A method for monitoring anti-blocking of a molten salt heat storage system of a thermal power plant

Abstract

The present disclosure provides a kind of power plant molten salt heat storage system anti-blocking monitoring method, by fusing distributed temperature measurement optical fiber, pressure sensor and acoustic vibration sensor, real-time acquisition system key position temperature field, pressure and vibration and other multidimensional data;Further calculate and fuse temperature gradient, temperature drop rate, flow resistance coefficient and other key state characteristic quantities, construct comprehensive blockage risk index quantitative model for intelligent analysis.The method realizes early, sensitive perception to molten salt flow state and solidification risk, can change from traditional single parameter-dependent lag alarm to active early warning and hierarchical response based on multi-source information fusion and quantitative evaluation, ultimately achieves the purpose of identifying risk signs before blockage actually occurs, guiding precise intervention, effectively preventing system shutdown and equipment damage, improving the safety and reliability of system operation, realizing the fundamental change from "after treatment" to "prevention".

Description

Technical Field

[0001] This invention relates to the field of energy technology and power systems, and in particular to a method for monitoring and preventing blockage in a molten salt thermal storage system of a thermal power plant. Background Technology

[0002] Molten salt thermal energy storage systems in thermal power plants typically refer to molten salt thermal energy storage devices coupled to the traditional coal-fired power plant side. They are used to enable flexible operation of the unit, peak shaving, and time shifting of energy. Through charging and heat storage, when the grid load is low, electrical energy or extracted steam is used to heat the molten salt and store the energy in a high-temperature molten salt tank. When the grid load is high, the high-temperature molten salt and water steam exchange heat in a heat exchanger to generate superheated steam, which drives the steam turbine to generate electricity, thereby increasing the unit output.

[0003] Coupling molten salt thermal energy storage technology with thermal power units is an important means to improve grid flexibility. However, this system faces a common and severe challenge in actual operation—the freezing and blockage of molten salt in pipelines and equipment. Molten salt (usually a mixture of nitrates) has a high freezing point, typically between 120°C and 220°C. Once the local temperature of the system drops below this freezing point, the molten salt will solidify, leading to serious consequences, such as system shutdown, blockage of pipelines or equipment interrupting the circulation of molten salt, making the entire heat storage or release process impossible, resulting in system failure; damage to equipment, blockage can cause abnormal pressure increases in pipelines, pump overload operation, and may even cause physical damage to valves, pipelines, or heat exchangers; maintenance is difficult and costly, unblocking solidified molten salt is extremely difficult, requiring a lot of manpower and time for external heating and cleaning, resulting in huge economic losses and power generation losses.

[0004] Currently, the methods for monitoring and preventing molten salt blockage in engineering are relatively simple and outdated. The main problems are as follows: relying on a single temperature monitoring method, the most common method is to install point thermocouples or resistance temperature detectors on the outer wall of the pipe to monitor the temperature at a certain point. This can only reflect the temperature at the installation point and cannot sense the flow state and cross-sectional temperature distribution inside the pipe. When the point temperature measurement shows an abnormality, a serious blockage has often already formed inside, resulting in a delayed warning. In addition, it cannot distinguish between "low-temperature flow" and "non-flowing but not yet solidified" states. Summary of the Invention

[0005] The first aspect of this disclosure provides a method for monitoring and preventing blockage in a molten salt thermal storage system of a thermal power plant, comprising the following steps: S1: Real-time acquisition of monitoring data from preset locations in the molten salt thermal storage system, wherein the monitoring data includes at least temperature field data and pressure data; S2: Based on the monitoring data, calculate at least one state characteristic quantity to characterize the molten salt flow state and solidification risk; S3: Compare and analyze the state feature quantity with a preset threshold or model; S4: When the results of the comparison analysis meet the preset blockage risk conditions, generate and output the corresponding level of anti-blockage warning information.

[0006] In conjunction with the first aspect, in step S1, the temperature field data is acquired by a distributed temperature-measuring optical fiber laid along the outer wall of the molten salt pipe to obtain the axial continuous real-time temperature distribution of the molten salt pipe.

[0007] In conjunction with the first aspect, in step S2, the state characteristic quantity includes the temperature gradient along the axial direction of the molten salt pipeline. The temperature gradient is obtained by calculating the ratio of the temperature difference to the distance between adjacent monitoring points. When the absolute value of the temperature gradient exceeds a first preset threshold, it is determined that there is a risk in the corresponding area.

[0008] In conjunction with the first aspect, in step S2, the state characteristic quantity includes the axial temperature drop rate of the molten salt pipeline. The temperature drop rate is the decrease in temperature at a specific point or area per unit time. When the temperature drop rate exceeds a second preset threshold, the risk of blockage is determined to be increased.

[0009] In conjunction with the first aspect, in step S2, the state characteristic quantity includes the flow resistance coefficient K of the molten salt thermal storage system. The flow resistance coefficient K is based on the inlet and outlet pressure difference ΔP of the pump and the real-time volumetric flow rate Q of the molten salt, expressed by the formula... Calculations show that when the rate of change of the flow resistance coefficient K relative to the reference value exceeds a third preset threshold, the internal flow condition of the system is determined to be abnormal.

[0010] In conjunction with the first aspect, in step S1, the monitoring data also includes vibration signals of pipes and equipment collected by acoustic vibration sensors. In step S2, the vibration signals are subjected to spectral analysis to identify characteristic frequency changes caused by changes in flow state, and the analysis results are used as auxiliary characteristic quantities characterizing the flow state.

[0011] In conjunction with the first aspect, in step S4, the anti-blocking warning information includes multiple levels; Among them, the first-level warning corresponds to local temperature abnormalities or abnormal temperature drop rates, and it is recommended to conduct insulation checks. A Level 2 warning corresponds to increased flow resistance or the presence of a significant low-temperature area, and it is recommended to adjust operating parameters or enhance local heating. A Level 3 warning indicates that a congestion is imminent or has already occurred, and it is recommended to follow emergency procedures.

[0012] In conjunction with the first aspect, the method further includes a predictive maintenance step: Based on historical operating data, ambient temperature, and unit load data, machine learning models are used to predict the probability of blockage in different parts of the system within a set future time period and generate preventative operation recommendations.

[0013] A second aspect of this disclosure provides an electronic device, comprising: One or more processors; A storage unit is used to store one or more programs, which, when executed by one or more processors, enable the one or more processors to implement the anti-blocking monitoring method for the molten salt thermal storage system of a thermal power plant.

[0014] A third aspect of this disclosure provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, enables the anti-blocking monitoring method for the molten salt thermal storage system of a thermal power plant.

[0015] Beneficial Effects: This disclosure provides a method for monitoring and preventing blockage in a molten salt thermal power plant. By integrating distributed temperature-measuring optical fibers, pressure sensors, and acoustic vibration sensors, it collects multi-dimensional data such as temperature field, pressure, and vibration from key parts of the system in real time. It then calculates and integrates key state characteristics such as temperature gradient, temperature drop rate, and flow resistance coefficient to construct a comprehensive blockage risk index quantitative model for intelligent analysis. This method achieves early and sensitive awareness of the molten salt flow state and solidification risk, transforming traditional delayed alarms relying on single parameters into proactive early warning and graded response based on multi-source information fusion and quantitative assessment. Ultimately, it achieves the goal of identifying risk signs and guiding precise intervention before actual blockage occurs, effectively preventing system shutdowns and equipment damage, improving the safety and reliability of system operation, and realizing a fundamental shift from "post-event handling" to "pre-event prevention." Attached Figure Description

[0016] Figure 1 This is a schematic flowchart of a method for monitoring and preventing blockage in a molten salt thermal storage system of a thermal power plant, according to an embodiment of this disclosure. Figure 2 An electronic device according to an embodiment of this disclosure. Detailed Implementation

[0017] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with those disclosed herein.

[0018] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the present disclosure. The singular forms “a,” “the,” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0019] Figure 1 A deep learning model defense method based on feature layer adaptive denoising, as described in this disclosure, includes the following steps: S1: Real-time acquisition of monitoring data from preset locations in the molten salt thermal storage system, wherein the monitoring data includes at least temperature field data and pressure data; S2: Based on the monitoring data, calculate at least one state characteristic quantity to characterize the molten salt flow state and solidification risk; S3: Compare and analyze the state feature quantity with a preset threshold or model; S4: When the results of the comparison analysis meet the preset blockage risk conditions, generate and output the corresponding level of anti-blockage warning information.

[0020] Specifically, step S1: Real-time collection of monitoring data.

[0021] This step forms the basis of risk perception. Sensor networks are deployed at key locations in the molten salt thermal storage system to collect data continuously and synchronously.

[0022] Monitoring locations: The “preset locations” or “critical locations” include, but are not limited to: the inlet and outlet of the molten salt circulation pump, the main circulation pipeline (especially elbows, valves, low points and other easily blocked parts), the molten salt side inlet and outlet of the molten salt-steam heat exchanger, and the outlet pipelines of the high-temperature molten salt tank and the low-temperature molten salt tank.

[0023] Monitoring data content: Temperature field data: A distributed fiber optic temperature sensing system (DTS) is preferred for data acquisition. Specifically, the sensing fiber is tightly laid along the axial direction of the molten salt pipe (e.g., secured with metal armor or thermally conductive adhesive) to measure the continuous spatial temperature distribution on the pipe surface. This system can provide temperature data per second over a range of several kilometers along the fiber path with a spatial resolution down to the meter level, thus forming a continuous temperature distribution curve along the pipe's axis and accurately locating "cold spots" or temperature anomaly zones.

[0024] Pressure data: Acquired through pressure transmitters installed at the aforementioned key locations (especially pump inlets and outlets, and both ends of main pipe sections). The pressure data is used to reflect the overall pressure drop and local resistance changes of the system.

[0025] Optional auxiliary data: To further enhance the comprehensiveness of monitoring, acoustic vibration sensors can also be integrated to collect vibration signals from pipelines and equipment to help determine the flow status.

[0026] Step S2: Calculate the state characteristic quantities.

[0027] This step involves extracting information from the raw data, transforming direct physical measurements into characteristic indicators that directly reflect the system's health status and risk level.

[0028] Feature type: Based on the data collected by S1, calculate one or more of the following state features: Temperature gradient ( T): Based on the axial temperature distribution curve obtained from distributed temperature measurement, calculate the ratio of temperature difference to distance between adjacent measurement points. For example, for DTS data with a spatial resolution of 1 meter, the temperature change per meter of pipe length (°C / m) can be calculated. Significant abrupt increases in temperature gradient usually indicate insulation failure or the presence of flow dead zones at that location.

[0029] Temperature drop rate (-ΔT / Δt): Calculates the temperature drop per unit time (e.g., per minute) for a specific monitoring point or area (such as a valve group). This rate is particularly critical during system shutdowns, sudden load drops, or heat tracing failures.

[0030] Flow resistance coefficient (K): This is a comprehensive characteristic quantity reflecting the overall hydraulic properties of the system. Based on the pump inlet and outlet pressure difference ΔP (obtained from pressure data) and the real-time volumetric flow rate Q of the molten salt (which can be measured by a flow meter or estimated through pump characteristic curves and operating parameters), it is calculated using the formula... Calculations were performed. A continuous increase in the K value suggests that there may be fouling deposits inside the system or that the flow cross-section may have decreased due to localized solidification.

[0031] Calculation frequency: The calculation of the above-mentioned features is carried out simultaneously with data acquisition, usually at the second or minute level, in order to achieve real-time or near real-time monitoring.

[0032] Step S3: Comparison and analysis.

[0033] This step involves comparing the calculated feature values ​​with preset criteria to quantify and assess the risk.

[0034] Preset threshold: Set one or more security thresholds for each state feature.

[0035] For example, for temperature gradient T sets the first threshold G_set (e.g., 10℃ / m), sets the second threshold R_set (e.g., 5℃ / min) for the rate of temperature drop, and sets the third threshold (e.g., 20% higher than the normal operating baseline value) for the rate of change of the flow resistance coefficient K.

[0036] These thresholds can be determined comprehensively based on system design parameters, molten salt properties (freezing point), historical operating data, and safety margins, and can be optimized and adjusted during system debugging and long-term operation.

[0037] Analysis Model (Optional Enhancement): In addition to simple threshold comparisons, more complex analysis models can be used: Statistical model: Based on historical normal operation data, establish the statistical distribution of each characteristic quantity (such as mean and standard deviation), and use control charts or sigma criteria to determine whether the current value deviates from the normal range.

[0038] Machine learning models: These models are trained using historical data (including normal and abnormal operating conditions). The current multidimensional features are input into the model, and the model directly outputs the probability or level of congestion risk.

[0039] Blockage risk conditions: When a single feature exceeds its threshold, or when multiple features combine to meet specific logical conditions (such as "temperature gradient exceeds limit" and "temperature drop rate exceeds limit"), or when the risk value output by the analysis model reaches the threshold, it is determined that the "preset blockage risk conditions" are met.

[0040] Step S4: Generate and output warning information.

[0041] This step transforms the analysis results into decision-making information that operations and maintenance personnel can understand and implement.

[0042] Warning Information Generation: Once step S3 determines that the risk conditions are met, the system automatically generates structured warning information. This information includes at least: the location of the risk (e.g., pipeline meter markers based on DTS positioning), the main characteristic quantities that triggered the warning and their values, and the current risk level.

[0043] Tiered early warning: Early warning information is output in tiers based on the severity of the risk, typically divided into three levels: Level 1 Warning (Alert / Attention Level): For example, an alert may appear stating "Local temperature is low at 150-155 meters of pipeline XX, approaching the lower safety limit" or "The temperature drop rate at valve YY is slightly higher than normal." Recommended actions include "Checking the insulation layer in this area" or "Confirming the operating status of the heat tracing system."

[0044] Level 2 Warning (Warning / Operational Level): For example, warnings such as "The overall system flow resistance coefficient continues to rise, increasing by 15% compared to the baseline value" or "The temperature at point ZZ is XX°C below the freezing point, and the temperature drop is accelerating." Recommended measures include "Appropriately increasing the circulation pump frequency to increase flow rate flushing," "Activating the backup heat tracing for this area," or "Preparing for online flushing."

[0045] Level 3 Warning (Severe / Emergency): For example, the alarm "The temperature gradient at point AA has increased sharply, and the temperature has fallen below the freezing point, suggesting a possible localized solidification blockage." Recommended actions include "Immediately execute the emergency shutdown procedure," "Activate the highest level of electric heat tracing," and "Notify maintenance personnel to prepare for emergency handling" to prevent the blockage from worsening and causing equipment damage.

[0046] Information output: The generated early warning information is transmitted to the operation and duty personnel in real time through various means such as human-machine interface (HMI), monitoring and data acquisition (SCADA) system alarm window, industrial SMS platform, and mobile application push.

[0047] Through the closed-loop execution of steps S1 to S4 above, this method constructs a complete monitoring and early warning chain from data perception, feature extraction, intelligent diagnosis to decision support, realizing effective, early and precise prevention and control of blockage risks in molten salt thermal storage systems of thermal power plants.

[0048] Furthermore, in step S1, the temperature field data is acquired by a distributed temperature-measuring optical fiber laid along the outer wall of the molten salt pipe to obtain the axial continuous real-time temperature distribution of the molten salt pipe.

[0049] The temperature field data is collected through a distributed optical fiber temperature measurement system tightly laid along the outer wall of the molten salt pipe. Specifically, high-temperature resistant and corrosion-resistant armored temperature measurement cables are selected and firmly attached to the outer wall of the pipe using specialized metal clamps or high-temperature thermally conductive adhesive, ensuring good thermal contact between the optical fiber and the pipe wall. This system is based on the principle of backscattering Raman scattering in optical fibers. When a laser pulse propagates in the fiber, the intensity of its backscattered light is directly related to the temperature at the fiber's location. By analyzing the scattered light signal using demodulation equipment, the temperature value at each point along the fiber path (i.e., the pipe axis) can be calculated in real time, thus obtaining a spatially continuous and temporally updated temperature distribution curve. The advantage of this method is that it achieves full-coverage temperature monitoring of the entire pipe "line" rather than the traditional "point," with spatial resolution reaching meter-level or even sub-meter-level. It can capture "cold spots" formed by local insulation layer damage, abnormal heat loss, or fluid stagnation without omission, providing high-precision basic temperature field data for early risk identification.

[0050] In step S2, the state characteristic quantity includes the temperature gradient along the axial direction of the molten salt pipeline. The temperature gradient is obtained by calculating the ratio of the temperature difference to the distance between adjacent monitoring points. When the absolute value of the temperature gradient exceeds a first preset threshold, it is determined that there is a risk in the corresponding area.

[0051] The temperature gradient is a key characteristic representing the severity of axial temperature changes in the pipeline. Based on continuous temperature distribution data obtained from a distributed fiber optic temperature measurement system, the system selects adjacent monitoring points at fixed spatial intervals (e.g., every 1 meter or corresponding to the inherent spatial resolution of the fiber). Temperature gradient ( The formula for calculating T is: T = (T2 - T1) / L, where T1 and T2 are the temperatures of two adjacent monitoring points, and L is the actual distance between the two points. The calculated gradient value is a vector, and its absolute value reflects the rate of temperature change. The first preset threshold (G_set) is set based on the pipeline insulation design standards, the solidification point of molten salt, and safety margin, for example, 10°C / m. When the absolute value of the temperature gradient of a certain section of the pipeline calculated by the system in real time continuously exceeds G_set, it indicates that there is abnormal heat loss in that section of the pipeline or that the internal fluid is close to stagnation, with the heat dissipation rate far exceeding the normal flow state. Therefore, it is determined that there is a high risk of local solidification in this area, which requires close attention.

[0052] In step S2, the state characteristic quantity includes the axial temperature drop rate of the molten salt pipeline. The temperature drop rate is the decrease in temperature at a specific point or area per unit time. When the temperature drop rate exceeds a second preset threshold, the risk of blockage is determined to be increased.

[0053] The temperature drop rate is a sensitive indicator for assessing the rate of risk evolution under dynamic operating conditions. The system records the temperature value at fixed time intervals (e.g., every minute) for specific key monitoring points (such as valves, low-temperature points) or a defined area (such as a section of pipeline). The temperature drop rate (-ΔT / Δt) is obtained by calculating the difference between the current temperature and the temperature at the previous time point and dividing by the time interval, i.e., (-ΔT / Δt) = (T_(t-1) - T_t) / Δt, where Δt is the sampling period. The second preset threshold (R_set) is determined based on the molten salt properties, the system's allowable cooling rate, and anti-condensation operating procedures, for example, set to 5°C / min. When the system detects a temperature drop rate exceeding R_set at a certain point, it indicates that the temperature at that point is rapidly approaching the freezing point, possibly due to a sudden system shutdown, heat tracing failure, or extreme environmental cooling. This characteristic provides a race-against-time warning signal, prompting immediate intervention to prevent the temperature from dropping to a dangerous level in a short period.

[0054] In step S2, the state characteristic quantity includes the flow resistance coefficient K of the molten salt thermal storage system. The flow resistance coefficient K is based on the inlet and outlet pressure difference ΔP of the pump and the real-time volumetric flow rate Q of the molten salt, and is expressed by the formula... Calculations show that when the rate of change of the flow resistance coefficient K relative to the reference value exceeds a third preset threshold, the internal flow condition of the system is determined to be abnormal.

[0055] The flow resistance coefficient K is a macroscopic state quantity characterizing the hydraulic properties of the entire molten salt circulation loop. Its implementation relies on the real-time measurement of two key parameters: the pump inlet and outlet pressure difference ΔP, obtained by a high-precision pressure transmitter; and the molten salt volumetric flow rate Q, obtained by an electromagnetic flowmeter or by conversion from the pump's speed and power characteristic curves. The system operates according to the formula... The K-value is calculated in real time. First, a baseline K0 value is calculated and stored when the system is clean and operating stably. A third preset threshold is typically set as the limit for the rate of change of the K-value relative to the baseline value, for example, 20%. When the real-time calculated K-value continues to rise, and its rate of change (K-K0) / K0 exceeds this threshold, it indicates a significant increase in system circulation resistance. This is not caused by changes in flow rate, but rather suggests possible fouling deposits on the inner wall of the pipe, slight localized solidification leading to a reduction in the flow cross-section, or blockage by foreign objects. This characteristic provides an indirect but effective assessment of the health of the internal flow channels from the perspective of overall system dynamics.

[0056] In step S1, the monitoring data also includes vibration signals of pipes and equipment collected by acoustic vibration sensors. In step S2, the vibration signals are subjected to spectrum analysis to identify characteristic frequency changes caused by changes in flow state, and the analysis results are used as auxiliary characteristic quantities to characterize the flow state.

[0057] To compensate for the shortcomings of temperature and pressure monitoring, acoustic vibration monitoring is added as an auxiliary dimension. Broadband acoustic vibration sensors are installed on key pipe sections such as pump outlets, elbows, heat exchanger interfaces, and the pump body. These sensors collect the sound and vibration signals generated by the structural vibration of the pipeline / equipment. In step S2, the system performs a fast Fourier transform on the collected raw time-domain vibration signal, converting it into a frequency domain spectrum. Anomalies are identified by comparing the current spectrum with the baseline spectrum characteristics established under normal, unobstructed flow conditions. For example, when the molten salt flow rate decreases or stagnation occurs, the energy of specific frequency bands (such as low-frequency bands related to fluid pulsation) will weaken; while when solid particles begin to appear in the fluid or local initial condensation occurs, the particles impacting the pipe wall will generate specific high-frequency impact signals, and their spectral energy will increase. The system uses a preset algorithm or model (such as a neural network) to quantify and score these characteristic frequency changes, outputting an "anomaly score S," which serves as an auxiliary feature input to the subsequent comprehensive risk assessment model, thereby improving the detection sensitivity for flow stagnation and early phase changes.

[0058] In step S4, the anti-blocking warning information includes multiple levels; Among them, the first-level warning corresponds to local temperature abnormalities or abnormal temperature drop rates, and it is recommended to conduct insulation checks. A Level 2 warning corresponds to increased flow resistance or the presence of a significant low-temperature area, and it is recommended to adjust operating parameters or enhance local heating. A Level 3 warning indicates that a congestion is imminent or has already occurred, and it is recommended to follow emergency procedures.

[0059] The tiered early warning mechanism aims to achieve differentiated and precise risk management. The triggering conditions, information content, and handling recommendations for each level of early warning are as follows: Level 1 Warning (Indication Level): Typically triggered by a slight exceedance of a single characteristic quantity, such as a temperature at a certain point below the lower limit of normal operation but still above the freezing point with a certain safety margin, or a temperature drop rate briefly exceeding the threshold. The warning message clearly identifies the abnormal location and parameter, and suggests that operators "inspect the integrity of the insulation layer in this area" or "confirm the status of electric heat tracing operation," serving as a preventative inspection reminder.

[0060] Level 2 Warning (Alert): Triggered by multiple abnormal characteristics simultaneously or a single characteristic significantly exceeding the limit, such as the rate of change of the flow resistance coefficient K continuously exceeding the threshold, or the temperature at a certain point approaching the freezing point with a high rate of temperature drop. The warning message will clearly indicate the escalation of risk and recommend proactive operational intervention measures such as "increasing the circulation pump frequency to increase the flow rate for flushing," "starting the backup heat tracing circuit for this pipe section," or "preparing to execute an online hot brine flushing procedure."

[0061] Level 3 Warning (Emergency Level): Triggered by strong characteristic signals indicating impending or actual blockage, such as a temperature below the molten salt freezing point with an extreme temperature gradient, or a surge in flow resistance causing the pump to nearly shut down. The warning message is given the highest priority and automatically linked to the emergency plan, recommending "immediately execute the emergency pump shutdown and system isolation procedure," "activate the maximum power emergency heat tracing," and "notify the maintenance team for emergency intervention" to prevent the accident from escalating and causing permanent equipment damage.

[0062] The method also includes a predictive maintenance step: Based on historical operating data, ambient temperature, and unit load data, machine learning models are used to predict the probability of blockage in different parts of the system within a set future time period and generate preventative operation recommendations.

[0063] Specifically, the system continuously collects and stores historical datasets, including time series of all state characteristics (temperature gradient, temperature drop rate, drag coefficient K, vibration score), ambient temperature, unit load plans, start-up and shutdown operation records, etc. Using this historical data, a time series prediction machine learning model (e.g., a Long Short-Term Memory network, LSTM) is trained. This model learns the evolution of state characteristics of various parts of the system under different ambient temperatures, different unit operating load patterns, and different operational histories. In practical applications, based on the current and recent system status, combined with known future unit load plans and weather forecasts, the model predicts the probability of risk characteristics (such as excessively low temperatures or increased drag) appearing in key parts of the system within the next few hours to tens of hours. Based on the prediction results, the system can generate preventative operation suggestions in advance, such as: "An ambient temperature drop is expected at 3:00 AM; it is recommended to increase the heat tracing power of pipe section XX in advance at 1:00 AM"; or "Based on future peak-shaving plans, it is recommended to perform preventative hot flushing of the inlet section of heat exchanger YY after the next discharge process." This shifts the maintenance mode from post-incident inspection and in-process handling to pre-incident prediction and planning.

[0064] Electronic device 200 can be a desktop computer, laptop, handheld computer, cloud server, or other electronic device. Electronic device 200 may include, but is not limited to, processor 201 and memory 202. Those skilled in the art will understand that... Figure 2 This is merely an example of electronic device 200 and does not constitute a limitation on electronic device 200. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device may also include input / output devices, network access devices, buses, etc.

[0065] The processor 201 can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor.

[0066] The memory 202 can be an internal storage unit of the electronic device 200, such as a hard disk or RAM of the electronic device 200. The memory 202 can also be an external storage device of the electronic device 200, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card equipped on the electronic device 200. Furthermore, the memory 202 can include both internal and external storage units of the electronic device 200. The memory 202 is used to store the computer program 203 and other programs and data required by the electronic device. The memory 202 can also be used to temporarily store data that has been output or will be output.

[0067] The above embodiments are only used to illustrate the technical solutions of this disclosure, and are not intended to limit it. Although this disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this disclosure, and should all be included within the protection scope of this disclosure.

Claims

1. A method for monitoring and preventing blockage in a molten salt thermal storage system of a thermal power plant, characterized in that, Includes the following steps: S1: Real-time acquisition of monitoring data from preset locations in the molten salt thermal storage system, wherein the monitoring data includes at least temperature field data and pressure data; S2: Based on the monitoring data, calculate at least one state characteristic quantity to characterize the molten salt flow state and solidification risk; S3: Compare and analyze the state feature quantity with a preset threshold or model; S4: When the results of the comparison analysis meet the preset blockage risk conditions, generate and output the corresponding level of anti-blockage warning information.

2. The method for monitoring and preventing blockage in a molten salt thermal storage system of a thermal power plant according to claim 1, characterized in that, In step S1, the temperature field data is acquired by a distributed temperature-measuring optical fiber laid along the outer wall of the molten salt pipe to obtain the axial continuous real-time temperature distribution of the molten salt pipe.

3. The method for monitoring and preventing blockage in a molten salt thermal storage system of a thermal power plant according to claim 2, characterized in that, In step S2, the state characteristic quantity includes the temperature gradient along the axial direction of the molten salt pipeline. The temperature gradient is obtained by calculating the ratio of the temperature difference to the distance between adjacent monitoring points. When the absolute value of the temperature gradient exceeds a first preset threshold, it is determined that there is a risk in the corresponding area.

4. The method for monitoring and preventing blockage in a molten salt thermal storage system of a thermal power plant according to claim 2, characterized in that, In step S2, the state characteristic quantity includes the axial temperature drop rate of the molten salt pipeline. The temperature drop rate is the decrease in temperature at a specific point or area per unit time. When the temperature drop rate exceeds a second preset threshold, the risk of blockage is determined to be increased.

5. The method for monitoring and preventing blockage in a molten salt thermal storage system of a thermal power plant according to claim 1, characterized in that, In step S2, the state characteristic quantity includes the flow resistance coefficient K of the molten salt thermal storage system. The flow resistance coefficient K is based on the inlet and outlet pressure difference ΔP of the pump and the real-time volumetric flow rate Q of the molten salt, and is expressed by the formula... Calculations show that when the rate of change of the flow resistance coefficient K relative to the reference value exceeds a third preset threshold, the internal flow condition of the system is determined to be abnormal.

6. The method for monitoring and preventing blockage in a molten salt thermal storage system of a thermal power plant according to claim 1, characterized in that, In step S1, the monitoring data also includes vibration signals of pipes and equipment collected by acoustic vibration sensors. In step S2, the vibration signals are subjected to spectrum analysis to identify characteristic frequency changes caused by changes in flow state, and the analysis results are used as auxiliary characteristic quantities to characterize the flow state.

7. The method for monitoring and preventing blockage in a molten salt thermal storage system of a thermal power plant according to claim 1, characterized in that, In step S4, the anti-blocking warning information includes multiple levels; Among them, the first-level warning corresponds to local temperature abnormalities or abnormal temperature drop rates, and it is recommended to conduct insulation checks. A Level 2 warning corresponds to increased flow resistance or the presence of a significant low-temperature area, and it is recommended to adjust operating parameters or enhance local heating. A Level 3 warning indicates that a congestion is imminent or has already occurred, and it is recommended to follow emergency procedures.

8. The method for monitoring and preventing blockage in a molten salt thermal storage system of a thermal power plant according to claim 1, characterized in that, It also includes predictive maintenance steps: Based on historical operating data, ambient temperature, and unit load data, machine learning models are used to predict the probability of blockage in different parts of the system within a set future time period and generate preventative operation recommendations.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method as described in any one of claims 1 to 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method as described in any one of claims 1 to 8.

Images