Visual simulation training method and system based on thermal power plant

Abstract

The invention discloses a visual simulation training system and method based on a thermal power plant, and the system comprises a data collection module which comprises a multi-source heterogeneous sensor array and is used for collecting various key numbers of actual operation equipment of the thermal power plant; the data processing module is used for carrying out cleaning, normalization and feature extraction processing on the collected data and fusing the data; the visual model construction module is used for constructing a model containing equipment three-dimensional geometry, a connection relation and dynamic operation; the training scene setting module is used for setting various training scenes and adjusting operation parameters of the visual simulation model; and the training interaction module is used for realizing interaction operation between the trainees and the visual simulation model and giving feedback information, the trainees can operate the model, and the system gives feedback conforming to actual specifications according to the operation. The invention belongs to the technical field of thermal power plants, and particularly provides a method for solving the problems that in the prior art, on-site practical operation has safety risks, models are not accurate enough, and training scenes are single.

Description

Technical Field

[0001] The present invention belongs to the technical field of thermal power plants, and specifically refers to a visualization simulation training method and system based on thermal power plants. Background Art

[0002] The equipment in thermal power plants is complex and the operation process is cumbersome. Traditional training methods mainly rely on theoretical explanations and on-site practical operations. However, on-site practical operations have safety risks and the operating status of equipment is limited.

[0003] At the same time, traditional practical training needs to be carried out on operating units, there is a risk of exposure to high-temperature and high-pressure environments. Moreover, existing simulation training systems mostly adopt simplified physical models, and there are problems with inaccurate models. Conventional systems only preset typical working conditions and lack the ability to generate dynamic working conditions, resulting in a single training scenario and cannot well meet the training needs of different levels of employees in thermal power plants. Summary of the Invention

[0004] The technical problem to be solved by the present invention is the problems of safety risks in on-site practical operations, inaccurate models, and single training scenarios in the prior art.

[0005] To solve the above problems, the technical solutions adopted by the present invention are as follows:

[0006] The present invention proposes a visualization simulation training system based on thermal power plants, including:

[0007] Data acquisition module: It includes a multi-source heterogeneous sensor array, which is used to collect various key data of the actual operating equipment in the thermal power plant, collect operating data from the actual operating equipment in the thermal power plant to ensure the comprehensiveness of the data;

[0008] Data processing module: Clean, normalize, and extract features from the collected data, and fuse the data to improve the data quality for constructing an accurate visualization model;

[0009] Visualization model construction module: Based on the processed data, construct a visualization simulation model of the thermal power plant, construct a model including three-dimensional geometry, connection relationships, and dynamic operations of the equipment, so that trainees can intuitively see the overall operation of the thermal power plant;

[0010] Training scenario setting module: Set a variety of training scenarios and adjust the operating parameters of the visualization simulation model;

[0011] Training interaction module: Realize the interactive operation between trainees and the visualization simulation model and give feedback information. Trainees can operate the model, and the system gives feedback that conforms to actual specifications according to the operation.

[0012] Preferably, the data acquisition module includes vibration sensors, temperature sensors, pressure sensors, flow sensors, rotational speed sensors, infrared thermal imagers, and ultrasonic flaw detectors; it also includes a data acquisition terminal, which uses an industrial-grade edge computing gateway.

[0013] Preferably, the data cleaning is based on an outlier rejection algorithm using the K-S test, and the data fusion uses the D-S evidence theory to fuse multi-sensor data.

[0014] Preferably, the model includes a three-dimensional geometric model of the equipment, a connection relationship model between the equipment, and a dynamic operation model based on data. Among them, the three-dimensional geometric model of the equipment accurately reflects the appearance and internal structure of the actual equipment, the connection relationship model between the equipment accurately presents the layout and connection method of each equipment in the thermal power plant, and the dynamic operation model presents the real-time operation state of the equipment according to the data.

[0015] Preferably, the visualization model construction module uses 3D modeling software to construct the three-dimensional geometric model of the equipment, and realizes the connection relationship model and the dynamic operation model between the equipment through a scripting language.

[0016] Preferably, the training scenario setting module sets multiple training scenarios according to different training needs, such as normal operation scenarios, fault troubleshooting scenarios, and emergency handling scenarios, and adjusts the operation parameters of the visualization simulation model under each scenario to meet the scenario requirements.

[0017] Preferably, the training scenario setting module generates dynamic scenarios based on a scenario knowledge graph and an abnormal condition generator based on the GAN network.

[0018] Preferably, the training interaction module is based on a multi-modal interaction interface, including VR handle force feedback, voice command recognition, etc.

[0019] Preferably, the feedback information is based on the operation logic of the visualization simulation model and the operation specifications of the actual thermal power plant.

[0020] As a further solution, the present invention proposes a training method for the above system, including the following steps:

[0021] S1: The data acquisition module deploys sensors to the thermal power plant equipment, starts collecting data and transmits it to the data processing module;

[0022] S2: The data processing module processes the data according to the set algorithm and then transmits it to the visualization model construction module;

[0023] S3: After the visualization model construction module constructs the model, it hands it over to the training scenario setting module for scenario setting;

[0024] S4: The trainee enters the training interaction module and starts the interactive training with the visualization model in the set scenario.

[0025] The beneficial effects achieved by the present invention with the above solution are as follows:

[0026] 1. By operating in the virtual environment through this system, the trainees avoid the safety risks of on-site practical operations.

[0027] 2. The visual models and various scenario settings enable the trainees to better understand the operation principles and operation processes of thermal power plants.

[0028] 3. The training scenarios and contents can be quickly adjusted according to the needs of employees at different levels and with different capabilities, reducing the dependence on actual equipment and lowering the equipment wear problem. Description of the Drawings

[0029] Figure 1 It is the system framework diagram of a visualization simulation training system based on a thermal power plant provided for this embodiment;

[0030] Figure 2 It is the flowchart of a visualization simulation training method based on a thermal power plant provided for this embodiment.

[0031] The drawings are used to provide further understanding of the present invention, and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention, but do not constitute a limitation to the present invention. Detailed Embodiments

[0032] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments; based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0033] In the description of the present invention, it should be noted that the orientation or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the orientation or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be understood as a limitation to the present invention.

[0034] Embodiment 1

[0035] As Figure 1As shown in the figure, this embodiment proposes a visualization simulation training system based on a thermal power plant, including a data acquisition module: which includes a multi-source heterogeneous sensor array for collecting various key data of the actual operating equipment in the thermal power plant, collecting operating data from the actual operating equipment in the thermal power plant to ensure the comprehensiveness of the data.

[0036] Among them, the data acquisition module includes vibration sensors, temperature sensors, pressure sensors, flow sensors, speed sensors, infrared thermal imagers, and ultrasonic flaw detectors; it also includes a data acquisition terminal, which uses an industrial-grade edge computing gateway.

[0037] The system also includes a data processing module: which cleans, normalizes, and extracts features from the collected data, and fuses the data to improve the data quality for building an accurate visualization model; the data cleaning is based on an outlier removal algorithm using the K-S test, and the data fusion uses the D-S evidence theory to fuse multi-sensor data.

[0038] The system also includes a visualization model construction module: which constructs a visualization simulation model of the thermal power plant based on the processed data, constructs a model including the three-dimensional geometry of the equipment, connection relationships, and dynamic operations, enabling trainees to intuitively see the overall operation of the thermal power plant.

[0039] Among them, the model includes a three-dimensional geometric model of the equipment, a connection relationship model between the equipment, and a dynamic operation model based on data. The three-dimensional geometric model of the equipment accurately reflects the appearance and internal structure of the actual equipment, the connection relationship model between the equipment accurately presents the layout and connection methods of each equipment in the thermal power plant, and the dynamic operation model presents the real-time operation status of the equipment according to the data.

[0040] In a preferred embodiment, the visualization model construction module uses three-dimensional modeling software to construct the three-dimensional geometric model of the equipment, and realizes the connection relationship model and the dynamic operation model between the equipment through a scripting language.

[0041] The system also includes a training scenario setting module: which is used to set various training scenarios and adjust the operation parameters of the visualization simulation model. The training scenario setting module sets various training scenarios according to different training needs, such as normal operation scenarios, fault troubleshooting scenarios, and emergency handling scenarios, and adjusts the operation parameters of the visualization simulation model under each scenario to meet the scenario requirements.

[0042] Among them, the training scenario setting module realizes the generation of dynamic scenarios based on a scenario knowledge graph and an abnormal working condition generator based on a GAN network.

[0043] Among them, the scenario knowledge graph:

[0044] Contains 137 typical nodes (such as "turbine runaway", "desulfurization tower blockage"); node association strength matrix W = [w ij 137 × 137 .

[0045] Dynamic scenario generation:

[0046] An abnormal condition generator based on a GAN network, input: random noise with N(μ, σ) distribution, output: non-preset accident scenarios (such as a compound fault of "feedwater pump cavitation + air preheater fire").

[0047] The system also includes a training interaction module: realizing the interactive operation between the trainee and the visual simulation model and giving feedback information. The trainee can operate the model, and the system gives feedback that conforms to the actual specifications according to the operation.

[0048] Among them, the training interaction module is based on a multimodal interaction interface, including VR handle force feedback, voice command recognition, etc. The feedback information is based on the operation logic of the visual simulation model and the operation specifications of the actual thermal power plant.

[0049] Embodiment 2

[0050] As Figure 2 shown, this embodiment provides a training method for a visual simulation training system based on a thermal power plant, including the following steps:

[0051] S1: The data acquisition module deploys sensors to the thermal power plant equipment, starts collecting data and transmits it to the data processing module;

[0052] S2: The data processing module processes the data according to the set algorithm and then transmits it to the visual model construction module;

[0053] S3: After the visual model construction module constructs the model, it hands it over to the training scenario setting module for scenario setting;

[0054] S4: The trainee enters the training interaction module and starts interactive training with the visual simulation model in the set scenario.

[0055] Embodiment 3

[0056] A more detailed description of the training method is given based on Embodiment 1 and Embodiment 2:

[0057] S1. In the data acquisition stage, install sensors on each key equipment of the thermal power plant:

[0058] ​For boiler equipment, install temperature sensors on the furnace wall to monitor the temperature distribution inside the furnace, install pressure sensors and flow sensors on the steam pipeline to obtain the state parameters of the steam, and install speed sensors at the feed pump to monitor the operating speed of the pump. For steam turbine equipment, install pressure sensors at the inlet and outlet respectively, install vibration sensors at the bearing parts (which can be used as supplementary special speed-related data), and install flow sensors at the regulating valve. For generator equipment, install temperature sensors on the stator winding and speed sensors at the rotor, etc.

[0059] S2. Start the sensor data acquisition system:

[0060] Set appropriate data acquisition frequencies. For example, acquire temperature data every 5 minutes, pressure and flow data every 2 minutes, and speed data every 1 minute. Ensure that the acquisition system communicates normally with each sensor and transmit the acquired data to the data processing module in real time.

[0061] S3. Data processing stage:

[0062] First, conduct preliminary screening and cleaning on the acquired data to remove obvious outliers that do not conform to physical laws. For example, if the temperature sensor suddenly shows data exceeding 50% of the upper limit of the normal operating temperature of the equipment, it is determined as an outlier and removed. Exclude duplicate data records. Since the acquisition system may have occasional repeated transmissions, remove the duplicates by comparing the acquisition time and key parameter values of the data.

[0063] Secondly, perform normalization processing on the cleaned data and determine the normalization formula for different types of data. For example, for pressure data, if the actual measurement range is 0 - 10 MPa, use the formula: normalized value = (actual pressure value - 0) / (10 - 0) to map the pressure value to the interval [0, 1].

[0064] Then, carry out the feature extraction step. For temperature data, extract the temperature change rate as a feature by calculating the ratio of the difference between two adjacent acquired temperature values to the time interval. For pressure data, extract the pressure fluctuation amplitude feature, that is, the difference between the maximum and minimum pressure values within a certain period of time. For flow data, calculate the flow stability coefficient feature by statistically calculating the ratio of the standard deviation to the average value of the flow within a certain period of time, etc.

[0065] S4. Visualization model construction stage:

[0066] Use professional 3D modeling software (such as SolidWorks or 3ds Max) to accurately model equipment such as boilers, steam turbines, and generators. In the boiler model, detailedly construct the 3D shapes of components such as the furnace, water wall, superheater, and reheater, accurately reflecting their dimensional ratios and internal structures. According to the actual layout of the thermal power plant, determine the connection orientation of the steam pipeline between the boiler and the steam turbine, and accurately present the connection points and orientation of the pipeline in the model.

[0067] Secondly, construct a dynamic operation model. Based on the characteristic data obtained from data processing, write a scripting language to implement the dynamic operation logic of the equipment.

[0068] For example, when the steam pressure of the boiler increases, set the opening of the steam inlet valve of the steam turbine to be automatically adjusted through the script, thereby affecting the rotational speed of the steam turbine and the output power of the generator, and reflecting these change relationships in real time in the model.

[0069] S5. Training scenario setting:

[0070] In the visual simulation model, set the initial parameters of each device to the values under normal startup conditions. For example, set the initial water level of the boiler to the lower limit value within the normal range, the initial rotational speed of the steam turbine to 0, and the initial output voltage of the generator to 10% of the rated voltage.

[0071] Provide an example of a troubleshooting scenario:

[0072] Steam pipeline leakage troubleshooting scenario: Randomly set a leakage point in the steam pipeline model, and simulate the pressure drop and abnormal flow during leakage by adjusting sensor data. Some misleading parameter changes can also be set in the model. For example, the temperature sensor near the leakage point shows a slightly increased temperature due to heat dissipation and other factors, increasing the difficulty of troubleshooting.

[0073] S6. Training interaction:

[0074] Step 1) The new employee logs in to the training system and enters the visual simulation model under the steam pipeline leakage troubleshooting scenario.

[0075] Step 2) The new employee first views the parameter display panels of each device and finds that the steam pipeline pressure continues to drop and the flow is abnormal.

[0076] Step 3) Try to close some valves connected to the leaking pipeline, and the system judges whether the valve closing is correct according to the operation logic. If the correct valve is closed, the system displays feedback information indicating that the pressure drop trend slows down; if the closing is incorrect, the system prompts a warning message that may affect other devices.

[0077] Step 4) The new employee further checks the temperature sensor data. After discovering misleading information about a slightly rising temperature, the new employee finally determines the location of the real leakage point by analyzing the pipeline layout and equipment connection relationships.

[0078] Step 5) During the whole operation process, the system records information such as the operation steps, operation time, and operation results of the new employee for evaluation and analysis after the training.

[0079] Through the above detailed steps of the embodiments, new employees can gradually and deeply learn knowledge and skills such as equipment operation and fault troubleshooting in a thermal power plant in the visualization simulation training system based on a thermal power plant.

[0080] The above describes the present invention and its implementation manners. Such a description is not restrictive. What is shown in the drawings is only one of the implementation manners of the present invention, and the actual structure is not limited thereto. Generally speaking, if those of ordinary skill in the art are inspired by it and design similar structural manners and embodiments to this technical solution without creative efforts without departing from the purpose of the present invention, they shall fall within the protection scope of the present invention.

Claims

1. A visualization simulation training system based on a thermal power plant, characterized in that Including: Data acquisition module: It includes a multi-source heterogeneous sensor array, which is used to collect various key data of the actual operating equipment in the thermal power plant and collect operation data from the actual operating equipment in the thermal power plant. Data processing module: It cleans, normalizes, and extracts features from the collected data, and fuses the data. Visualization model construction module: Based on the processed data, it constructs a visualization simulation model of the thermal power plant, constructs a model including the three-dimensional geometry, connection relationship, and dynamic operation of the equipment, so that trainees can intuitively see the overall operation of the thermal power plant. Training scenario setting module: It sets multiple training scenarios and adjusts the operation parameters of the visualization simulation model. Training interaction module: It realizes the interactive operation between trainees and the visualization simulation model and gives feedback information. Trainees can operate the model, and the system gives feedback that conforms to the actual specifications according to the operation.

2. The visual simulation training system based on a thermal power plant according to claim 1, wherein: The data acquisition module includes vibration sensors, temperature sensors, pressure sensors, flow sensors, speed sensors, infrared thermal imagers, ultrasonic flaw detectors; it also includes a data acquisition terminal using an industrial-grade edge computing gateway.

3. The visual simulation training system based on a thermal power plant according to claim 1, characterized in that: The data cleaning is based on an outlier rejection algorithm using the K-S test, and the data fusion uses the D-S evidence theory to fuse multi-sensor data.

4. A visualization simulation training system based on a thermal power plant according to claim 1, characterized in that: The model includes a three-dimensional geometric model of the equipment, a connection relationship model between the equipment, and a dynamic operation model based on data. Among them, the three-dimensional geometric model of the equipment accurately reflects the appearance and internal structure of the actual equipment, the connection relationship model between the equipment accurately presents the layout and connection method of each equipment in the thermal power plant, and the dynamic operation model presents the real-time operation state of the equipment according to the data.

5. A visual simulation training system based on a thermal power plant according to claim 1 or 4, characterized in that: The visualization model construction module uses 3D modeling software to construct the three-dimensional geometric model of the equipment, and realizes the connection relationship model and dynamic operation model of the equipment through a scripting language.

6. The visual simulation training system based on a thermal power plant according to claim 1, wherein: The training scenario setting module sets multiple training scenarios according to different training needs, such as normal operation scenarios, fault troubleshooting scenarios, and emergency handling scenarios, and adjusts the operation parameters of the visualization simulation model under each scenario to meet the scenario requirements.

7. A visualization simulation training system based on a thermal power plant according to claim 1 or 6, characterized in that: The training scenario setting module realizes the generation of dynamic scenarios based on a scenario knowledge graph and an abnormal working condition generator based on the GAN network.

8. A visualization simulation training system based on a thermal power plant according to claim 1 or 6, characterized in that: The training interaction module is based on a multi-modal interaction interface, including VR handle force feedback and voice command recognition.

9. The visual simulation training system based on a thermal power plant according to claim 1 or 6, wherein: The feedback information is based on the operation logic of the visualization simulation model and the operation specifications of the actual thermal power plant.

10. A training method for a visualization simulation training system based on a thermal power plant as described in any one of claims 1-9, characterized in that, Including the following steps: S1: The data acquisition module deploys sensors to the thermal power plant equipment, starts collecting data, and transmits it to the data processing module. S2: The data processing module processes the data according to the set algorithm and transmits it to the visualization model construction module. S3: After the visualization model construction module constructs the model, it hands it over to the training scenario setting module for scenario setting. S4: Trainees enter the training interaction module and start interactive training with the visualization model in the set scenario.

Images