A method and system for evaluating the operating state of a fluidized bed unit deep peak shaving process

Abstract

The application discloses a kind of fluidized bed unit depth peak shaving process operating state evaluation method and system, it is related to fluidized bed technical field, this method includes: collecting the combustion operating condition data of fluidized bed unit depth peak shaving process;Particle flow velocity vector diagram is generated, and different flow patterns and paths in particle flow velocity vector diagram are identified and labeled;Through the particle flow velocity and particle concentration distribution of each flow mode area in flow field, the uneven index of entire flow field is generated;Bed temperature distribution diagram is generated, and high temperature-flow mode area is identified in combination with particle flow velocity vector diagram, and the coking risk index of high temperature-flow mode area is calculated;The coking risk index of entire flow field is calculated, and the coking risk index of entire flow field is compared with risk threshold value, when the coking risk index of entire flow field is greater than or equal to risk threshold value, send early warning, identify potential risk, avoid the occurrence of serious problems such as coking.

Description

Technical Field

[0001] This invention relates to the field of fluidized bed technology, specifically to a method and system for evaluating the operational status of a fluidized bed unit during deep peak shaving. Background Technology

[0002] In recent years, new energy power generation such as wind power and photovoltaic power has been developed vigorously. However, the supply of new energy power has inherent characteristics of intermittency, fluctuation and seasonality, which puts forward higher requirements for the grid's regulation and absorption capacity. Fluidized bed units have become an important target for the flexible transformation of thermal power plants due to their advantages such as wide fuel adaptability, high combustion efficiency, low stable combustion load and low pollutant control cost.

[0003] Chinese invention application CN116663776A discloses a method and system for determining the deep peak-shaving capacity of a circulating fluidized bed (CFB) boiler. The method includes: Step 1: Determining the set of main parameters for determining the deep peak-shaving capacity of the CFB boiler unit under deep peak-shaving operation conditions; Step 2: Analyzing and determining the method and range for each parameter; Step 3: Determining the deep peak-shaving capacity of the CFB boiler unit. By establishing a comprehensive determination system, the deep peak-shaving capacity of the CFB boiler can be accurately determined, enabling the CFB unit to better adapt to the requirements of grid-source coordinated and safe operation under future high-proportion renewable energy conditions.

[0004] However, during deep peak shaving operations, the fluid flow pattern undergoes complex changes with the dynamic changes in load. This makes the current monitoring strategies for static and fixed areas inadequate and no longer effective. In particular, when the load decreases, the flow characteristics and distribution of particles inside the bed may become more uneven, and channeling and flow deviation may be significantly enhanced. The direct consequence of this uneven flow is that the particle concentration in local areas increases abnormally, forming so-called "hot spots". These hot spots are very likely to become breeding grounds for coking problems, thus posing a serious threat to the stable operation of the system. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] To address the shortcomings of existing technologies, this invention provides a method and system for evaluating the operational status of a fluidized bed unit during deep peak shaving. By analyzing the particle flow velocity and particle concentration distribution during the deep peak shaving process of the fluidized bed unit, different flow patterns and paths in the particle flow velocity vector diagram are identified and marked. Combined with the bed temperature distribution, the coking risk of the entire flow field is analyzed, potential risks are identified, and serious problems such as coking are avoided.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the present invention provides the following technical solution: a method for evaluating the operational status of a fluidized bed unit's deep peak-shaving process, comprising the following steps:

[0009] Collect combustion condition data during the deep peak shaving process of the fluidized bed unit;

[0010] Acquire particle flow velocity data during the deep peak shaving process of the fluidized bed unit, generate a particle flow velocity vector map, and identify and label different flow patterns and paths in the particle flow velocity vector map;

[0011] Based on the identified flow patterns, the flow field is divided into different flow pattern regions. By analyzing the particle flow velocity and particle concentration distribution in each flow pattern region within the flow field, the non-uniformity index of the entire flow field is obtained.

[0012] Obtain bed temperature data of fluidized bed unit during deep peak shaving process, generate bed temperature distribution map, combine with particle flow velocity vector map, identify high temperature-flow mode region, and calculate coking risk index of high temperature-flow mode region;

[0013] By combining the non-uniformity index of the entire flow field and the coking risk index of all high-temperature-flow mode regions, the coking risk index of the entire flow field is calculated. The coking risk index of the entire flow field is compared with the risk threshold. When the coking risk index of the entire flow field is greater than or equal to the risk threshold, an early warning is issued.

[0014] Furthermore, the collection of combustion condition data includes: setting up several temperature measuring points at different heights and regions of the fluidized bed, using thermocouples to monitor changes in bed temperature; measuring the flow velocity of particles in the fluidized bed using a laser Doppler velocimeter; and using an optical probe to monitor the particle concentration in the fluidized bed in real time.

[0015] Furthermore, the generation of the particle flow velocity vector map includes: mapping the particle flow velocity data onto a grid model, where each grid point corresponds to the position coordinates and velocity vector of different particles; and filling in the missing grid points using a spatial interpolation algorithm to obtain the particle flow velocity vector map.

[0016] Furthermore, different flow patterns and paths in the particle flow velocity vector diagram were identified and labeled, including:

[0017] A local window is used to extract velocity vector features of each grid point and within the window region. The extracted velocity vector features are then used to perform pattern recognition through machine learning algorithms to classify different flow patterns in the flow field.

[0018] The streamline tracing algorithm is used to generate flow paths in the flow field, and the traced flow paths are labeled according to the classification results of the flow patterns.

[0019] Furthermore, for each flow mode region, the average particle flow velocity is calculated. By calculating the ratio between the maximum and minimum values ​​of the average particle flow velocity in all flow mode regions, the non-uniformity index of particle flow velocity in the entire flow field is obtained.

[0020] Furthermore, the particle concentration distribution data of the fluidized bed unit during the deep peak shaving process is obtained. For each flow mode region, the average particle concentration is calculated. By calculating the ratio between the maximum and minimum values ​​of the average particle concentration in all flow mode regions, the non-uniformity index of particle concentration distribution in the entire flow field is obtained.

[0021] The non-uniformity index of the entire flow field is calculated by combining the non-uniformity index of particle flow velocity and the non-uniformity index of particle concentration distribution.

[0022] Furthermore, the generation of the bed temperature distribution map includes: mapping temperature data onto a grid model of the fluidized bed, with each grid point corresponding to a different location coordinate and its temperature value; for grid points with missing temperature data, spatial interpolation algorithms are used to fill in the missing data to obtain the bed temperature distribution map.

[0023] Furthermore, the process of identifying the high-temperature-flow pattern region includes:

[0024] The average temperature and standard deviation of the bed temperature are calculated using the temperature value of each grid point. The sum of the average temperature and standard deviation is used as the temperature threshold. Areas with temperatures higher than the temperature threshold are marked as local high-temperature areas. The bed temperature distribution map and the particle flow velocity vector map are overlaid to determine and mark the flow mode regions corresponding to the local high-temperature areas as high-temperature-flow mode regions.

[0025] Furthermore, the generation of the coking risk index in the high-temperature-flow mode region includes:

[0026] By integrating particle concentration distribution data and bed temperature data for each high-temperature-flow mode region through spatial matching, the average particle concentration and average temperature of the marked high-temperature-flow mode region are calculated, and the product of the average particle concentration and average temperature is used as the coking risk index of that high-temperature-flow mode region.

[0027] Furthermore, the formula for calculating the coking risk index of the entire flow field is as follows:

[0028]

[0029] Where Cr represents the coking risk index of the entire flow field, and U represents the non-uniformity index of the entire flow field. kω represents the coking risk index of the k-th high-temperature-flow mode region. k The weight represents the k-th high-temperature-flow pattern region. The weight is the proportion of the area of ​​the k-th high-temperature-flow pattern region to the total area of ​​all high-temperature-flow pattern regions. k = 1, 2, ..., K, where K represents the number of high-temperature-flow pattern regions.

[0030] An operational status assessment system for the deep peak shaving process of a fluidized bed unit, comprising:

[0031] The data acquisition module collects combustion condition data during the deep peak shaving process of the fluidized bed unit;

[0032] The particle flow analysis module acquires particle flow velocity data during the deep peak shaving process of the fluidized bed unit, generates a particle flow velocity vector map, and identifies and marks different flow patterns and paths in the particle flow velocity vector map.

[0033] The flow field uniformity analysis module divides the flow field into different flow mode regions based on the identified flow patterns. By analyzing the particle flow velocity and particle concentration distribution in each flow mode region within the flow field, the non-uniformity index of the entire flow field is obtained.

[0034] The coking risk assessment module acquires bed temperature data of the fluidized bed unit during the deep peak shaving process, generates a bed temperature distribution map, and, combined with the particle flow velocity vector map, identifies the high-temperature-flow mode region and calculates the coking risk index of the high-temperature-flow mode region.

[0035] The early warning module integrates the non-uniformity index of the entire flow field and the coking risk index of all high-temperature-flow mode regions to calculate the coking risk index of the entire flow field. It then compares the coking risk index of the entire flow field with the risk threshold. When the coking risk index of the entire flow field is greater than or equal to the risk threshold, an early warning is issued.

[0036] (III) Beneficial Effects

[0037] This invention provides a method and system for evaluating the operational status of a fluidized bed unit during deep peak shaving, which has the following beneficial effects:

[0038] (1) By generating a particle flow velocity vector diagram, the flow state and change trend of particles in the fluidized bed can be monitored in real time, and different flow patterns and paths such as eddies, mainstreams and backflows can be identified and marked. Potential uneven fluidization phenomena such as channel flow and deflection can be detected in time, so as to provide early warning and take corresponding measures.

[0039] (2) By using the non-uniformity index of the entire flow field, the non-uniformity of the flow field during the deep peak shaving process of the fluidized bed can be evaluated, including both particle flow velocity and particle concentration distribution. This allows for a more comprehensive understanding of the actual state of the flow field. After dividing the flow field into different flow mode regions, specific analysis can be performed on each region, which helps to accurately identify areas with problems in the flow field, such as areas where particle flow velocity is too fast or too slow, or where particle concentration is too high or too low.

[0040] (3) By mapping the bed temperature data onto the grid model, local high-temperature areas in the fluidized bed can be accurately identified, which helps to discover potential hot spots in a timely manner and take corresponding measures to prevent coking and other adverse situations. By calculating the coking risk index of the high-temperature-flow mode area, the degree of coking risk in the area can be more accurately reflected, so as to make targeted adjustments and optimizations.

[0041] (4) By calculating the coking risk index of the entire flow field, the operating status and potential risks of the fluidized bed unit during the deep peak shaving process can be comprehensively reflected. When the coking risk index exceeds the preset risk threshold, an early warning is issued to remind relevant personnel to take measures in a timely manner, thereby effectively avoiding serious problems such as coking. Attached Figure Description

[0042] Figure 1 This is a schematic diagram of the operational status evaluation method for the deep peak shaving process of the fluidized bed unit according to the present invention;

[0043] Figure 2 This is a schematic diagram of the operational status evaluation system for the deep peak shaving process of the fluidized bed unit of the present invention. Detailed Implementation

[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0045] Please see Figure 1 This invention provides a method for evaluating the operational status of a fluidized bed unit during deep peak shaving, comprising the following steps:

[0046] Step 1: Collect combustion condition data of the fluidized bed unit during the deep peak shaving process, including but not limited to bed temperature, pressure distribution, particle flow velocity, particle concentration distribution, fuel input, air flow, bed height and fluctuation;

[0047] Step one includes:

[0048] Step 101: Arrange several temperature measuring points at different heights and regions of the fluidized bed, and monitor the temperature changes of the bed using thermocouples or infrared thermometers; measure the flow velocity of particles at key sections of the fluidized bed using laser Doppler velocimetry (LDA) or other non-contact velocimetry technologies; and monitor the particle concentration in the fluidized bed in real time using optical probes, capacitance tomography (ECT), and other technologies.

[0049] Step 102: Install flow meters in the fuel input system to record real-time data of fuel input; install flow meters in the air supply system to monitor changes in airflow in different air ducts; and use a bed height measuring device (such as an ultrasonic rangefinder) to obtain real-time changes and fluctuations in bed height.

[0050] It should be noted that when arranging temperature measuring points and selecting velocity and concentration monitoring technologies, the specific structure and operating conditions of the fluidized bed unit must be considered. The location of the temperature measuring points should be able to comprehensively reflect the changes in bed temperature, avoiding the neglect of areas with local overheating or uneven temperature. At the same time, the measurement of particle flow velocity and concentration distribution should cover the entire fluidized bed as much as possible to obtain comprehensive particle motion information.

[0051] Step 103: Set the data acquisition frequency and period. For parameters that require simple measurement methods, such as temperature and pressure, adopt a high-frequency acquisition method, such as acquiring data once or multiple times per second. For parameters that require complex measurement methods, such as particle flow velocity and concentration distribution, set a periodic acquisition plan, such as acquiring data once per hour or adjusting it according to the operating conditions.

[0052] When using this method, refer to the content of steps 101 to 103:

[0053] By collecting key parameters such as bed temperature, pressure distribution, particle flow velocity, and particle concentration distribution, the operating status of the fluidized bed unit during deep peak shaving can be monitored in real time. This helps to promptly identify potential operating problems, such as abnormal temperature, pressure fluctuations, and poor particle flow, and thus take corresponding measures to address them, ensuring the stable operation of the unit.

[0054] Step 2: Obtain particle flow velocity data during the deep peak shaving process of the fluidized bed unit, generate a particle flow velocity vector map, and identify and label different flow patterns and paths in the particle flow velocity vector map;

[0055] Step two includes:

[0056] Step 201: Obtain particle flow velocity data during the deep peak shaving process of the fluidized bed unit, preprocess the particle flow velocity data, including cleaning, removing outliers and noise, map the particle flow velocity data onto a grid model, where each grid point corresponds to the position coordinates and velocity vector (including velocity magnitude and direction) of different particles, and fill in the missing grid points using a spatial interpolation algorithm (such as linear interpolation) to generate a particle flow velocity vector map.

[0057] It should be noted that the grid model is based on the actual size and shape of the fluidized bed, dividing it into several regular grid cells. Each grid point has a specific coordinate position. The collected bed temperature data is mapped onto the grid model, and each grid point corresponds to a temperature value.

[0058] Step 202: Using a local window, extract the velocity vector features of each grid point and within the window region. Use machine learning algorithms (such as support vector machine, decision tree, etc.) to perform pattern recognition on the extracted velocity vector features to classify different flow patterns in the flow field. For example, identify eddies by recognizing the rotation direction and intensity of the velocity vector; identify the main flow direction by tracing the main flow direction of the velocity vector; and identify backflow by detecting velocity vectors in the opposite direction to the main flow direction.

[0059] It should be noted that a local window refers to a small, limited spatial range selected in flow field data analysis for detailed analysis of a specific area. This range contains multiple grid points, each of which corresponds to a specific location in the flow field.

[0060] It should be noted that the mainstream is the main flow path in the flow field. In the velocity vector diagram, the velocity is usually large and the direction is relatively consistent. The mainstream region can be identified by the magnitude and direction distribution of the velocity vector. The vector diagram of the mainstream region is usually represented by continuous arrows with the same direction.

[0061] Vortexes are rotating flows in a flow field. Within a vortex region, the velocity vector will exhibit circular or spiral motion around a certain center. In a velocity vector diagram, the vortex region is represented by a pattern of vector arrows rotating around a certain center. The presence and intensity of vortices can be determined by the direction of rotation and the magnitude of the velocity of the vector arrows.

[0062] Backflow refers to the flow of a portion of the fluid in the flow field in the opposite direction to the mainstream. In the velocity vector diagram, the backflow region is represented by a vector arrow in the opposite direction to the mainstream. These arrows may form a clear reverse flow path or may form a certain angle with the mainstream direction.

[0063] Step 203: Use streamline tracing algorithms (such as streamline and trace algorithms) to generate flow paths in the flow field. According to the classification results of flow patterns, mark the traced flow paths. For example, mark the main flow path with a specific color or line type, mark the eddy path as a closed curve around a certain center, and mark the return flow path as a line segment opposite to the direction of the main flow, etc.

[0064] When using this method, refer to steps 201 to 203:

[0065] By generating a particle flow velocity vector diagram, the flow state and changing trend of particles in the fluidized bed can be monitored in real time. Different flow patterns and paths, such as eddies, mainstreams and backflows, can be identified and marked. Potential non-uniform fluidization phenomena such as channeling and eccentric flow can be detected in a timely manner, thereby providing early warning and taking corresponding measures.

[0066] Step 3: Based on the identified flow patterns, the flow field is divided into different flow pattern regions. The non-uniformity index of the entire flow field is generated by analyzing the particle flow velocity and particle concentration distribution in each flow pattern region.

[0067] Step three includes:

[0068] Step 301: Based on the identified flow patterns, the flow field is divided into different flow pattern regions. For each flow pattern region, the average particle flow velocity is calculated. By calculating the ratio between the maximum and minimum average particle flow velocities across all flow pattern regions, the non-uniformity index of particle flow velocity in the entire flow field is obtained. The calculation formula is as follows:

[0069]

[0070] Among them, V i V represents the average particle flow velocity within the i-th flow mode region, where i = 1, 2, ..., M, and M represents the number of flow mode regions. xj V yj V zj Let x, y, and z represent the velocity components of the j-th measurement point, respectively, j = 1, 2, ..., N, where N represents the number of measurement points, and Uv represents the non-uniformity index of the particle flow velocity in the entire flow field.

[0071] It should be noted that the potential risks of non-uniform flow phenomena such as channeling and deflection are assessed by the magnitude of the particle flow velocity non-uniformity index. Generally, the larger the non-uniformity index, the more uneven the velocity distribution, and the higher the risk of channeling and deflection.

[0072] Step 302: Obtain particle concentration distribution data of the fluidized bed unit during the deep peak shaving process. For each flow mode region, calculate the average particle concentration. By calculating the ratio between the maximum and minimum values ​​of the average particle concentration in all flow mode regions, obtain the non-uniformity index of particle concentration distribution in the entire flow field. The calculation formula is as follows:

[0073]

[0074] Among them, C i C represents the average particle concentration in the i-th flow pattern region, i = 1, 2, ..., M, where M represents the number of flow pattern regions. j Let j represent the particle concentration at the j-th measurement point, where j = 1, 2, ..., N, N represents the number of measurement points, and Uc represents the non-uniformity index of particle concentration distribution throughout the flow field.

[0075] Step 303: Combining the non-uniformity index of particle flow velocity and the non-uniformity index of particle concentration distribution throughout the flow field, the non-uniformity index of the entire flow field is calculated using the geometric mean method; Geometric mean method: Where U represents the non-uniformity index of the entire flow field;

[0076] It should be noted that if the non-uniformity index of particle flow velocity is high, the velocity distribution in the flow field will show a large difference. The velocity in some areas may be much higher than that in other areas. This velocity difference will cause the fluid to generate severe friction and collision during the flow, thereby increasing energy loss and noise. At the same time, the non-uniform velocity may also cause the fluid to separate or vortex during the flow, further reducing the stability and efficiency of the flow field.

[0077] Similarly, if the non-uniformity index of particle concentration is high, the concentration distribution in the flow field will also show a large difference. This concentration difference may lead to a decrease in heat and mass transfer efficiency because the heat and mass transfer rate between areas with high concentration and areas with low concentration will be affected. In addition, non-uniformity of concentration may also cause non-uniformity of chemical reactions.

[0078] When using this method, refer to the content of steps 301 to 303:

[0079] By using the non-uniformity index of the entire flow field, the non-uniformity of the flow field during deep peak shaving in a fluidized bed can be assessed, including both particle flow velocity and particle concentration distribution. This allows for a more comprehensive understanding of the actual state of the flow field. After dividing the flow field into different flow mode regions, specific analysis can be performed on each region, which helps to accurately identify areas with problems in the flow field, such as areas where particle flow velocity is too fast or too slow, or where particle concentration is too high or too low.

[0080] Step 4: Obtain bed temperature data of the fluidized bed unit during the deep peak shaving process, generate a bed temperature distribution map, combine it with the particle flow velocity vector map, identify the high temperature-flow mode region, and generate the coking risk index of the high temperature-flow mode region.

[0081] Step four includes:

[0082] Step 401: Obtain bed temperature data of the fluidized bed unit during the deep peak shaving process, map the temperature data onto the grid model of the fluidized bed, each grid point corresponds to a different location coordinate and its temperature value, and for missing data points, use spatial interpolation algorithms (such as inverse distance weighted interpolation, etc.) to fill in the missing data points and generate a bed temperature distribution map.

[0083] Step 402: Calculate the average temperature and standard deviation of the bed temperature using the temperature value of each grid point. Use the sum of the average temperature and standard deviation as the temperature threshold, where temperature threshold = average temperature + standard deviation. Mark the area with temperature higher than the temperature threshold as a local high temperature area. Overlay the bed temperature distribution map and the particle flow velocity vector map to determine and mark the flow mode area corresponding to the local high temperature area as a high temperature-flow mode area.

[0084] Step 403: Integrate the particle concentration distribution data and bed temperature data of each high-temperature-flow mode region through spatial matching, calculate the average particle concentration and average temperature of the marked high-temperature-flow mode region, and use the product of the average particle concentration and average temperature as the coking risk index of the high-temperature-flow mode region.

[0085] It should be noted that by spatial matching, the particle concentration and temperature data of each high-temperature-flow mode region are matched one by one to form a comprehensive dataset. By matching and integrating two sets of data with different but related attributes (particle concentration distribution data and bed temperature data) under the same spatial framework, we can more accurately understand the relationship between particle concentration and temperature in different regions of the fluidized bed, thereby identifying those regions that have both high particle concentration and high temperature. These regions are usually places with a high risk of coking.

[0086] When using this method, please refer to the content of steps 401 to 403:

[0087] By mapping bed temperature data onto a grid model, local high-temperature regions in the fluidized bed can be accurately identified, which helps to promptly detect potential hotspots and take corresponding measures to prevent coking and other adverse conditions. By calculating the coking risk index of the high-temperature-flow mode region, the degree of coking risk in that region can be more accurately reflected, thereby enabling targeted adjustments and optimizations.

[0088] Step 5: Combine the non-uniformity index of the entire flow field and the coking risk index of all high-temperature-flow mode regions to calculate the coking risk index of the entire flow field. Compare the coking risk index of the entire flow field with the risk threshold. When the coking risk index of the entire flow field is greater than or equal to the risk threshold, issue an early warning.

[0089] Step five includes:

[0090] Step 501: Obtain the non-uniformity index of the entire flow field and the coking risk index of all high-temperature-flow mode regions. Calculate the coking risk index of the entire flow field using the non-uniformity index and the coking risk index of all high-temperature-flow mode regions. The calculation formula is as follows:

[0091]

[0092] Where Cr represents the coking risk index of the entire flow field, and U represents the non-uniformity index of the entire flow field. k ω represents the coking risk index of the k-th high-temperature-flow mode region. k The weight of the k-th high-temperature-flow mode region is given by k = 1, 2, ..., K, where K represents the number of high-temperature-flow mode regions and the weight is the proportion of the area of ​​the k-th high-temperature-flow mode region to the total area of ​​all high-temperature-flow mode regions.

[0093] Step 502: Pre-set a risk threshold and compare the coking risk index of the entire flow field with the risk threshold. When the coking risk index of the entire flow field is greater than or equal to the risk threshold, it indicates that the current operating status of the fluidized bed unit has a high risk of coking. An early warning is issued to remind relevant personnel to take immediate measures to adjust or optimize, such as adjusting parameters such as fuel input, air flow, and bed height, or carrying out necessary maintenance and cleaning work.

[0094] When the coking risk index of the entire flow field is less than the risk threshold, it indicates that the current operating status of the fluidized bed unit is relatively stable and the coking risk is low. The current operating status should continue to be monitored.

[0095] It should be noted that, based on new historical data and expert experience, an initial risk threshold is set. During actual operation, the set risk threshold is verified, and it is observed whether the operating status and coking of the fluidized bed unit meet expectations when the risk threshold is near. As the operating time increases and the status of the fluidized bed unit changes, the risk threshold is updated regularly.

[0096] When using this method, refer to the content of steps 501 to 502:

[0097] By calculating the coking risk index of the entire flow field, the operating status and potential risks of the fluidized bed unit during the deep peak shaving process can be comprehensively reflected. When the coking risk index exceeds the preset risk threshold, an early warning is issued to promptly remind relevant personnel to take measures, thereby effectively avoiding the occurrence of serious problems such as coking.

[0098] Please see Figure 2 The present invention also provides an operational status assessment system for the deep peak shaving process of a fluidized bed unit, comprising: a data acquisition module, a particle flow analysis module, a flow field uniformity analysis module, a coking risk assessment module, and an early warning module;

[0099] Among them, the data acquisition module collects combustion condition data of the fluidized bed unit during the deep peak shaving process;

[0100] The particle flow analysis module acquires particle flow velocity data during the deep peak shaving process of the fluidized bed unit, generates a particle flow velocity vector map, and identifies and marks different flow patterns and paths in the particle flow velocity vector map.

[0101] The flow field uniformity analysis module divides the flow field into different flow mode regions based on the identified flow patterns. By analyzing the particle flow velocity and particle concentration distribution in each flow mode region within the flow field, the non-uniformity index of the entire flow field is obtained.

[0102] The coking risk assessment module acquires bed temperature data of the fluidized bed unit during the deep peak shaving process, generates a bed temperature distribution map, and, combined with the particle flow velocity vector map, identifies the high-temperature-flow mode region and calculates the coking risk index of the high-temperature-flow mode region.

[0103] The early warning module integrates the non-uniformity index of the entire flow field and the coking risk index of all high-temperature-flow mode regions to calculate the coking risk index of the entire flow field. It then compares the coking risk index of the entire flow field with the risk threshold. When the coking risk index of the entire flow field is greater than or equal to the risk threshold, an early warning is issued.

[0104] In the application, the various formulas mentioned are all calculated by removing dimensions and taking their numerical values. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The coefficients in the formulas are set by those skilled in the art according to the actual situation.

[0105] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, and combinations thereof. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0106] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0107] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for operating state assessment of a fluidized bed unit deep peak shaving process, characterized in that: The method comprises the following steps: collecting combustion condition data of the fluidized bed unit during the deep peak shaving process; obtaining particle flow velocity data of the fluidized bed unit during the deep peak shaving process, generating a particle flow velocity vector diagram, and identifying and labeling different flow patterns and paths in the particle flow velocity vector diagram; dividing the flow field into different flow pattern regions according to the identified flow patterns, and generating a non-uniformity index of the entire flow field through the particle flow velocity and particle concentration distribution of each flow pattern region in the flow field; obtaining bed temperature data of the fluidized bed unit during the deep peak shaving process, generating a bed temperature distribution diagram, including: mapping the temperature data onto a grid model of the fluidized bed, each grid point corresponding to different position coordinates and its temperature value, filling the grid points with missing temperature data using a spatial interpolation algorithm to obtain the bed temperature distribution diagram; in combination with the particle flow velocity vector diagram, identifying a high-temperature-flow pattern region, including: calculating the average temperature and standard deviation of the bed temperature through the temperature value of each grid point, taking the sum of the average temperature and the standard deviation as a temperature threshold, marking the region with a temperature higher than the temperature threshold as a local high-temperature region, superimposing the bed temperature distribution diagram and the particle flow velocity vector diagram to determine the flow pattern region corresponding to the local high-temperature region and mark it as a high-temperature-flow pattern region; calculating the coking risk index of the high-temperature-flow pattern region, including: integrating the particle concentration distribution data and the bed temperature data of each high-temperature-flow pattern region through spatial matching, calculating the average particle concentration and the average temperature of the marked high-temperature-flow pattern region, and taking the product of the average particle concentration and the average temperature as the coking risk index of the high-temperature-flow pattern region; comprehensively calculating the coking risk index of the entire flow field from the non-uniformity index of the entire flow field and the coking risk index of all high-temperature-flow pattern regions, comparing the coking risk index of the entire flow field with a risk threshold, and issuing a warning when the coking risk index of the entire flow field is greater than or equal to the risk threshold.

2. The method according to claim 1, wherein the combustion condition data is collected by: arranging a plurality of temperature measuring points at different heights and regions of the fluidized bed, and using thermocouples to monitor the change of the bed temperature; measuring the flow velocity of the particles in the fluidized bed by a laser Doppler velocimeter; and using an optical probe to monitor the particle concentration in the fluidized bed in real time. The particle flow velocity vector diagram is generated by: mapping the particle flow velocity data onto a grid model, each grid point corresponding to different particle position coordinates and its velocity vector, and filling the grid points with missing particle flow velocity data using a spatial difference algorithm to obtain the particle flow velocity vector diagram.

3. The method of claim 1, wherein the method further comprises: determining the operating state of the fluidized bed unit based on the determined operating state of the fluidized bed unit. The different flow patterns and paths in the particle flow velocity vector diagram are identified and labeled by:

4. The method of claim 3, wherein the method further comprises: using a local window to extract the velocity vector features in each grid point and the window region, and classifying different flow patterns in the flow field through machine learning algorithm pattern recognition on the extracted velocity vector features; ​ The flow path in the flow field is generated by using a streamline tracking algorithm, and the flow path tracked is labeled according to the classification result of the flow mode.

5. The method of claim 4, wherein the method further comprises: For each flow mode region, the average particle flow velocity is calculated, and the non-uniformity index of the particle flow velocity in the entire flow field is obtained by calculating the ratio between the maximum value and the minimum value of the average particle flow velocity in all flow mode regions.

6. The method of claim 5, wherein the method further comprises: The particle concentration distribution data of the fluidized bed unit in the deep peak shaving process is obtained, for each flow mode region, the average particle concentration is calculated, and the non-uniformity index of the particle concentration distribution in the entire flow field is obtained by calculating the ratio between the maximum value and the minimum value of the average particle concentration in all flow mode regions; The non-uniformity index of the entire flow field is calculated by the geometric mean method by combining the non-uniformity index of the particle flow velocity and the non-uniformity index of the particle concentration distribution in the entire flow field.

7. The method of claim 1, wherein the method further comprises: determining a current operating state of the fluidized bed unit based on the determined operating state of the fluidized bed unit. The formula for calculating the coking risk index of the entire flow field is as follows: ； wherein, Cr represents the coking risk index of the entire flow field, U represents the unevenness index of the entire flow field, represents the coking risk index of the first k high-temperature flow pattern region, represents the weight of the first k high-temperature flow pattern region, the weight being the proportion of the area of the first k high-temperature flow pattern region to the total area of all high-temperature flow pattern regions, , and K represents the number of high-temperature flow pattern regions.

8. A system for operating state assessment of a fluidized bed unit deep peak shaving process, for implementing the method of any one of claims 1 to 6, characterized in that: The method comprises: a data acquisition module for acquiring combustion condition data of the fluidized bed unit in the deep peak shaving process; a particle flow analysis module for obtaining particle flow velocity data of the fluidized bed unit in the deep peak shaving process, generating a particle flow velocity vector diagram, and identifying and labeling different flow modes and paths in the particle flow velocity vector diagram; a flow field uniformity analysis module for dividing the flow field into different flow mode regions according to the identified flow modes, and obtaining the non-uniformity index of the entire flow field through the particle flow velocity and the particle concentration distribution of each flow mode region in the flow field; a coking risk assessment module for obtaining bed temperature data of the fluidized bed unit in the deep peak shaving process, generating a bed temperature distribution diagram, including: mapping the temperature data onto a grid model of the fluidized bed, each grid point corresponding to different position coordinates and its temperature value, filling the grid points with missing temperature data using a spatial interpolation algorithm to obtain the bed temperature distribution diagram; combining the particle flow velocity vector diagram, identifying the high-temperature-flow mode region, including: calculating the average temperature and the standard deviation of the bed temperature through the temperature value of each grid point, taking the sum of the average temperature and the standard deviation as the temperature threshold, marking the region with a temperature higher than the temperature threshold as a local high-temperature region, superimposing the bed temperature distribution diagram and the particle flow velocity vector diagram to determine the flow mode region corresponding to the local high-temperature region and mark it as the high-temperature-flow mode region; calculating the coking risk index of the high-temperature-flow mode region, including: integrating the particle concentration distribution data and the bed temperature data of each high-temperature-flow mode region through spatial matching, calculating the average particle concentration and the average temperature of the marked high-temperature-flow mode region, and taking the product of the average particle concentration and the average temperature as the coking risk index of the high-temperature-flow mode region; The early warning module integrates the non-uniformity index of the entire flow field and the coking risk index of all high-temperature-flow mode regions, calculates the coking risk index of the entire flow field, compares the coking risk index of the entire flow field with a risk threshold, and issues a warning when the coking risk index of the entire flow field is greater than or equal to the risk threshold.

Images