734 results about "Nuclear plant" patented technology

The invention provides a nuclear power regulation human factor trap identification method and device based on a knowledge graph, and the method comprises the steps: determining a current nuclear power regulation according to a to-be-executed task, and extracting an operation target in the current nuclear power regulation; on the basis of a pre-constructed interface knowledge graph, according to the current nuclear power regulation, calculating all regulation paths from a task starting point node to a task end point node; the interface knowledge graph takes interaction elements related to operation in the interface of the nuclear power plant simulator as nodes and takes the relationship between the interaction elements as edges; the task end node is an operation target; and obtaining a human factor trap identification result of the current nuclear power regulation according to a path structure in the regulation path. According to the method, the interaction elements in the nuclear power plant simulator interface are subjected to structured mapping through the interface knowledge graph, automatic recognition of the structural human factor trap in the nuclear power regulation can be effectively achieved, data support and logic basis are provided for regulation optimization and interface design, and finally safety risks caused by human errors are reduced.

The invention provides an anomaly diagnosis method for a nuclear power plant gamma energy spectrum continuous monitoring process, and the method comprises the following steps: S1, determining at least one target gamma characteristic peak which needs to be monitored under the current monitoring condition of a nuclear power plant, and calculating the anomaly of the target gamma characteristic peak based on the performance index of a gamma spectrometer and the expected energy of the target gamma characteristic peak; setting a parameter allowable range of each target gamma characteristic peak; the parameter permissible range at least comprises a peak center energy permissible range and a resolution permissible range; s2, acquiring a gamma energy spectrum output by the gamma spectrometer, and processing the energy spectrum to obtain actual measurement parameters of a target gamma characteristic peak; the measured parameters at least comprise a peak center energy measured value and a resolution measured value; s3, the actually measured parameters are compared with the corresponding parameter allowable ranges; and S4, if the actually measured parameter of any target gamma characteristic peak exceeds the corresponding parameter allowable range, determining that the state of the gamma spectrometer is abnormal. Automatic and real-time abnormity diagnosis of the monitoring state of the gamma spectrometer is realized.

The invention provides a nuclear power plant safety system based on causal reasoning, and relates to the technical field of nuclear power plant safety monitoring and fault diagnosis. The nuclear power plant safety system comprises a multi-source sensing layer and a causal engine layer, the multi-source sensing layer is used for collecting operation parameters of a nuclear power plant, and the causal engine layer comprises a dynamic causal network construction module and an online reasoning module. The dynamic causal network construction module is used for constructing a dynamic causal network according to the operation parameters, a continuous causal edge generated based on an equipment physical equation and a discrete event edge generated based on a probabilistic safety evaluation event tree; and the online reasoning module is used for determining a key fault propagation path according to the dynamic causal network so as to locate a fault source. According to the method, the equipment physical equation is embedded into the dynamic causal network, so that the reasoning process is ensured to accord with thermal hydraulic and safety physical rules, the problem of high false alarm rate caused by correlation misjudgment in deep learning is avoided, and the safety of the nuclear power plant is improved.

The invention belongs to the technical field of nuclear power, and particularly relates to a nuclear power plant vibration measurement method and device based on a dynamic range technology. Comprising the following steps: sampling vibration signals after gain amplification by using a plurality of analog-to-digital converters with different measuring ranges to obtain an initial digital sequence; the initial digital sequence comprises a signal amplitude; performing amplitude region division on the initial digital sequence according to the magnitude of the signal amplitude to obtain a plurality of signal segments in different amplitude regions, and endowing the analog-to-digital converter with a target weight according to the saturation of the signal segments in the measuring range of each analog-to-digital converter; the saturation is the ratio of the signal amplitude to the upper limit of the measuring range of the analog-to-digital converter; the target weight comprises sub-weights respectively corresponding to a plurality of different signal segments of the analog-to-digital converter; and performing weighted fusion and splicing on the signal segments based on the sub-weights of the analog-to-digital converters in the different signal segments to obtain a target sampling signal.

The invention discloses a risk early warning method and system for an emergency diesel generator of a nuclear power plant, and the method comprises the steps: collecting operation parameters of a generator set in a hot standby state, and when the parameters are not in a preset normal range, indicating that the parameters are abnormal and the generator set breaks down; preprocessing the acquired operating parameters, and extracting effective features of the preprocessed operating parameters; on the basis of the generator set with the fault, combining with the effective characteristics, and utilizing a high-frequency data flow transient anomaly recognition algorithm to capture an anomaly type label to obtain fault information; evaluating the risk degree of the generator set according to the effective features and the abnormal type labels of the preprocessed operation parameters; and carrying out risk early warning and response on the generator set according to the risk degree of the generator set. According to the invention, the operation state of the emergency diesel generator set can be monitored in real time, accurate risk assessment and early warning information is provided, and the risk management capability of managers is enhanced.

The invention belongs to the technical field of reliability management of nuclear power plants, and particularly relates to a multi-focal-length visible light array cameras, a cooperative control unit and a central analysis platform, the multiple multi-focal-length visible light array cameras are located on the end side, and conversion from physical world real scenes to digital signal images is completed; the cooperative control unit is located on the side and is responsible for receiving, processing and analyzing data of the multiple multi-focal-length visible light array cameras, executing AI reasoning with high delay requirements, driving a holder and a lens to act and uploading related result data. The central analysis platform is located at a cloud side, gathers alarm data of all side devices, and carries out global situation presentation, historical data storage and backtracking, long-term trend analysis and multi-system linkage. The system has the beneficial effects that the tradeoff limitation among the view field, the distance and the precision in the prior art is overcome, and the industrial problem that the large view field cannot be considered is creatively solved through the cooperative working mode of the multi-focal-length optical array.

The invention provides a nuclear power plant thermal discharge dynamic change prediction analysis method, and relates to the technical field of data processing, and the method comprises the steps: collecting a temperature observation data set, extracting a temperature maximum value sequence and a temperature minimum value sequence in an equal-tide period, calculating a wave amplitude stable factor, extracting a plurality of trend inflection points of a temperature change first-order derivative curve, and calculating the temperature change first-order derivative curve. The method comprises the following steps: determining thermal expansion boundary candidate areas caused by thermal drainage, evaluating a time interval and amplitude difference between a lag response inflection point in each thermal expansion boundary candidate area and a previous dominant temperature rise point to obtain a thermal lag offset value, extracting a boundary sensitive area, performing multipoint communication fitting, determining a maximum space extrapolation boundary influenced by the thermal drainage effect, and performing thermal expansion on the maximum space extrapolation boundary. And carrying out trajectory tracking and change direction regression on the extrapolation boundaries at multiple different time points, and identifying a heat disturbance movement trend to obtain dynamic evolution data. The method can identify the diffusion trend of the warm discharged water and predict the temperature change.

The invention relates to the technical field of nuclear power plant automatic control and fluid regulation, in particular to a nuclear power plant condensation water regulating valve automatic control system which comprises a sensor data acquisition and synchronization module which acquires and synchronizes parameters in a nuclear power plant condensation water system; the reinforcement learning control module reads parameters of the sensor data acquisition and synchronization module, and generates a regulating valve control strategy by using a reinforcement learning algorithm; the real-time prediction and control decision module controls and predicts the opening degree of the condensed water regulating valve according to the regulating strategy of the reinforcement learning control module; and the anomaly detection and safety protection module reads the parameters acquired by the data acquisition and synchronization module, monitors the parameters and gives an alarm according to an anomaly monitoring result. According to the system, valve oscillation and unreasonable adjustment in a traditional method can be avoided while a valve adjustment strategy is learned and optimized in real time, and the precision and stability of condensed water flow adjustment are improved.

A knowledge graph-based nuclear power plant fault diagnosis and display method and system, an electronic device, and a storage medium. The method comprises the following steps: monitoring, in real time, a key device to be monitored, and acquiring real-time information of said key device (S101); on the basis of the real-time information, determining whether there is an anomaly early-warning for said key device (S102); if there is an anomaly early-warning, identifying a fault data feature of said key device (S103); acquiring a related knowledge graph model in a knowledge base and calling the knowledge graph model (S104); performing complex knowledge decomposition and reasoning on the basis of the related knowledge graph model and the fault data feature to obtain a fault diagnosis result (S105); and displaying the fault diagnosis result (S106).

The invention discloses a nuclear power plant graph attention network diagnosis method based on fault causal guidance, and relates to the technical field of nuclear power plant fault diagnosis. In order to solve the technical defect that a nuclear power plant fault diagnosis method in the prior art does not fully utilize causal information in a knowledge-driven method, the technical scheme provided by the invention comprises the following steps: acquiring operation data and high-precision simulation data of a preset number of subsystems of a nuclear power plant; carrying out fault state labeling on the simulation data; establishing a causal propagation path between the sensor nodes; extracting time sequence features of the sensor data, and constructing a time-space fused fault propagation graph; hierarchical aggregation of fault features is realized; feature extraction is carried out on the sensor data, and local modes and key fault features in the data are captured; optimizing the connection weight of the same-order neighbor nodes in the fault propagation graph; and calculating hidden variable representation, and outputting a fault diagnosis result. The method is suitable for nuclear power plant operation monitoring, intelligent fault diagnosis, safety early warning and the like.

The invention relates to the field of nuclear power operation, in particular to a control method for quantitative ammonia potassium saturation of cation resin of a primary loop purification system of a nuclear power plant. The method comprises the following steps: before starting ammonia-potassium saturation, calculating the total amount of ammonia absorbed by resin, the total amount of potassium absorbed by resin and the ideal concentration of a potassium hydroxide solution by utilizing the total exchange amount of nuclear-grade cation resins in a series I and a series II and the ammonia and potassium selectivity coefficients of the nuclear-grade cation resins; the flow of the first series and the flow of the second series are adjusted, the theoretical ammonia adding amount of the first loop under different flows of the second series is calculated, and continuous ammonia adding is conducted on the first loop with the theoretical ammonia adding amount as the target; when the series of diamino potassium is saturated, sampling and analyzing the total alkali metal content and the ammonia content from the outlets of the primary loop and the series II, and calculating and adjusting the adding amount and the frequency of the potassium hydroxide solution to ensure that the total alkali metal concentration of the main loop is in the area A and does not deviate; after the potassium ammonia saturation of the series II is finished, stopping the operation of the series I, calculating the required ammonia adding amount under the normal working condition, timely adjusting the continuous ammonia adding amount of the loop I, quitting the series I, and keeping the operation of the series II. According to the invention, smooth implementation of saturation of series of diamino potassium and safety of primary loop water quality are ensured.

The invention belongs to the technical field of nuclear power DCS health management, and aims to provide a DCS health management platform for nuclear power plant DCS operation and maintenance business application. The invention discloses a DCS (Distributed Control System) health management platform based on agile development and a general-branch architecture, which comprises a display layer, a business application layer, a model and service layer, an operation and maintenance deployment layer and an infrastructure layer, and is characterized in that intelligent operation and maintenance management, software and hardware state remote monitoring and service life prediction of a nuclear power plant DCS are realized through the business application layer; and a dynamic logic deduction and fault tracing mechanism is constructed, full-process digital operation management and control, overall planning of a maintenance plan, authority auditing and intelligent guiding are carried out, and a full-life-cycle data system is established. According to the invention, DCS operation and maintenance service application is provided for the nuclear power plant, the reliability management of the DCS of the nuclear power plant and the risk analysis level of an instrument control system can be improved, and the probability of unexpected unit operation events caused by the DCS is reduced.

The invention discloses an acoustic-optical-electric multi-dimensional sensing integrated underwater disaster-causing object monitoring system which is suitable for water areas with strict stability requirements such as nuclear power plants. The system comprises a floating body, a control integrated cabin, a solar power supply module and a liftable regulation and control cabin. The regulation and control cabin is driven by a rack-gear mechanism and can drive the acoustic, optical and electrical monitoring modules installed on the outer wall of the regulation and control cabin to dive as a whole, and flexible adjustment of the monitoring depth is achieved. When stormy waves are large, the gravity center of the system can be lowered through the descending regulation and control cabin, and the anti-overturning stability is further enhanced by combining the active water injection function of a water storage cavity in the descending regulation and control cabin. In addition, by coordinating multidirectional drainage of the water storage cavity, active impact resistance and low-speed position fine adjustment can be achieved. Through collaborative design of the structure and control, the stability of the floating body is improved, meanwhile, continuous, stable and collaborative work of acoustic, optical and electric sensors under complex hydrological conditions is guaranteed, and reliable, multi-dimensional and in-situ monitoring of underwater disaster-causing objects is achieved.

This invention discloses a passive spray system for aerosol removal in nuclear power plants, comprising: a condensate collector, a guide pipe, a spray water tank, a spray pipe, a burst valve, and nozzles. The condensate collector is connected to the spray water tank via the guide pipe and is used to collect condensate within the containment. The spray water tank is connected to the nozzles via the spray pipes. The burst valve is installed on the spray pipes. When the pressure inside the containment reaches the activation trigger pressure of the burst valve, the burst valve opens, and water from the spray water tank flows into the nozzles through the spray pipes, where it is sprayed to remove aerosols from the containment. The beneficial effects of this system are as follows: compared with natural removal methods, this system provides more efficient removal of suspended aerosols within the containment; compared with active spray systems, this system does not require pumps and power supplies, and has higher reliability under accident conditions; this system can collect condensate near the walls, providing a more sustainable water source for the spray system.

The invention belongs to the technical field of nuclear power plant subcritical degree determination, and particularly relates to a method for determining a subcritical degree by adopting a spatial effect of an external neutron source. Comprising the following steps: 1, obtaining a curve of the ratio of the subcritical degree to near-source and passive neutron detector signals through theoretical calculation; 2, obtaining measurement values of four source range neutron detectors and four intermediate range neutron detectors; 3, calculating a near-source and passive ratio; and 4, according to the ratio calculated in the step 3, searching a ratio curve of the subcritical degree and the signals of the near-source and passive neutron detectors so as to obtain the reactivity of the reactor core. The method has the beneficial effect that the subcritical degree cannot be obtained by directly measuring through an instrument when the power level of the reactor is lower. According to the method, the subcriticality can be quickly obtained in a chart form through the ratio of the near-source neutron detector signals to the passive neutron detector signals or the ratio of the passive neutron detector signals to the near-source neutron detector signals.

This invention belongs to the field of vibration reduction technology for nuclear power plant fans, specifically relating to a high-damping integrated vibration reduction foundation suitable for nuclear power fan fans. It consists of a steel frame connected to a height-adjustable vibration damper. The steel frame is welded from high-damping composite channel steel, which is composed of conventional channel steel and particulate dampers. The height-adjustable vibration damper uses the housing as the installation reference. An upper annular metal rubber is inserted into its inner cavity, followed by the upper end of the mandrel being inserted into the inner hole of the upper annular metal rubber. Then, a lower annular metal rubber is inserted into the inner cavity of the housing, with its upper end face against the mandrel. A base is then placed on top, and rivets are used to connect the base and housing through four mounting holes. Finally, the internal thread of the mounting block is screwed into the external thread at the upper end of the mandrel. This invention effectively reduces the vibration level of the fan, ensuring safe operation of the fan.

The application discloses a coating structure of a valve rod of a nuclear power plant, a valve rod and a preparation method thereof, and the coating structure comprises a transition layer and a hydrogen-free diamond layer which are sequentially arranged from inside to outside, the microstructure of the transition layer is nanotwin, and the coating structure is uniformly covered on the surface of the valve rod. The preparation method comprises the following steps: after the valve rod is cleaned, the valve rod is subjected to plasma surface activation, the transition layer and the hydrogen-free diamond layer are sequentially deposited, and the valve rod is obtained. The coating structure of the valve rod of the nuclear power plant, the valve rod and the preparation method thereof effectively improve the hardness of the surface of the valve rod, reduce the friction loss of the valve rod, make the valve rod have good wear resistance and a reduced friction coefficient, improve the work efficiency, and further improve the service life of the valve rod, can meet the technical performance requirements of the valve rod on impact toughness, corrosion resistance and wear resistance in a high-temperature and high-pressure water environment, and effectively reduce the production cost.

This invention relates to the field of structural health monitoring of nuclear power plant equipment, and in particular to a method for testing the performance of high-temperature resistant long-wire strain gauges. The method includes: Step 1: Determining the temperature step for strain testing of the strain gauge; Step 2: Installing the strain gauge on a measuring beam; Step 3: Conducting room-temperature performance testing of the strain gauge; Step 4: Increasing the temperature of the high-temperature loading furnace, and repeating Step 3 under different temperature step conditions, respectively measuring and collecting the strain output, strain gauge lead-end resistance, and lead resistance of the strain gauge at each temperature step; Step 5: Calculating the actual strain; Step 6: Calculating the loaded strain; Step 7: Calculating the relative sensitivity coefficient. This invention enables the testing and calibration of the thermal output and sensitivity coefficient of high-temperature resistant long-wire strain gauges as a function of temperature. It provides accurate test data for the accurate testing and calibration of the thermal output and sensitivity coefficient of high-temperature resistant long-wire strain gauges as a function of temperature, and is of great significance.

The utility model relates to a nuclear power plant anemograph comparison device, the nuclear power plant anemograph comparison device comprises a support, an air pipe, a fan, a calibration anemograph and a clamp, the support is provided with a containing cavity, a ventilation hole and an avoiding hole; the air pipe comprises a measuring part, a connecting part and a mounting part which are sequentially connected and communicated to form an air cavity; the measuring part is provided with an air inlet; the pipe wall of the measuring part is provided with a calibration assembling hole and a to-be-measured assembling hole. The connecting part is positioned outside the side wall of the bracket; the mounting part penetrates through the avoiding hole, and an air outlet is formed in the mounting part; the fan is arranged at the air outlet of the mounting part; the calibration anemograph is arranged corresponding to the calibration assembly hole; the clamp is used for clamping a to-be-tested anemograph arranged in the to-be-tested assembly hole in a penetrating mode. Therefore, the length size of the anemograph comparison device of the nuclear power plant can be reduced, and miniaturization is realized; a simulation environment is formed in the air cavity through cooperation of the fan and the air pipe, the anemograph is calibrated to measure the real-time air speed, and the accuracy of comparison of the anemograph to be measured can be improved.

The utility model discloses a nuclear power plant radioactive waste barrel gamma scanning device and measuring equipment, and the nuclear power plant radioactive waste barrel gamma scanning device comprises a supporting seat, a sealing cover, a rotating disc, a control mechanism and a fixing assembly. The sealing cover is detachably mounted on the supporting seat, and a sealing cavity is formed between the sealing cover and the supporting seat; the turntable is detachably mounted on the supporting seat and rotates in the sealing cavity; the fixing assembly is installed on the rotating disc, the radioactive waste barrel is detachably fixed to the rotating disc, and the axis of the radioactive waste barrel is parallel to the plane of the rotating disc. Therefore, the total activity of the radioactive raw material in the radioactive waste barrel can be quickly obtained only through twice measurement, the measurement efficiency and the accuracy of activity reconstruction are improved, the layering frequency is reduced, and the measurement time is saved.

The invention belongs to the technical field of nuclear power plant equipment maintenance, and particularly relates to an underwater electric spark maintenance device based on rigid-flexible coupling positioning in an irradiation environment. Comprising an underwater carrying device, an electric spark machining device, a camera monitoring device and a control system. The underwater carrying device provides carrying and positioning supporting for the electric spark machining device, the electric spark machining device achieves underwater cutting of the instrument casing pipe, the control system is arranged on the water, the underwater cutting operation of the instrument casing pipe is completed through remote operation, and the maintenance process is monitored and recorded in real time through the camera monitoring device. The method is simple to operate and short in consumed time, secondary damage to the lower reactor internals is avoided, and the lower reactor internals are not affected to return to the reactor for reuse.

The invention belongs to the technical field of nuclear power, and particularly relates to an intelligent parameter monitoring method and device for a nuclear power plant. According to the intelligent parameter monitoring method for the nuclear power plant provided by the invention, parameter monitoring under two different working conditions of a steady state and a transient state is realized, different monitoring methods are introduced for different working conditions, and the precision and reliability of parameter monitoring are improved; parameter abnormity can be found in advance, operators can have sufficient intervention and processing time, and safe operation of the nuclear power plant is facilitated; the whole-process dynamic monitoring and abnormity identification of the parameters are realized, the monitoring pressure of operators is reduced, and the working efficiency is improved; the influence of human factors on the monitoring result is reduced, the inherent limitation of people is eliminated, and the parameter monitoring is more comprehensive and timely.

The invention belongs to the technical field of nuclear power plant containment penetration piece isolation valve sealing tests, and particularly relates to a rubber sealing and blocking structure and a test method. Comprising a self-sealing type rubber sealing plug and a pressing type rubber sealing plug and is used for sealing and plugging an outlet of the check valve. The self-sealing type rubber sealing plug comprises first chains, a rubber sealing ring and a plug structural body, the plug structural body is a main body part of the plug, the annular rubber sealing ring is installed in a groove of the plug structural body to play a sealing and plugging role, and the multiple first chains made of metal materials are fastened in the plug structural body. The device has the advantages that a penetration piece swing check valve sealing test method is optimized, a systematic test method is optimized into a valve single body sealing test, and the device has the functions that a check valve rocker arm is not disassembled, and a rubber part can be replaced on site for reuse.

The embodiment of the present application provides a kind of fire automatic alarm method and system, it is applied to the fire automatic alarm system of nuclear power plant, the method comprises: the alarm module obtains fire information;The fire-fighting linkage module collects the equipment parameter of fire-fighting equipment;The control module receives the fire information and the equipment parameter, according to the fire information and the equipment parameter, first control instruction is sent to the fire-fighting linkage module, and the first control instruction is used to control the fire-fighting equipment;After the fire-fighting equipment executes the first control instruction, the fire-fighting linkage module collects the state information of the fire-fighting equipment;The control module receives the state information, and the state information is sent to the DCS module.By control module, the flexibility of system can be improved to the fire automatic alarm system.

The invention provides a training method and device of a simulation test model, and relates to the technical field of nuclear power simulation. The method comprises the following steps: acquiring historical test operation sequence information and historical test result information in a historical test procedure; acquiring feature information of the nuclear power plant simulation system; the historical test operation sequence information and the historical feature information are subjected to feature fusion and then input into a simulation test model, and first test result information is obtained; and training the simulation test model according to the historical test result information and the first test result information. Through training, the simulation test model learns to obtain the capability of predicting the test result corresponding to the test operation sequence information based on the characteristic information of the system, the method can adapt to different states of the nuclear power plant simulation system, and the test result is accurately predicted. The efficiency, accuracy and adaptability of simulation testing of the nuclear power plant simulation system are improved, the intelligent level of testing is improved, and therefore the workload of manual testing is reduced.

The application provides a fire control method and device and electronic equipment, and relates to the technical field of nuclear power, and the method comprises the following steps: in the case that a fire alarm signal sent by a fire detector is received by a fire alarm system, the fire alarm system is used to send the fire alarm signal to a distributed control system (DCS); the fire alarm system is used to receive a device state control instruction sent by the DCS based on the fire alarm signal; and the fire alarm system is used to control the fire equipment to execute a fire operation indicated by the device state control instruction, wherein the fire equipment is the fire equipment indicated by the device state control instruction. According to the application, the fire alarm system is used to control the fire equipment to execute the fire operation, the fire alarm system can execute corresponding fire operations according to the device state control instruction of the DCS, and the fire safety of the nuclear power plant can be improved.

The present application relates to a kind of for nuclear power plant reactor head exhaust control cabinet debugging method, device and system, wherein the debugging method includes: action debugging step: S1-1: receive the action debugging instruction output by the control cabinet, and according to the action debugging instruction trigger simulation exhaust valve module action, to generate position signal;S1-2: the control cabinet receives the position signal and carries out logic processing, to obtain action debugging result;Fault debugging step: S2-1: trigger fault debugging action, and generate corresponding position fault signal;S2-2: the control cabinet receives the position fault signal and carries out logic processing, to obtain fault debugging result;Obtain and display the action debugging result or the fault debugging result.Can realize the test of the logic channel of reactor head exhaust control cabinet, reduce on-site disconnection, reduce the risk of control cabinet damage caused by operation error, test effect is not good, optimizes system available period.

The invention discloses a nuclear power plant accident analysis method and device, and relates to the technical field of nuclear power safety analysis. The method includes: responding to user triggering to perform accident analysis of a first accident type on a nuclear power plant, and acquiring a first input card and a first reference card; the first input card is used for representing an initial analysis condition corresponding to accident analysis of a first accident type of the nuclear power plant, and the first reference card is used for representing an accident verification standard of the first accident type; based on the first input card, calling a first calculation configuration template corresponding to the first accident type to perform accident calculation to obtain a first accident calculation result; the first calculation configuration template is used for representing a calculation process rule of the first accident type; and determining an accident analysis result of the nuclear power plant in the first accident type based on the first accident calculation result and the first reference card. According to the invention, the accident analysis efficiency can be improved.

The invention belongs to the technical field of image processing and segmentation, and particularly relates to a nuclear power plant equipment appearance corrosion defect segmentation method based on an attention mechanism and an edge detection operator. Comprising the following steps: collecting appearance images of nuclear power plant key equipment; performing semantic-level annotation on the acquired image, and constructing a nuclear power equipment corrosion segmentation data set; a TransUnett model is adopted as a basic network framework, and a pre-training weight is obtained by using a data set; an improved TransUnet model is obtained; obtaining a final weight file suitable for nuclear power equipment corrosion defect segmentation; and inputting a to-be-detected nuclear power equipment image into the trained model, outputting a semantic segmentation mask and an edge response result by the model at the same time, and generating a final corrosion defect segmentation result through post-processing operation. The method has the beneficial effects that defect segmentation can be realized by combining semantic segmentation and edge detection at the same time, and the defects of fuzzy edge and inaccurate segmentation of an existing model structure in a complex environment are overcome.

The present application belongs to the technical field of nuclear power engineering design, and provides a nuclear power plant wind pipe support load automatic calculation and analysis method and system, comprising: obtaining data information of a nuclear power plant wind pipe; calling a corresponding three-dimensional support model in a preset wind pipe information library according to the data information of the nuclear power plant wind pipe; automatically calculating the wind pipe support load according to the three-dimensional support model; and automatically completing the mechanical analysis of the wind pipe support load according to the calculated wind pipe support load; the present application can realize the data information of the wind pipe support, automatically call the support model in the wind pipe information library to complete three-dimensional support modeling; according to the three-dimensional support model, the automatic calculation and automatic analysis of the wind pipe support load can be realized, the complex process of drawing, mapping, data extraction and coupling iterative calculation of the designer is cancelled, the cycle of mechanical calculation and analysis of the wind pipe support is shortened, and the accuracy of the calculation result is greatly improved.