Hydropower station intelligent desktop drilling method, system, equipment and medium based on three-dimensional simulation technology

Abstract

The invention discloses a hydropower station intelligent desktop drilling method, system and equipment based on a three-dimensional simulation technology and a medium, and belongs to the technical field of hydropower station desktop drilling. Connecting and constructing an event chain according to a time sequence; establishing a three-dimensional virtual scene according to equipment and places related to each event in the event chain, and adding a virtual special effect to generate an emergency drilling scene; setting a standard emergency measure corresponding to each event in the event chain according to the plan disposal process; displaying the emergency drilling scene to the participants, and obtaining countermeasures proposed by the participants for the current event; and generating a drill evaluation report. According to the invention, systematic expression of the accident evolution process is realized. The adjacent events are associated with the processing result through the current event state.

Description

Technical Field

[0001] This invention relates to the field of tabletop exercise technology for hydropower stations, specifically to a method, system, equipment, and medium for intelligent tabletop exercise of hydropower stations based on three-dimensional simulation technology. Background Technology

[0002] Emergency drills are a crucial component of emergency management, playing a vital role in testing contingency plans, improving preparedness, training personnel, and disseminating knowledge. Hydropower stations, as a high-risk industry prone to frequent emergencies, require strengthened emergency drills to enhance their ability to prevent and respond to such incidents. This is essential for economic and social development and the safety of people's lives and property. Flooding of the powerhouse is a significant safety hazard in hydropower station operation; once it occurs, it can lead to equipment damage, casualties, and substantial economic losses.

[0003] Traditional tabletop exercises for hydropower stations primarily employ two methods: conference discussion and video-assisted methods. Conference discussion involves describing the emergency situation verbally, with participants in different roles proposing different responses. Video-assisted methods involve a director setting up the emergency scenario through verbal descriptions and pre-recorded videos, with participants proposing responses. Conference discussion, relying solely on verbal descriptions, lacks the opportunity for observation or perception of the accident scenario, making it difficult to accurately assess the consequences and the effectiveness of responses, potentially leading to insufficient understanding of the accident. While video-assisted methods utilize video aids to facilitate observation and assessment, the lack of hands-on participation results in a lower level of engagement and experience for participants. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention provides a method, system, equipment and medium for intelligent desktop simulation of hydropower stations based on three-dimensional simulation technology.

[0005] Therefore, the technical problem solved by this invention is: how to solve the technical problems of existing hydropower station tabletop exercises, such as insufficient intuitiveness of accident scenarios, poor participation and experience of participants, strong subjectivity in exercise evaluation, difficulty in quantitative analysis and evaluation, difficulty in recording data during the exercise process, and low evaluation efficiency.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a method for intelligent desktop simulation of hydropower stations based on three-dimensional simulation technology, comprising, Collect historical cases of factory flooding accidents, extract the causes, development process and emergency response results from these cases, and build an accident case database. Select a target accident case from the accident case database, identify the initial triggering event, state change event, and emergency response completion event in the target accident case, and connect them in chronological order to construct an event chain. Adjacent events in the event chain are associated through the current event state and the response result. A three-dimensional virtual scene is created based on the equipment and locations involved in each event in the event chain. Virtual effects are added to the three-dimensional virtual scene based on the equipment status information of each event to generate an emergency drill scenario. Standard emergency measures are set for each event in the event chain according to the contingency plan's handling procedures; Show the participants the emergency drill scenario and obtain their proposed response measures for the current event; Extract keywords and key sentences from response measures and standard emergency measures, calculate keyword matching degree and key sentence semantic similarity, and calculate text similarity based on keyword matching degree and key sentence semantic similarity; When the text similarity is greater than a preset threshold, the response measures are deemed effective. Based on the response measures, the next event in the event chain is determined and the exercise continues. When the text similarity is less than or equal to a preset threshold, the countermeasures are deemed invalid, the deviation is recorded, and an exercise evaluation report is generated.

[0007] As a preferred embodiment of the intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology described in this invention, the construction of the accident case library includes classifying accident causes according to the triggering mechanism of the accident causes; Establish a correspondence between accident causes and equipment types; The accident development process is time-series labeled, and the time information of each state node during the accident development process is labeled; Establish the relationship between state nodes and their changes over time; Identify the emergency response measures and their effects in the emergency response outcome; Establish the correlation between emergency response measures and their effectiveness; An accident case library is constructed based on the correspondence between accident causes and equipment types, the relationship between status nodes and changes over time, and the correlation between emergency response measures and their effects.

[0008] As a preferred embodiment of the intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology described in this invention, the step of connecting and constructing the event chain in chronological order includes selecting target accident cases from an accident case library; Identify the time points in the accident development process of the target accident case where the state nodes change as event segmentation points; The accident development process is divided into multiple event segments at the event inflection point; Extract the time and status information of each event segment; Identify the earliest event segment based on time information as the initial trigger event; Identify event segments where the state changes based on the state information as state change events; Identify event segments that indicate the completion of emergency response as emergency response completion events; Connect the initial triggering event, the state change event, and the emergency response completion event according to the time information to construct an event chain.

[0009] As a preferred embodiment of the intelligent desktop exercise method for hydropower stations based on three-dimensional simulation technology described in this invention, the step of setting standard emergency measures corresponding to each event in the event chain according to the emergency response process includes obtaining the text of the emergency response plan for flooded powerhouses. Semantic analysis is performed on emergency response plan texts to identify statements describing emergency response procedures. Extract emergency response measures from the handling process statements; Based on the equipment and operations involved in the emergency response measures, determine the event corresponding to the emergency response measures; By associating emergency response measures with corresponding events, standard emergency measures are formed for each event in the event chain.

[0010] As a preferred embodiment of the intelligent desktop exercise method for hydropower stations based on three-dimensional simulation technology described in this invention, the calculation of keyword matching degree and key sentence semantic similarity includes performing word segmentation on the response measures and standard emergency measures to obtain a word set; Extract words from the word set that represent emergency response actions, emergency response targets, and emergency response departments as keywords; Syntactic analysis was performed on response measures and standard emergency measures to identify sentences containing keywords as key sentences; Calculate the number of matches between the keywords of the response measures and the keywords of the standard emergency measures, and calculate the keyword matching degree based on the ratio of the number of matches to the total number of keywords; The key sentences of the response measures and the key sentences of the standard emergency measures are converted into sentence vectors. The cosine distance between the sentence vectors is calculated, and the semantic similarity of the key sentences is calculated based on the cosine distance. The text similarity is calculated by weighted summation of keyword matching degree and key sentence semantic similarity.

[0011] The beneficial effects of this preferred technical solution are as follows: By segmenting response measures and standard emergency measures, words representing emergency response actions, emergency response targets, and emergency response departments are extracted as keywords, clarifying the basic elements of emergency response and avoiding the inefficient method of comparing the entire text word by word. Calculating the keyword matching degree reflects the completeness of the emergency response elements. Through syntactic analysis, key sentences containing keywords are identified, converted into sentence vectors, and their semantic similarity is obtained by calculating cosine distance. This achieves semantic understanding of the emergency response logic, unaffected by differences in textual expression.

[0012] As a preferred embodiment of the intelligent desktop exercise method for hydropower stations based on three-dimensional simulation technology described in this invention, the step of recording deviation content and generating an exercise evaluation report includes comparing text similarity with a preset threshold. When the text similarity is greater than a preset threshold, the response measures are deemed effective, and the outcome of the current event is identified based on the handling operations contained in the response measures. Based on the outcome of the current event and the relationship between adjacent events in the event chain through their current status and outcome, determine the next event in the event chain for the current event; Treat the next event as the current event, return to the steps of showing the emergency drill scenario to the participants, and continue the drill; When the text similarity is less than or equal to a preset threshold, the response measures are deemed ineffective. By comparing the response measures with the standard emergency response measures, the missing and incorrect emergency response measures in the response measures are extracted to form the deviation content; Record the response measures, text similarity and deviations for each event during the exercise, and generate an exercise evaluation report.

[0013] As a preferred embodiment of the intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology described in this invention, the establishment of the three-dimensional virtual scene includes identifying the equipment types and locations involved in each event in the event chain; Create a 3D model of the equipment based on its type, and a 3D model of the factory based on its location. Place the 3D model of the equipment into the 3D model of the factory according to its location to form a 3D virtual scene. The 3D models of devices in the 3D virtual scene are classified according to their level of detail, and different display distance ranges are set for 3D models of devices with different levels of detail. The 3D virtual scene is baked with light and shadow, and the shadows and reflections formed by light shining on the surface of the 3D model of the equipment and the 3D model of the factory are recorded as textures. Retrieve device status information for the current event in the event chain; The type of virtual effect is determined based on the equipment status information. When the equipment status information indicates that the equipment is leaking, a water flow effect is generated; when the equipment status information indicates that the equipment is vibrating, a vibration effect is generated; when the equipment status information indicates that the equipment is broken, a breakage effect is generated. Add virtual effects to the locations of devices involved in the current event in a 3D virtual scene to generate an emergency drill scenario.

[0014] The beneficial effects of this preferred technical solution are as follows: By identifying the equipment types and locations involved in each event in the event chain, 3D models of the equipment and the plant are established, forming a 3D virtual scene, thus realizing the 3D visualization of the hydropower station accident site. Employing a level-accuracy technology, different display distance ranges are set for 3D equipment models of varying accuracy, automatically switching model precision based on viewing distance. This reduces the rendering burden and improves the smoothness of scene display while ensuring visual quality. Furthermore, by using light and shadow baking processing to record shadows and reflections formed by light as textures, the amount of real-time lighting calculations is reduced, improving rendering efficiency.

[0015] This invention provides an intelligent desktop simulation system for hydropower stations based on three-dimensional simulation technology.

[0016] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a smart desktop simulation system for hydropower stations based on three-dimensional simulation technology, comprising: an accident case library construction module, used to collect historical flooding accident cases, extract the accident causes, accident development process and emergency response results from the accident cases, and construct an accident case library; The event chain construction module is used to select target accident cases from the accident case library, identify the initial triggering event, state change event and emergency response completion event in the target accident case, and connect them in chronological order to construct an event chain. Adjacent events in the event chain are associated through the current event state and the response result. The virtual scene generation module is used to create a 3D virtual scene based on the equipment and locations involved in each event in the event chain, and to add virtual effects to the 3D virtual scene based on the equipment status information of each event to generate an emergency drill scene. The standard measures setting module is used to set standard emergency measures for each event in the event chain according to the contingency plan's handling procedures. The response measures acquisition module is used to display emergency drill scenarios to the participants and obtain the response measures proposed by the participants in response to the current event. The text similarity calculation module is used to extract keywords and key sentences from response measures and standard emergency measures, calculate keyword matching degree and key sentence semantic similarity, and calculate text similarity based on keyword matching degree and key sentence semantic similarity; The exercise evaluation module is used to determine the effectiveness of the response measures when the text similarity is greater than a preset threshold, and to determine the next event in the event chain based on the response measures and continue the exercise. When the text similarity is less than or equal to a preset threshold, the countermeasures are deemed invalid, the deviation is recorded, and an exercise evaluation report is generated.

[0017] The present invention provides a computer device, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology.

[0018] The present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology.

[0019] The beneficial effects of this invention are as follows: By constructing an emergency accident scenario based on event chains, the continuous accident development process is divided into discrete event segments. Initial triggering events, state change events, and emergency response completion events are identified, and these are connected in chronological order to construct an event chain, thus achieving a systematic expression of the accident evolution process. Adjacent events are associated through their current event state and response outcome.

[0020] By creating a 3D virtual scene and adding virtual effects based on equipment status information to generate an emergency drill scenario, an intuitive display of the accident scene was achieved. The scene rendering effect was optimized using tiered detail and light and shadow baking techniques. Corresponding virtual effects were generated based on different equipment states such as water leakage, vibration, and cracking, allowing participants to intuitively perceive the accident scene and enhancing their sense of participation and experience during the drill.

[0021] By extracting keywords and key sentences from response measures and standard emergency measures, calculating keyword matching degree and key sentence semantic similarity, and comprehensively calculating text similarity, the objectivity and automation of exercise evaluation are achieved. Keyword matching degree reflects the completeness of emergency response actions, targets, and departments, while key sentence semantic similarity reflects the accuracy of emergency response logic. This two-level evaluation method reduces the subjectivity of manual evaluation and improves the accuracy of the evaluation.

[0022] By judging the effectiveness of response measures based on text similarity, and determining the next event in the event chain based on the outcome when the response measures are effective, dynamic tabletop exercises are achieved. The exercise process dynamically evolves based on the response measures of the participants, with different response measures leading to different event evolution paths. Compared with static pre-set scenario exercises, dynamic exercises can test the emergency response capabilities of participants in complex and ever-changing situations. Attached Figure Description

[0023] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 The above is a flowchart of an intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology, provided as an embodiment of the present invention.

[0025] Figure 2 This is a virtual reality image of a hydropower station, provided as an embodiment of the present invention, for a smart desktop simulation method for hydropower stations based on three-dimensional simulation technology.

[0026] Figure 3 This is a schematic diagram of an emergency case event chain for an intelligent desktop exercise method for hydropower stations based on three-dimensional simulation technology, provided as an embodiment of the present invention. Detailed Implementation

[0027] To make the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0028] Example 1, referring to Figure 1 This is one embodiment of the present invention, which provides a smart desktop simulation method for hydropower stations based on three-dimensional simulation technology, including: Step 1: Collect historical cases of factory flooding accidents, extract the causes, development process, and emergency response results from these cases, and build an accident case database; Step 2: Select a target accident case from the accident case database, identify the initial triggering event, state change event, and emergency response completion event in the target accident case, and connect them in chronological order to construct an event chain. Adjacent events in the event chain are associated through the current event state and the response result. Step 3: Create a 3D virtual scene based on the equipment and locations involved in each event in the event chain, and add virtual effects to the 3D virtual scene based on the equipment status information of each event to generate an emergency drill scenario; Step 4: Set standard emergency measures for each event in the event chain according to the contingency plan's handling procedures; Step 5: Show the participants the emergency drill scenario and obtain their proposed response measures for the current event; Step 6: Extract keywords and key sentences from the response measures and standard emergency measures, calculate keyword matching degree and key sentence semantic similarity, and calculate text similarity based on keyword matching degree and key sentence semantic similarity; when the text similarity is greater than the preset threshold, the response measures are deemed effective, the next event in the event chain is determined based on the response measures, and the exercise continues; when the text similarity is less than or equal to the preset threshold, the response measures are deemed ineffective, the deviation is recorded, and an exercise evaluation report is generated.

[0029] Traditional tabletop exercises for hydropower stations rely on meetings or video presentations, where participants learn about the accident through verbal descriptions or video demonstrations. Meetings, however, rely solely on verbal descriptions, making it difficult for participants to intuitively perceive the situation and accurately assess crucial information such as equipment damage and water spread. This can lead to inaccurate emergency response plans. While video presentations provide visual information, the fixed content fails to dynamically adjust the accident's evolution based on participants' responses, resulting in a lack of participation. Evaluation depends on subjective judgment, leading to differing assessments of the same responses and difficulties in standardizing evaluation criteria. Manual data recording is essential, easily overlooking key information and hindering quantitative analysis of the exercise's effectiveness. This invention addresses the problems of insufficient intuitiveness, poor participation, and subjective evaluation in traditional exercises by constructing an event chain-based accident scenario and creating a 3D virtual scene to simulate a real accident. It dynamically predicts the event's development based on participants' responses and objectively evaluates the effectiveness of responses through text similarity calculations.

[0030] The method provided by this invention constructs an accident case library in step 1, providing a foundation of real accident data for drills. Step 2 establishes an event chain by identifying initial triggering events, state change events, and emergency response completion events, discretizing the continuous accident development process into an ordered sequence of events. Adjacent events are linked through their current state and response results, establishing a logical framework for dynamic drill simulations. Step 3 creates a 3D virtual scene and adds virtual effects, allowing participants to intuitively observe equipment state changes and the accident evolution process, improving their perception of accident scenarios. Step 4 sets standard emergency measures, providing objective criteria for subsequent evaluation. Step 5 presents the emergency drill scenario to participants and provides them with response measures, enabling interaction between participants and the drill system. Step 6 calculates text similarity by extracting keywords and key sentences, employing a two-level evaluation method to objectively judge the effectiveness of response measures. When the response measures are effective, the next event is determined based on the response results to continue the drill, achieving dynamic simulation. When the response measures are ineffective, deviations are recorded to generate an evaluation report, providing a basis for improving participants' capabilities. These steps form a complete drill process, realizing the intelligence and objectivity of hydropower station emergency drills.

[0031] Example 2, an embodiment of the present invention, provides a method for intelligent desktop simulation of hydropower stations based on three-dimensional simulation technology, based on the previous embodiment, including: Step 1: Collect historical cases of factory flooding accidents, extract the causes, development process, and emergency response results from these cases, and construct an accident case database, including the following steps A1-A7: A1: Classify the causes of accidents according to their triggering mechanisms; A2: Establish the correspondence between accident causes and equipment types; A3: Perform time-series annotation on the accident development process, annotating the time information of each state node during the accident development process; A4: Establish the relationship between state nodes and their changes over time; A5: Identify the emergency response measures and their effects in the emergency response outcome; A6: Establish the correlation between emergency response measures and their effectiveness; A7: Construct an accident case library based on the correspondence between accident causes and equipment types, the relationship between status nodes and changes over time, and the correlation between emergency response measures and their effects.

[0032] In this embodiment of the application, in step A7, the accident case library is constructed by: Each historical flooded factory accident case is assigned a unique case identifier. The correspondence between accident causes and equipment types is used as the case's equipment attribute, recording the equipment types involved and their corresponding accident cause categories. The relationship between state nodes and their changes over time is used as the case's temporal attribute, storing the time information and state descriptions of each state node in the accident case in chronological order. The correlation between emergency response measures and their effects is used as the case's handling attribute, storing the various emergency response measures taken in the accident case and their corresponding effects. Based on the case identifier, the equipment attribute, temporal attribute, and handling attribute are linked to form a complete accident case record. All accident case records are stored in an accident case database.

[0033] In an optional implementation, in step A7, the accident case database can be constructed by: establishing spatial attributes for the cases based on the spatial structure of the hydropower station powerhouse; recording the location of the accident, including the turbine floor, generator floor, spiral casing floor, tailrace gallery floor, and other powerhouse floor locations; recording the specific location of the equipment involved in the powerhouse in the equipment attributes based on the correspondence between the accident cause and equipment type; recording the time points and water level changes as the water level gradually rises from the lowest floor to submerge each floor in the time-series attributes based on the relationship between state nodes and time changes; and recording the emergency measures taken for different submersion levels and the water level control effects in the treatment attributes based on the correlation between emergency response measures and their effects. Establishing the association between spatial attributes and equipment attributes, time-series attributes, and treatment attributes forms an accident case record containing spatial information.

[0034] In another optional implementation, in step A7, the construction of the accident case library can also be achieved by: establishing operating condition attributes for cases based on the hydropower station's operating conditions. These attributes record the unit's operating status at the time of the accident, including normal power generation, unit maintenance, and standby operation. The operating condition attributes record the pressure status of the water intake system, the open / closed status of the tailrace system, and the operating status of the drainage system under that operating condition. The operating condition attributes are associated with the correspondence between accident causes and equipment types, recording the types of accidents prone to occur under different operating conditions. The operating condition attributes are associated with the correlation between emergency response measures and their effectiveness, recording the focus of emergency response and differences in effectiveness under different operating conditions. Finally, the relationships between operating condition attributes and equipment attributes, time-series attributes, and response attributes are established to form an accident case record containing operating condition information.

[0035] Step 2: Select a target accident case from the accident case database, identify the initial triggering event, state change event, and emergency response completion event in the target accident case, and connect them in chronological order to construct an event chain, including the following steps B1-B8: B1: Select a target accident case from the accident case database; B2: Identify the time points in the development process of the target accident case where the state nodes change as event segmentation points; B3: Divide the accident development process into multiple event segments at the event inflection point; B4: Extract the time and status information of each event segment; B5: Identify the earliest event segment based on time information as the initial trigger event; B6: Identify event segments where the state changes based on state information as state change events; B7: Identify event segments that indicate the completion of emergency response as emergency response completion events; B8: Connect the initial triggering event, the state change event, and the emergency response completion event according to the time information to build an event chain.

[0036] In this embodiment, step B8 involves constructing an event chain by: extracting the state information of the initial triggering event as the current event state, and extracting the emergency response measures involved in the initial triggering event as the response result. Based on the current event state and the response result, finding event fragments in the state change events whose state information is associated with the current event state as the next event. Establishing a connection between the initial triggering event and the next event, and recording the association information of the current event state and the response result as the connection relationship. Taking the next event as the new current event, extracting its state information and response result, and repeating the above search and connection process until it is connected to the emergency response completion event. Arranging the initial triggering event, each state change event, and the emergency response completion event in the connection order to form an event chain.

[0037] In an optional implementation, in step B8, the event chain can be constructed by: constructing the event chain according to the typical evolution path of a hydropower station flooding accident. The initial triggering event is a equipment leakage event, recording the type of leaking equipment and the location of the leakage. The first type of state change event is a water level rise event, calculating the water level rise rate based on the leakage rate and the drainage capacity of the sump, and marking the time nodes when the water level rises from the bottom of the turbine floor to each key elevation. The second type of state change event is an equipment submersion event, determining the submersion status of equipment on each floor based on the water level height, and marking the time nodes when key equipment such as turbines, generators, and transformers begin to be submerged. The third type of state change event is a function loss event, determining the time of function loss of each system based on the equipment submersion time, and marking the time nodes when key functions such as power generation loss, plant power loss, and communication system failure are lost. The emergency response completion event is either a successful water level control event or a safe personnel evacuation event. Connecting the above events in chronological order constructs the event chain of the flooding accident.

[0038] In another optional implementation, in step B8, the event chain can also be constructed by: constructing a branch event chain based on the emergency response level of the hydropower station. After the initial trigger event, the emergency response level is determined based on the amount of leakage and the rate of water level rise. When the leakage is less than the drainage capacity of the sump, it is marked as a general accident level, and the event chain includes four events: leakage detection, starting the drainage pump, leakage cessation, and emergency response completion. When the leakage is greater than the drainage capacity of the sump and the water level rises below the generator level, it is marked as a major accident level, and the event chain adds events such as shutdown and power outage, closing the inlet valve, and starting the backup drainage pump. When the water level rises above the generator level, it is marked as a serious accident level, and the event chain adds events such as plant-wide power outage, emergency evacuation of personnel, and requesting external support. Based on the handling results corresponding to each event, it is determined whether the water level has been successfully controlled or whether the emergency response level needs to be upgraded, and an event chain containing branch judgments is constructed.

[0039] Step 3: Establish a 3D virtual scene based on the equipment and locations involved in each event in the event chain. Add virtual effects to the 3D virtual scene based on the equipment status information of each event to generate an emergency drill scenario, including the following steps C1-C7: C1: Identify the device type and location involved in each event in the event chain; C2: Create a 3D model of the equipment based on the equipment type, and a 3D model of the factory based on the location. Place the 3D model of the equipment in the 3D model of the factory according to the location to form a 3D virtual scene. C3: Classify the 3D models of devices in the 3D virtual scene according to their level of detail, and set different display distance ranges for 3D models of devices with different levels of detail; C4: Performs light and shadow baking on the 3D virtual scene, recording the shadows and reflections formed by light shining on the surface of the 3D equipment model and the 3D factory model as textures; C5: Retrieves the device status information of the current event in the event chain; C6: Determine the type of virtual effect based on the device status information. When the device status information indicates that the device is leaking, generate a water flow effect; when the device status information indicates that the device is vibrating, generate a vibration effect; when the device status information indicates that the device is broken, generate a breakage effect. C7: Add virtual effects to the locations of devices involved in the current event in a 3D virtual scene to generate an emergency drill scenario.

[0040] In this embodiment, step 3, establishing a three-dimensional virtual scene involves: creating a layered three-dimensional model of the hydropower station powerhouse, including a turbine layer, a generator layer, an intermediate layer, and a hoisting layer, based on the powerhouse structure. The spatial relationships between each layer are established according to their actual elevation differences. Three-dimensional models of the turbine, spiral casing, and tailrace are created on the turbine layer; on the generator layer, models of the generator, thrust bearing, and control cabinet are created; on the intermediate layer, models of the transformer and switchgear are created; and on the hoisting layer, a trolley model is created. A three-dimensional model of the water intake system is created, including the water intake pipe, ball valves, quick-release gates, and bypass pipes. A three-dimensional model of the drainage system is created, including the collection well, drainage pumps, and drainage pipes. Based on the actual installation location of the equipment in the powerhouse, the three-dimensional models of each piece of equipment are placed in their corresponding positions on their respective layers. A three-dimensional model of the external environment of the powerhouse is created, including the upstream reservoir, downstream river channel, powerhouse slopes, and access roads.

[0041] In an optional implementation, in step 3, the creation of the 3D virtual scene can be achieved by: creating 3D models of three levels of detail for key equipment such as turbines, ball valves, and generators. The detailed model contains the complete geometry and detailed features of the equipment, used for close-up observation; it is displayed when the viewing distance is less than a first distance threshold. The standard model simplifies the secondary structure of the equipment while retaining the main features, used for medium-distance observation; it is displayed when the viewing distance is between the first and second distance thresholds. The simplified model retains only the basic outline of the equipment, used for long-distance observation; it is displayed when the viewing distance is greater than a second distance threshold. Two levels of detail are also created for auxiliary equipment such as pipes, valves, and cables. Based on the characteristics of the flooded power plant drill scenario, details are added to the detailed model for parts prone to leakage and rupture, such as the turbine top cover, ball valve body, and bypass pipe connections, to facilitate observation of equipment damage during drills.

[0042] In another optional implementation, in step 3, the creation of the 3D virtual scene can also be achieved by: establishing a water level simulation module within the 3D virtual scene, which calculates the real-time water level height based on the leakage rate of the current event and the drainage capacity of the sump. A 3D water body model is displayed in the 3D virtual scene based on the water level height, with the height of the 3D water body model corresponding to the calculated water level height. When the water level rises to the turbine level, the 3D water body model submerges the 3D equipment models on the turbine level, displaying the equipment models as submerged. When the water level continues to rise to the generator level, the 3D water body model submerges the 3D equipment models on the generator level. A water surface ripple effect is added to the surface of the 3D water body model to simulate the water surface state of accumulated water within the plant. A water flow effect is added at the location of the leaking equipment involved in the current event, with the flow rate determined by the leakage rate. A correlation is established between the water level and the equipment status; when equipment is submerged by the 3D water body model, the displayed status of that equipment is automatically changed to a fault state.

[0043] Step 4: Based on the contingency plan's handling procedures, set the standard emergency measures for each event in the event chain, including the following steps D1-D5: D1: Obtain the text of the emergency response plan for flooded factory buildings; D2: Perform semantic analysis on the emergency response plan text to identify the statements describing the emergency response process in the emergency response plan text; D3: Extract emergency response measures from the handling process statements; D4: Based on the equipment and operations involved in the emergency response measures, determine the event corresponding to the emergency response measures; D5: Link emergency response measures with corresponding events to form standard emergency measures for each event in the event chain.

[0044] In this embodiment, step 6 calculates text similarity by: establishing a hydropower station emergency response action terminology database, which includes verbs such as close, open, stop, start, cut off, evacuate, notify, and inspect; establishing a hydropower station emergency response object terminology database, which includes equipment names such as ball valves, fast gates, drainage pumps, standby drainage pumps, generators, excitation systems, and plant power supplies; and establishing a hydropower station emergency response department terminology database, which includes department and personnel names such as on-duty personnel, maintenance department, dispatch center, fire department, and emergency command center. Keywords of response measures and keywords of standard emergency measures are categorized and statistically analyzed according to three categories: action, object, and department. The ratio of the number of matching keywords in each category to the total number of keywords in that category is calculated to obtain the action matching degree, object matching degree, and department matching degree. The action matching degree, object matching degree, and department matching degree are then weighted and averaged according to action weight, object weight, and department weight to obtain the keyword matching degree. In this system, action weight represents the importance of the emergency response action, object weight represents the importance of the emergency response object, and department weight represents the importance of the emergency response department; the sum of these three weights is 1. Weight coefficients are set for keyword matching and key sentence semantic similarity, with the sum of these two weight coefficients being 1. The text similarity is then calculated by weighted summation of keyword matching and key sentence semantic similarity.

[0045] In an optional implementation, in step 6, text similarity can be calculated by: determining key emergency measures based on the flooded plant emergency plan. Key emergency measures include closing the inlet ball valve, closing the quick-release gate, starting the backup drainage pump, cutting off generator excitation, cutting off plant power, evacuating personnel from the turbine floor, evacuating personnel from the generator floor, and notifying the dispatch center. Importance weights are assigned to the key emergency measures in the standard emergency measures, with closing the inlet ball valve and closing the quick-release gate having the highest weights, followed by starting the backup drainage pump and cutting off power, and evacuating personnel and notifying departments having lower weights. When calculating keyword matching, it is determined whether the response measures contain the keywords corresponding to each key emergency measure. When a response measure contains the keyword of a key emergency measure, its matching score is the importance weight of that measure; otherwise, the score is zero. The matching scores of each key emergency measure are summed and divided by the sum of the weights of all key emergency measures to obtain the weighted keyword matching score. The weighted keyword matching score is used instead of the keyword matching score for text similarity calculation.

[0046] In another optional implementation, in step 6, text similarity can also be calculated by: performing a completeness check on the response measures to see if they contain keywords for the five essential steps: shutting off the water source, controlling the water level, restoring drainage, electrical safety, and personnel safety. If any keyword for a necessary step is missing from the response measures, a missing penalty coefficient is applied. The consistency of the order of occurrence of each keyword in the response measures with the corresponding order of occurrence of keywords in the standard emergency measures is calculated to obtain the order consistency score. The standard emergency measures follow the order of shutdown first, then control the water level, then restore drainage, then ensure electrical safety, and finally ensure personnel safety. A higher order consistency score is obtained when the order of occurrence of keywords in the response measures matches the standard order; otherwise, a lower order consistency score is obtained. The keyword matching score, key sentence semantic similarity, and order consistency score are weighted and summed, then multiplied by the missing penalty coefficient to calculate the comprehensive text similarity score.

[0047] Furthermore, recording deviations and generating exercise evaluation reports includes comparing text similarity with preset thresholds; When the text similarity is greater than a preset threshold, the response measures are deemed effective, and the outcome of the current event is identified based on the handling operations contained in the response measures. Based on the outcome of the current event and the relationship between adjacent events in the event chain through their current status and outcome, determine the next event in the event chain for the current event; Treat the next event as the current event, return to the steps of showing the emergency drill scenario to the participants, and continue the drill; When the text similarity is less than or equal to a preset threshold, the response measures are deemed ineffective. By comparing the response measures with the standard emergency response measures, the missing and incorrect emergency response measures in the response measures are extracted to form the deviation content; Record the response measures, text similarity and deviations for each event during the exercise, and generate an exercise evaluation report.

[0048] Example 3, referring to Figure 2 and Figure 3 This invention provides an intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology. To verify the beneficial effects of this invention, scientific demonstration is conducted through experiments.

[0049] Build as Figure 2 The virtual reality of a hydropower station, based on three-dimensional spatial topology, allows users to roam the plant scene from a first-person perspective. This facilitates quick and easy understanding of equipment and the plant environment for hydropower station personnel, enabling them to grasp the overall plant situation in a timely manner. With the help of external input devices, users can browse hydropower station accident scenarios with full freedom and from all angles.

[0050] like Figure 3As shown, an emergency scenario is constructed by building an event chain. In reality, an emergency case or sudden event can be viewed as a set of events E consisting of several event fragments e, such as... Figure 3 As shown, set E can represent the event development process as a "Start+Process+End" or "1+N+1" event chain according to the three states of "start, process, and end".

[0051] This embodiment uses a hydropower station flooding accident as an example to construct emergency drill scenarios for three typical accidents: turbine top cover rupture, bypass pipe bolt breakage, and drainage pump failure. Historical flooding accident cases were collected, including the turbine top cover rupture accident at the Saiyan Hydropower Station in Russia, the tailrace backflow accident at a power station in Guangxi, and the plug valve bursting accident at the Guanzhou Hydropower Station in Ganzi, Sichuan. The causes of accidents such as turbine top cover bolt breakage, tailrace backflow, and plug valve bursting were extracted from each case. The accident development process of water level rise leading to equipment flooding was analyzed, along with emergency response measures such as closing the quick-release gate, starting the backup drainage pump, cutting off power, and evacuating personnel, and the effectiveness of water level control. Based on the triggering mechanism of the accident causes, the causes were classified into equipment failure, natural disaster, and operational error. A correspondence was established between equipment failure and equipment such as turbines, ball valves, and drainage pumps; a correspondence was established between natural disaster and tailrace backflow; and a correspondence was established between operational error and misoperation of the quick-release gate. The timeline of the accident's development is marked, starting from the initial leak, then the water level rising to the turbine level, then to the generator level, and finally the complete submersion of equipment. A correlation is established between emergency response measures and the effectiveness of water level control; closing the rapid-release gates corresponds to cutting off the water source, and starting the backup drainage pumps corresponds to reducing the rate of water level rise.

[0052] A turbine roof rupture accident was selected as the target accident case. The initial triggering event was identified as the breakage of the turbine roof bolts. The subsequent events—a large influx of water into the powerhouse, a rapid rise in water level, submersion of the generator floor, and equipment damage—were identified as state change events. The successful closure of the intake quick-closing gate or the complete evacuation of personnel were identified as the emergency response completion event. Event segmentation points were set at the times the water level rose to the top of the turbine floor, to the bottom of the generator floor, and to the top of the generator floor, dividing the accident development process into three event segments: turbine roof rupture, turbine floor submersion, and generator floor submersion. The initial triggering event, the three state change events, and the emergency response completion event were connected chronologically to construct the powerhouse flooding accident event chain.

[0053] Based on the equipment involved in the event chain, such as turbines, generators, and transformers, and the locations of the turbine floor and generator floor, 3D models of the turbines, generators, transformers, and switchgear were created using 3ds Max. 3D models of the turbine floor, generator floor, intermediate floor, and hoisting floor were also created. 3D models of the water intake pipes, ball valves, quick-release gates, and bypass pipes were also created. 3D models of the sump wells, drainage pumps, and drainage pipes were also created. 3D models of the upstream reservoir, downstream river channel, and powerhouse slopes were also created. The created 3D models were imported into the Unity3D engine in FBX format, and the equipment 3D models were placed within the powerhouse 3D models according to their actual installation positions within the powerhouse 3D models, forming a 3D virtual scene of the hydropower station powerhouse.

[0054] Three levels of detail—high-resolution, standard, and simplified—were established for key equipment such as water turbines, ball valves, and generators. The high-resolution model includes the complete structure and detailed features of the equipment; the standard model simplifies secondary structures while retaining key features; and the simplified model only retains the basic outline. Using Unity's LODGroup component, the high-resolution model is displayed when the viewpoint is less than 10 meters from the equipment, the standard model is displayed when the distance is between 10 and 30 meters, and the simplified model is displayed when the distance is greater than 30 meters. Light and shadow baking was performed on the 3D virtual scene. The static equipment and factory 3D models were set as static objects, and Unity3D scene lighting baking technology was applied to record the shadows and reflections formed by the light as textures.

[0055] The system acquires equipment status information for a turbine top cover rupture event, including broken turbine top cover bolts, rapid outflow of high-pressure water, and a water flow rate of 50 cubic meters per minute. Based on this information, the virtual effects are determined to be water flow and rupture effects. A rupture effect is added to the turbine top cover to simulate its rupture. A water flow effect is added at the connection between the turbine top cover and the volute, using Unity3D's particle system with a particle emission rate of 5000 particles per second, a light blue color, and downward particle movement to simulate the rapid outflow of high-pressure water. Based on a leakage rate of 50 cubic meters per minute and a sump drainage capacity of 20 cubic meters per minute, the water level rise rate is calculated as 30 cubic meters per minute divided by the plant's floor area. A 3D model of the water body is created, with its height updated in real-time based on time and the water level rise rate. When the water body reaches the generator's bottom elevation, the generator's 3D model is displayed as submerged. Add water surface ripple effects to the surface of a 3D water model to generate an emergency drill scenario for flooding a factory.

[0056] Standard emergency measures were established based on the emergency plan for flooded power plants. These measures included: on-duty personnel immediately closing the inlet ball valve; on-duty personnel notifying the dispatch center; the operations and maintenance department starting the backup drainage pump; on-duty personnel cutting off generator excitation; the operations and maintenance department cutting off plant power; on-duty personnel evacuating personnel from the turbine floor; and the operations and maintenance department evacuating personnel from the generator floor. A flooded power plant emergency drill scenario was presented to the participants, who observed the virtual effects of a turbine roof rupture and water influx, and the dynamic process of rising water levels and gradual submersion of equipment. Participants input their proposed response measures for the turbine roof rupture incident through the Unity3D graphical user interface.

[0057] The response measures input by the participants are segmented into words to obtain a word set. From this word set, emergency response action words such as "shutdown," "start," "cut off," and "evacuate" are extracted; emergency response object words such as "ball valve," "fast gate," "drainage pump," "generator," and "plant power supply" are extracted; and emergency response department words such as "duty personnel," "maintenance department," and "dispatch center" are extracted as keywords. Syntactic analysis is performed on the response measures to identify sentences containing these keywords as key sentences. The number of matches between the keywords of the response measures and the keywords of the standard emergency measures is calculated. The response measures contain 8 keywords (shutdown, ball valve, start, drainage pump, cut off, generator, evacuate, personnel), while the standard emergency measures contain 10 keywords; the number of matches is 8, resulting in a keyword matching degree of 0.8. The key sentences of the response measures and the standard emergency measures are converted into sentence vectors. The cosine distance between the sentence vectors is calculated to be 0.25, and the semantic similarity of the key sentences is 0.75. The keyword matching degree and the semantic similarity of key sentences were weighted and summed, with the keyword matching degree having a weight of 0.4, the semantic similarity of key sentences having a weight of 0.6, and the text similarity having a weight of 0.85.

[0058] A preset threshold of 0.7 is set. A text similarity score of 0.85, greater than the preset threshold, is considered an effective response measure. Based on the action of closing the ball valve included in the response measure, the outcome of the current event is identified as water supply interruption. According to the relationship between adjacent events in the event chain through the current event state and the outcome, if the current event is a turbine top cover rupture and the outcome is water supply interruption, the next event is water level cessation and the activation of drainage pumps to lower the water level. The event of water level cessation is taken as the current event, the emergency drill scenario is updated, the 3D water model stops rising, and the process returns to the steps of showing the emergency drill scenario to the participants, continuing the drill. Participants propose a response measure of activating the backup drainage pump. After text similarity calculation determines its effectiveness, the next event is water level gradually decreasing. Participants then propose response measures of cutting off power and evacuating personnel. After these are deemed effective, the event chain reaches the emergency response completion event, and the drill ends. The response measures, text similarity scores, and deviations for each event during the drill are recorded, and a drill evaluation report is generated. The exercise evaluation report includes the names of the participants, the exercise subject being a turbine top cover rupture accident, the start and end times of the exercise, the response measures for each event and the text similarity, the total score of the exercise being 85 points, and the evaluation being good.

[0059] This embodiment verifies the effectiveness of the intelligent desktop simulation method for hydropower stations based on 3D simulation technology by conducting a complete drill of a turbine top cover rupture accident. It achieves a systematic representation of the accident evolution process by constructing an event chain, provides an intuitive display of the accident scene through 3D virtual scenes and virtual effects, realizes objectivity and automation of drill evaluation through text similarity calculation, and enables a drill cycle that determines the next event based on response measures through dynamic deduction.

[0060] An emergency case or sudden event can be regarded as an event set E consisting of several event fragments e, and the event development process of set E can be represented as a "Start + Process + End" or "1+N+1" event chain according to the three states of "start, process, and end".

[0061] In this sequence, the first "1" represents the starting event, which usually reflects the entire theme of the event; "N" represents one or more process events in the event chain; and the last node "1" represents the ending event, which means that the entire emergency response to the sudden event has been completed.

[0062] In the "Start + Process + End" event chain, each sudden event 'e' is considered a node. Nodes are interconnected using pointers, forming a directed acyclic development trend. The relationship between the current node and the next node can be expressed as: Here, en represents the current event situation, including the event type (earthquake, fire, rainstorm, etc.) and the conditions for occurrence (flood season, power outage, crustal movement, etc.); Resultn refers to the emergency measures taken for the current event and the corresponding results, including the departments involved (public security, transportation, fire protection, etc.) and specific measures (closing entertainment venues in response to the epidemic, carrying out equipment repairs in response to power outages, etc.).

[0063] Example 4 is an embodiment of the present invention, which provides a smart desktop simulation system for hydropower stations based on three-dimensional simulation technology, comprising: The accident case database construction module is used to collect historical cases of factory flooding accidents, extract the causes of accidents, the development process of accidents, and the results of emergency response from the accident cases, and build an accident case database. The event chain construction module is used to select target accident cases from the accident case library, identify the initial triggering event, state change event and emergency response completion event in the target accident case, and connect them in chronological order to construct an event chain. Adjacent events in the event chain are associated through the current event state and the response result. The virtual scene generation module is used to create a 3D virtual scene based on the equipment and locations involved in each event in the event chain, and to add virtual effects to the 3D virtual scene based on the equipment status information of each event to generate an emergency drill scene. The standard measures setting module is used to set standard emergency measures for each event in the event chain according to the contingency plan's handling procedures. The response measures acquisition module is used to display emergency drill scenarios to the participants and obtain the response measures proposed by the participants in response to the current event. The text similarity calculation module is used to extract keywords and key sentences from response measures and standard emergency measures, calculate keyword matching degree and key sentence semantic similarity, and calculate text similarity based on keyword matching degree and key sentence semantic similarity; The exercise evaluation module is used to determine the effectiveness of the response measures when the text similarity is greater than a preset threshold, and to determine the next event in the event chain based on the response measures and continue the exercise. When the text similarity is less than or equal to a preset threshold, the countermeasures are deemed invalid, the deviation is recorded, and an exercise evaluation report is generated.

[0064] This embodiment also provides an electronic device applicable to a smart desktop simulation method for hydropower stations based on three-dimensional simulation technology, including: a memory and a processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology proposed in the above embodiments.

[0065] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements a smart desktop simulation method for hydropower stations based on three-dimensional simulation technology as proposed in the above embodiment.

[0066] The storage medium proposed in this embodiment belongs to the same inventive concept as the intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology proposed in the above embodiments. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0067] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0068] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for intelligent desktop simulation of hydropower stations based on three-dimensional simulation technology, characterized in that: include, Collect historical cases of factory flooding accidents, extract the causes, development process and emergency response results from these cases, and build an accident case database. Select a target accident case from the accident case database, identify the initial triggering event, state change event, and emergency response completion event in the target accident case, and connect them in chronological order to construct an event chain. Adjacent events in the event chain are associated through the current event state and the response result. A three-dimensional virtual scene is created based on the equipment and locations involved in each event in the event chain. Virtual effects are added to the three-dimensional virtual scene based on the equipment status information of each event to generate an emergency drill scenario. Standard emergency measures are set for each event in the event chain according to the contingency plan's handling procedures; Show the participants the emergency drill scenario and obtain their proposed response measures for the current event; Extract keywords and key sentences from response measures and standard emergency measures, calculate keyword matching degree and key sentence semantic similarity, and calculate text similarity based on keyword matching degree and key sentence semantic similarity; When the text similarity is greater than a preset threshold, the response measures are deemed effective. Based on the response measures, the next event in the event chain is determined and the exercise continues. When the text similarity is less than or equal to a preset threshold, the countermeasures are deemed invalid, the deviation is recorded, and an exercise evaluation report is generated.

2. The intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology as described in claim 1, characterized in that: The construction of the accident case library includes classifying accident causes according to the triggering mechanism of the accident causes; Establish a correspondence between accident causes and equipment types; The accident development process is time-series labeled, and the time information of each state node during the accident development process is labeled; Establish the relationship between state nodes and their changes over time; Identify the emergency response measures and their effects in the emergency response outcome; Establish the correlation between emergency response measures and their effectiveness; An accident case library is constructed based on the correspondence between accident causes and equipment types, the relationship between status nodes and changes over time, and the correlation between emergency response measures and their effects.

3. The intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology as described in claim 2, characterized in that: The step of constructing an event chain in chronological order includes selecting target accident cases from an accident case library; Identify the time points in the accident development process of the target accident case where the state nodes change as event segmentation points; The accident development process is divided into multiple event segments at the event inflection point; Extract the time and status information of each event segment; Identify the earliest event segment based on time information as the initial trigger event; Identify event segments where the state changes based on the state information as state change events; Identify event segments that indicate the completion of emergency response as emergency response completion events; Connect the initial triggering event, the state change event, and the emergency response completion event according to the time information to construct an event chain.

4. The intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology as described in claim 3, characterized in that: The standard emergency measures for each event in the event chain according to the contingency plan include obtaining the text of the emergency plan for flooded factory buildings. Semantic analysis is performed on emergency response plan texts to identify statements describing emergency response procedures. Extract emergency response measures from the handling process statements; Based on the equipment and operations involved in the emergency response measures, determine the event corresponding to the emergency response measures; By associating emergency response measures with corresponding events, standard emergency measures are formed for each event in the event chain.

5. The intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology as described in claim 4, characterized in that: The calculation of keyword matching degree and key sentence semantic similarity includes performing word segmentation on response measures and standard emergency measures to obtain a word set; Extract words from the word set that represent emergency response actions, emergency response targets, and emergency response departments as keywords; Syntactic analysis was performed on response measures and standard emergency measures to identify sentences containing keywords as key sentences; Calculate the number of matches between the keywords of the response measures and the keywords of the standard emergency measures, and calculate the keyword matching degree based on the ratio of the number of matches to the total number of keywords; The key sentences of the response measures and the key sentences of the standard emergency measures are converted into sentence vectors. The cosine distance between the sentence vectors is calculated, and the semantic similarity of the key sentences is calculated based on the cosine distance. The text similarity is calculated by weighted summation of keyword matching degree and key sentence semantic similarity.

6. The intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology as described in claim 5, characterized in that: The process of recording deviations and generating an exercise evaluation report includes comparing text similarity with a preset threshold. When the text similarity is greater than a preset threshold, the response measures are deemed effective, and the outcome of the current event is identified based on the handling operations contained in the response measures. Based on the outcome of the current event and the relationship between adjacent events in the event chain through their current status and outcome, determine the next event in the event chain for the current event; Treat the next event as the current event, return to the steps of showing the emergency drill scenario to the participants, and continue the drill; When the text similarity is less than or equal to a preset threshold, the response measures are deemed ineffective. By comparing the response measures with the standard emergency response measures, the missing and incorrect emergency response measures in the response measures are extracted to form the deviation content; Record the response measures, text similarity and deviations for each event during the exercise, and generate an exercise evaluation report.

7. The intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology as described in claim 6, characterized in that: The establishment of the three-dimensional virtual scene includes identifying the device types and locations involved in each event in the event chain; Create a 3D model of the equipment based on its type, and a 3D model of the factory based on its location. Place the 3D model of the equipment into the 3D model of the factory according to its location to form a 3D virtual scene. The 3D models of devices in the 3D virtual scene are classified according to their level of detail, and different display distance ranges are set for 3D models of devices with different levels of detail. The 3D virtual scene is baked with light and shadow, and the shadows and reflections formed by light shining on the surface of the 3D model of the equipment and the 3D model of the factory are recorded as textures. Retrieve device status information for the current event in the event chain; The type of virtual effect is determined based on the equipment status information. When the equipment status information indicates that the equipment is leaking, a water flow effect is generated; when the equipment status information indicates that the equipment is vibrating, a vibration effect is generated; when the equipment status information indicates that the equipment is broken, a breakage effect is generated. Add virtual effects to the locations of devices involved in the current event in a 3D virtual scene to generate an emergency drill scenario.

8. A smart desktop simulation system for hydropower stations based on three-dimensional simulation technology, employing the smart desktop simulation method for hydropower stations based on three-dimensional simulation technology as described in any one of claims 1 to 7, characterized in that, include: The accident case database construction module is used to collect historical cases of factory flooding accidents, extract the causes of accidents, the development process of accidents, and the results of emergency response from the accident cases, and build an accident case database. The event chain construction module is used to select target accident cases from the accident case library, identify the initial triggering event, state change event and emergency response completion event in the target accident case, and connect them in chronological order to construct an event chain. Adjacent events in the event chain are associated through the current event state and the response result. The virtual scene generation module is used to create a 3D virtual scene based on the equipment and locations involved in each event in the event chain, and to add virtual effects to the 3D virtual scene based on the equipment status information of each event to generate an emergency drill scene. The standard measures setting module is used to set standard emergency measures for each event in the event chain according to the contingency plan's handling procedures. The response measures acquisition module is used to display emergency drill scenarios to the participants and obtain the response measures proposed by the participants in response to the current event. The text similarity calculation module is used to extract keywords and key sentences from response measures and standard emergency measures, calculate keyword matching degree and key sentence semantic similarity, and calculate text similarity based on keyword matching degree and key sentence semantic similarity; The exercise evaluation module is used to determine the effectiveness of the response measures when the text similarity is greater than a preset threshold, and to determine the next event in the event chain based on the response measures and continue the exercise. When the text similarity is less than or equal to a preset threshold, the countermeasures are deemed invalid, the deviation is recorded, and an exercise evaluation report is generated.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the intelligent desktop simulation method for hydropower stations based on three-dimensional simulation technology as described in any one of claims 1 to 7.

Images