within a five-minute time window in step S102, the system can adjust the time window condition in step S102 or the condition that the difference between the time of subsequent events and the time of the initial event does not exceed 300 seconds. When the false alarm marking feedback indicates that the false alarm originates directly from a certain alarm event itself, the system can adjust the alarm triggering conditions corresponding to that alarm event.

[0123] For example, when a target submits risk level correction feedback for EVENT-e99a8c7e with an initial risk level of HIGH and a revised risk level of MEDIUM, the system stores this feedback and adjusts the mapping relationship between the total severity score range and the risk level based on accumulated similar feedback within a preset statistical period. For instance, the system can adjust the range divisions for "80 > total score ≥ 60 is high risk" and "60 > total score ≥ 30 is medium risk".

[0124] When the target object submits false alarm labeling feedback information for the risk event EVENT-e99a8c7e, and the false alarm labeling feedback information indicates that the alarm code marked as a false alarm is wind_speed_exceed and the reason for the false alarm is "sensor calibration offset", the system updates the alarm triggering conditions and / or timing association conditions in the preset alarm rules corresponding to the wind speed exceeding the standard alarm. For example, the alarm triggering conditions corresponding to the "wind speed exceeding the standard alarm" can be adjusted; or the timing association conditions in step S102, where "wind speed exceeding the standard alarm" is the starting event, "overload rate alarm" is the subsequent event, and the difference between the occurrence time of the subsequent event and the occurrence time of the starting event does not exceed 300 seconds, can be adjusted.

[0125] In an optional embodiment, the crane risk event handling method of this application can also be demonstrated through the following specific scenario examples: At 12:20 PM one day, the tower crane CRANE001 at a construction site in Chaoyang District, Beijing, continuously reported sensor data: the wind speed rapidly increased from 8.2 m / s to 15.2 m / s, while the lifting load rate increased from 0.72 to 0.85. The Flink stream processing job accessed this data stream in real time. After protocol parsing, geographic information enhancement, and multi-dimensional cleaning, standardized data was obtained. Within a five-minute time window, a temporal correlation was detected between "excessive wind speed alarm" and "excessive load rate alarm" under the same device ID, and the match was successful. The severity scores of the two alarm events were accumulated, with one alarm event corresponding to the WARNING level and the other to the LOW level, resulting in a total score of 80 points. Based on preset classification rules, the risk level was determined to be severe risk, and a crane risk event was generated. Its event identifier was EVENT-e99a8c7e, the device identifier was CRANE001, the event type was high wind and high load operation risk, the risk level was high risk, and the triggered alarm list contained two records. After generating a structured crane risk event, the risk score can be calculated as approximately 0.72 based on the wind speed value of 15.2 and the load rate value of 0.85 extracted from multiple discrete alarm events, according to the preset calculation formula R=0.4×load rate+0.3×(wind speed value÷12.0). The risk score is then stored or displayed as a quantitative representation of the crane risk event.

[0126] The system then calls the instruction template library and matches the instruction template with the template identifier WIND_HIGH_LOAD; it fills in the wind speed of 15.2 m / s into {v}, the load ratio of 0.85 into {η}, and the boom length of 45 m into {L}, generating the instruction content "Current wind speed 15.2 m / s, load ratio 0.85, boom length 45 m, there is a risk of overturning. {action}"; based on the high risk level, the final recommended action is selected as PAUSE_OPERATION; combined with the equipment type being tower crane and the event type being high wind and high load operation risk, the target handling role list is determined to be driver, site_manager, and safety_supervisor; the instruction content, recommended action, role list, 60-second validity period, and globally unique string CMD-7b3f7d7f are encapsulated to generate the business handling instruction.

[0127] The system queries the permission database to obtain the online status of driver Zhang San (DingTalk ID: zhangsan) and site manager Li Si (DingTalk ID: lisi); it finds that Zhang San's online status is WEB_ONLINE, so it pushes the instruction to his monitoring terminal through the WebSocket channel; and simultaneously starts a 30-second countdown task. At 12:20:25, Zhang San clicked "Confirm." The system received the instruction confirmation information carrying CMD-7b3f7d7f and "zhangsan," immediately updated the instruction status to "Confirmed," and extracted the risk characteristics of EVENT-e99a8c7e and CMD-7b3f7d7f. A query was initiated into the document-based database, matching the contingency plan PLAN_WIND_01 (historical success rate 92%). Its three steps were read: "Step 1: Suspend the current hoisting operation," "Step 2: Shorten the boom to within 30 meters," and "Step 3: Check the anemometer reading." Using the globally unique string TASK-8a9b0c1d as the task identifier, EVENT-e99a8c7e as the risk event identifier, CMD-7b3f7d7f as the instruction identifier, CRANE001 as the device identifier, and PLAN_WIND_01 as the contingency plan identifier, with the initial state of "Pending Dispatch," a risk event handling task was encapsulated and sent to the downstream message queue.

[0128] The operator clicks on EVENT-e99a8c7e on the monitoring screen. The front-end system extracts CRANE001 and sends a request to the back-end: / api / data / history?craneId=CRANE001&minutes=5. The back-end returns standardized data for the past five minutes from the time-series database. The front-end uses ECharts to render a dual Y-axis trend chart: the left side displays the load ratio curve (marked with a red dashed line "overload line" at y=0.9), and the right side displays the wind speed curve (marked with an orange dashed line "warning line" at y=6.0), visually presenting the causes of the risk. Optionally, when displayed on the monitoring terminal, a risk score of 0.72 can also be simultaneously displayed as a quantitative representation of the crane risk event.

[0129] After Zhang San executes step 1, he submits the task execution result on the mobile device, and the system updates the status to pending verification. The staff on the platform verify that the boom length in the uploaded on-site video is 28 meters and the anemometer shows 4.8 meters per second. They then click "Verification Passed" and the status updates to completed.

[0130] After the handling is completed, Zhang San clicks "Level Correction" on the interface, providing feedback on the original risk level HIGH, the assessed risk level MEDIUM, and the reason "No hoisting operation on site, although the wind speed is high, there is no actual risk." The system stores this risk level correction feedback information in a document-based database and, based on multiple risk level correction feedback messages accumulated within a preset statistical period, updates the mapping relationship between the total severity score range and the risk level in the preset grading rules. Simultaneously, Zhang San clicks "False Alarm Mark," providing feedback that the alarm code marked as a false alarm is wind_speed_exceed (wind speed exceeding the limit alarm), and the reason for the false alarm is "sensor calibration offset." Based on this false alarm mark feedback information, the system updates the alarm triggering conditions and / or timing correlation conditions in the alarm rules corresponding to the wind speed exceeding the limit alarm. For example, adjusting the wind speed threshold corresponding to the wind speed exceeding the limit alarm, and / or adjusting the five-minute time window used in step S102 when matching the timing pattern of the wind speed exceeding the limit alarm and the overload rate alarm, or ensuring that the difference between the time of subsequent events and the time of the initial event does not exceed 300 seconds.

[0131] Figure 4 A schematic diagram of a crane risk event handling device according to an embodiment of this application is shown. Exemplarily, the device 100 includes: The alarm identification and processing module 110 is used to acquire the original operating data of multiple cranes, and to perform alarm identification processing on the original operating data of each crane to generate corresponding crane risk events. The business processing module 120 is used to call a preset instruction template according to the risk characteristics of the crane risk event, and fill the parameters in the crane risk event that match the content placeholder of the instruction template into the content placeholder of the instruction template to generate a business processing instruction carrying a unique instruction identifier. The instruction distribution module 130 is used to distribute the business processing instruction to the corresponding target processing object through a preset delivery channel; The contingency plan retrieval module 140 is used to, if it receives instruction confirmation information from the target disposal object regarding the business disposal instruction, retrieve risk disposal contingency plans from a preset risk case database based on the risk characteristics of the crane risk event and the unique instruction identifier of the business disposal instruction, so as to create a risk event disposal task and send it to the target disposal object for execution.

[0132] It is understood that the apparatus of this embodiment corresponds to the method of the above embodiments, and the options in the above embodiments are also applicable to this embodiment, so they will not be described again here.

[0133] This application also provides a terminal device, exemplary of which includes a processor and a memory, wherein the memory stores a computer program, and the processor executes the computer program to enable the terminal device to perform the functions of the various modules in the above-described method or apparatus.

[0134] The processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, including at least one of a Central Processing Unit (CPU), Graphics Processing Unit (GPU), Network Processor (NP), Digital Signal Processor (DSP), Application-Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The general-purpose processor can be a microprocessor or any conventional processor, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application.

[0135] The memory can be, but is not limited to, Random Access Memory (RAM), Read Only Memory (ROM), Programmable Read-Only Memory (PROM), Erasable Programmable Read-Only Memory (EPROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), etc. The memory is used to store computer programs, and the processor can execute the computer programs accordingly after receiving execution instructions.

[0136] This application also provides a computer-readable storage medium for storing the computer program used in the aforementioned terminal device. For example, the computer-readable storage medium may include, but is not limited to, various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0137] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that, in alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0138] In addition, the functional modules or units in the various embodiments of this application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0139] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a 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 to cause a computer device (which may be a smartphone, personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0140] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A crane risk event handling method, characterized by, The method includes: The system acquires raw operating data from multiple cranes, performs alarm identification processing on the raw operating data of each crane, and generates corresponding crane risk events. Based on the risk characteristics of the crane risk event, a preset instruction template is invoked, and the parameters in the crane risk event that match the content placeholder of the instruction template are filled into the content placeholder of the instruction template to generate a business processing instruction carrying a unique instruction identifier. The business processing instructions are distributed to the corresponding target objects through preset delivery channels; If the target object receives an instruction confirmation message for the business handling instruction, then based on the risk characteristics of the crane risk event and the unique instruction identifier of the business handling instruction, a risk handling plan is retrieved from the preset risk case database to create a risk event handling task and send it to the target object for execution. The process of acquiring raw operating data from multiple cranes and performing alarm identification processing on the raw operating data of each crane to generate corresponding crane risk events includes: The system acquires raw operating data reported by each crane device through a preset protocol, and performs protocol parsing, external geographic information enhancement, and multi-dimensional data cleaning on the raw operating data to obtain standardized data. Based on preset alarm rules, the standardized data is subjected to time-series pattern matching to identify multiple discrete alarm events that satisfy preset correlation relationships; Based on the risk characteristics of the multiple discrete alarm events, the severity scores of each alarm event are extracted and accumulated. The accumulated results are then mapped according to a preset grading rule to determine the corresponding risk level. The multiple discrete alarm events and their corresponding risk levels are aggregated to form a corresponding structured crane risk event; The process involves, based on the risk characteristics of the crane risk event, invoking a preset instruction template and filling the content placeholders of the instruction template with parameters from the crane risk event that match those in the instruction template, to generate a business processing instruction carrying a unique instruction identifier, including: Based on the equipment type and event type of the crane risk event, a target instruction template is matched from a preset instruction template library; The parameters in the crane risk event that match the content placeholder of the target instruction template are filled into the content placeholder of the target instruction template to generate the instruction content; Based on the risk level of the crane risk event, the final recommended action is determined from among the multiple recommended actions associated with the target instruction template; Based on the equipment type and event type of the crane risk event, a list of target handling roles is determined, and the instruction content, the final recommended action, the list of target handling roles, and the instruction validity period are encapsulated to generate a business handling instruction carrying a unique instruction identifier.

2. Crane risk event handling method according to claim 1, characterized in that, The step of distributing the business processing instructions to the corresponding target processing objects through preset delivery channels includes: Based on the list of handling roles with crane management authority specified in the business handling instruction and the equipment identifier of the crane risk event, query the target handling object; Based on the real-time online status of the target object, the optimal delivery channel is selected from the preset multi-level communication channels; The business processing instruction is distributed to the target processing object through the optimal delivery channel, and a timed task associated with the business processing instruction is triggered. If no instruction confirmation information carrying the unique instruction identifier and operator identity identifier of the business processing instruction is received within the preset time, the backup delivery channel is triggered to resend the instruction. Receive instruction confirmation information returned by the target object, which carries a unique instruction identifier and an operator identity identifier for the business processing instruction.

3. Crane risk event handling method according to claim 1, characterized in that, The process of creating a risk event handling task by retrieving a risk handling plan from a pre-set risk case database based on the risk characteristics of the crane risk event and the unique instruction identifier of the business handling instruction includes: In response to the instruction confirmation information, update the status of the service processing instruction according to the received confirmation status; Extract the risk characteristics of the crane risk event and the unique instruction identifier of the business handling instruction; Based on the extracted risk characteristics, a matching risk response plan is retrieved from a pre-set risk case database; Based on the historical success rate of each of the risk response plans, the risk response plans are sorted. Obtain the contents of the risk response plan that is ranked first; Based on the obtained contingency plan content, the crane risk event, and the unique instruction identifier of the business handling instruction, a risk event handling task is constructed.

4. The crane risk event handling method according to claim 1, characterized in that, After creating the risk event handling task, it also includes: The legal states and state transition rules of the risk event handling task are recorded using a state machine model. The types of states include pending dispatch, assigned, accepted, processing, pending verification, completed, upgraded, and canceled. When the risk event handling task is created, its status is initialized to pending dispatch; When the risk event handling task is assigned to the target object, the status of the risk event handling task is updated to "assigned". Upon receiving confirmation information from the target object regarding the business handling instruction, the status of the risk event handling task is updated to "accepted". Upon receiving a step execution request initiated by the target object, the status of the risk event handling task is updated to "processing". Upon receiving the task execution result information submitted by the target object, the status of the risk event handling task is updated to pending verification; When verifying the authenticity of the on-site handling by using evidence links based on the task execution result information, the status of the risk event handling task is updated to completed; Upon receiving an upgrade request initiated by the target object, or when the risk event handling task is not completed within a preset timeout period, the status of the risk event handling task is updated to upgraded. Upon receiving a cancellation request initiated by the target object, or when the business processing instruction is revoked, the status of the risk event processing task is updated to "cancelled".

5. The crane risk event handling method according to claim 1, characterized in that, The method further includes: In response to the selection operation of the monitoring terminal user on the crane risk event, based on the crane equipment identifier associated with the crane risk event, query the standardized operation data generated by data cleaning and standardization within a preset time window; Based on the load rate and wind speed values ​​contained in the standardized operating data, a visual display data carrying load rate curves, wind speed curves and corresponding reference lines is generated. The visualized data is pushed to the monitoring terminal through a real-time communication channel.

6. The crane risk event handling method according to claim 1, characterized in that, The method further includes: Receive risk level correction feedback information or false alarm labeling feedback information submitted by the target handling object in response to the crane risk event; wherein, the risk level correction feedback information includes at least the original risk level and the corrected risk level, and the false alarm labeling feedback information includes at least the alarm event identifier or alarm code marked as a false alarm and the reason for the false alarm; Based on the risk level correction feedback information, update the mapping relationship between the total severity score range and the risk level in the preset grading rules; Based on the false alarm label feedback information, update the alarm triggering conditions and / or timing association conditions in the preset alarm rules corresponding to the alarm events marked as false alarms.

7. A terminal device, characterized in that, The terminal device includes a processor and a memory, the memory storing a computer program, and the processor executing the computer program to implement the crane risk event handling method according to any one of claims 1-6.

8. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed on a processor, implements the crane risk event handling method according to any one of claims 1-6.

Images