A synergistic control method and system for circulating fluidized bed flue gas purification

By combining multi-dimensional feature analysis of operating parameters, acoustic emission signals, and infrared thermographic data, the problem of comprehensiveness and accuracy in the state assessment of circulating fluidized bed flue gas purification systems was solved, enabling multi-dimensional assessment and optimized control of the purification system and improving purification performance.

CN121742278BActive Publication Date: 2026-11-10新疆华电米东热电有限公司
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202511721089.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-11-21
Publication Date
2026-11-10
Estimated Expiration
2045-11-21

AI Technical Summary

Technical Problem

Existing circulating fluidized bed flue gas purification systems lack comprehensiveness and accuracy in condition assessment, and cannot effectively integrate multi-source information with different physical characteristics, resulting in a lag in the assessment of the overall health status of the system and insufficient optimization control.

Method used

By acquiring the operating parameters, acoustic emission signals, and infrared thermal imaging data of the purification system, performing projection processing and multidimensional probability distribution model analysis, determining the regional health index, and generating control strategies for coordinated control of the system.

Benefits of technology

This enables a comprehensive and accurate multi-dimensional assessment of the purification system, improving purification performance and system stability, and ensuring the foresight and effectiveness of the control strategy.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121742278B_ABST
    Figure CN121742278B_ABST
Patent Text Reader

Abstract

The application discloses a kind of synergic control method and system for circulating fluidized bed flue gas purification, to solve the problem of lack of comprehensiveness and accuracy of prior art to the state evaluation of circulating fluidized bed flue gas purification system.Therein, the method comprises: obtaining the first operating parameter set, the first acoustic emission signal set and the first infrared thermal image data set of the purification system;The first operating parameter set, the first acoustic emission signal set and the first infrared thermal image data set are projected to obtain the first feature data;According to the multi-dimensional probability distribution model constructed, the first feature data is filtered to obtain the first distribution region;In the first distribution region, based on the second operating parameter set, the second acoustic emission signal set and the second infrared thermal image data set, the regional health degree index corresponding to the first distribution region is determined;According to the regional health degree index, the control strategy of the first distribution region is determined, and the control instruction of the first distribution region is generated to control the purification system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of flue gas purification technology, and in particular to a collaborative control method and system for flue gas purification in circulating fluidized beds. Background Technology

[0002] Circulating fluidized bed flue gas purification technology is a highly efficient method for controlling acidic gaseous pollutants, widely used in industrial fields such as waste incineration power generation and coal-fired boilers. Its basic principle is to establish a fluidized bed within a reaction tower using circulating fly ash, and to activate the absorbent by injecting atomized water, allowing the acidic components in the flue gas to fully contact and react with the absorbent. This technology has significant advantages such as a compact system and no secondary wastewater pollution.

[0003] However, in practical engineering applications, existing circulating fluidized bed flue gas purification systems typically focus only on operating parameters, failing to effectively integrate multi-source information with different physical properties. For example, phenomena such as fluidization uniformity, wall scaling, or localized wet walls within the reaction tower are difficult to reflect accurately and promptly using traditional pressure and temperature measurement points. This results in a one-sided and delayed assessment of the overall "health status" of the system, failing to provide forward-looking guidance for optimized control. Summary of the Invention

[0004] This application provides a collaborative control method for circulating fluidized bed flue gas purification, which addresses the problem that existing technologies lack comprehensiveness and accuracy in assessing the state of circulating fluidized bed flue gas purification systems.

[0005] This application also provides a collaborative control system for circulating fluidized bed flue gas purification.

[0006] The embodiments of this application adopt the following technical solutions: In a first aspect, this application provides a synergistic control method for circulating fluidized bed flue gas purification, comprising: Acquire the first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal images of the purification system; The first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal images are projected to obtain the first feature data. Based on the constructed multidimensional probability distribution model, the first feature data is filtered to obtain the first distribution region; In the first distribution area, the regional health index corresponding to the first distribution area is determined based on the second set of operating parameters, the second set of acoustic emission signals, and the second set of infrared thermal imaging data. Based on the regional health index, the control strategy for the first distribution area is determined, and control instructions for the first distribution area are generated to control the purification system.

[0007] Optionally, the method includes: acquiring a third set of operating parameters, a third set of acoustic emission signals, and a third set of infrared thermal images of the purification system in a healthy operating state, and constructing a health status data matrix; The health status data matrix is ​​standardized to obtain the covariance matrix, and the covariance matrix is ​​decomposed to obtain the projection matrix. By using a projection matrix, the health status data matrix is ​​projected onto a low-dimensional health feature space to obtain the projected second feature data. Based on the second feature data, a multidimensional probability distribution model is obtained; the multidimensional probability distribution model is used to describe the health status of the purification system.

[0008] Optionally, based on the constructed multidimensional probability distribution model, the first feature data is filtered to obtain a first distribution region, including: Calculate the Mahalanobis distance of the first feature data in the multidimensional probability distribution model and the health probability value belonging to the historical health state; Based on the Mahalanobis distance and health probability value, the first feature data is filtered to determine the first distribution area.

[0009] Optionally, the calculation of the Mahalanobis distance of the first feature data in the multidimensional probability distribution model and the health probability value belonging to the historical health state includes: The formula for calculating the Mahalanobis distance of the first feature data in a multidimensional probability distribution model includes:

[0010] in, Let be the Maharanobis distance, which follows a chi-square distribution with n degrees of freedom. This is the first feature data. This is the average of all the second feature data.

[0011] Optionally, the calculation of the Mahalanobis distance of the first feature data in the multidimensional probability distribution model and the health probability value belonging to the historical health state includes: The formulas for calculating the health probability of the first feature data in a multidimensional probability distribution model include:

[0012] in, It is the cumulative distribution function of the chi-square distribution.

[0013] Optionally, the second set of operating parameters includes the calcium-to-acid ratio and the deacidification efficiency value; the second set of acoustic emission signals includes the reaction tower pressure difference; the second set of infrared thermal imaging data includes the reaction tower outlet temperature; in the first distribution area, based on the second set of operating parameters, the second set of acoustic emission signals, and the second set of infrared thermal imaging data, the regional health index corresponding to the first distribution area is determined, including: Based on the calcium-acid ratio and deacidification efficiency, a normal distribution was used to evaluate and determine the deacidification efficiency sub-evaluation value. Based on the pressure difference in the reaction tower, a normal distribution assessment was used to determine the fluidization stability sub-assessment value; Based on the reaction tower outlet temperature, a normal distribution assessment is used to determine the temperature field suitability sub-assessment value. The regional health index corresponding to the first distribution region is obtained by using the offset weight, deacidification efficiency sub-evaluation value, fluidization stability sub-evaluation value, and temperature field suitability sub-evaluation value corresponding to the first distribution region.

[0014] Secondly, this application provides a collaborative control system for circulating fluidized bed flue gas purification, the system comprising: The data acquisition unit is used to acquire the first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal images of the purification system. The feature processing unit is used to perform projection processing on the first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal image data to obtain the first feature data. The filtering unit is used to filter the first feature data according to the constructed multidimensional probability distribution model to obtain the first distribution region; The determining unit is used to determine the regional health index corresponding to the first distribution area based on the second set of operating parameters, the second set of acoustic emission signals, and the second set of infrared thermal imaging data in the first distribution area. The control command generation unit is used to determine the control strategy of the first distribution area based on the regional health index, and generate control commands for the first distribution area to control the purification system.

[0015] Optionally, the system also includes building blocks, accordingly, The data acquisition unit is also used to acquire the third set of operating parameters, the third set of acoustic emission signals, and the third set of infrared thermal images of the purification system in a healthy operating state, and to construct a health status data matrix. The feature processing unit is also used to standardize the health status data matrix to obtain a covariance matrix, and decompose the covariance matrix to obtain a projection matrix; through the projection matrix, the health status data matrix is ​​projected onto a low-dimensional health feature space to obtain the projected second feature data. The construction unit is used to obtain a multidimensional probability distribution model based on the second feature data; wherein the multidimensional probability distribution model is used to describe the health status of the purification system.

[0016] Optionally, this filtering unit is specifically used for: Calculate the Mahalanobis distance of the first feature data in the multidimensional probability distribution model and the health probability value belonging to the historical health state; Based on the Mahalanobis distance and health probability value, the first feature data is filtered to determine the first distribution area.

[0017] Optionally, the second set of operating parameters includes the calcium-to-acid ratio and the deacidification efficiency value; the second set of acoustic emission signals includes the reaction tower pressure difference; the second set of infrared thermal imaging data includes the reaction tower outlet temperature; the determination unit is specifically used for: Based on the calcium-acid ratio and deacidification efficiency, a normal distribution was used to evaluate and determine the deacidification efficiency sub-evaluation value. Based on the pressure difference in the reaction tower, a normal distribution assessment was used to determine the fluidization stability sub-assessment value; Based on the reaction tower outlet temperature, a normal distribution assessment is used to determine the temperature field suitability sub-assessment value. The regional health index corresponding to the first distribution region is obtained by using the offset weight, deacidification efficiency sub-evaluation value, fluidization stability sub-evaluation value, and temperature field suitability sub-evaluation value corresponding to the first distribution region.

[0018] The above-described technical solutions adopted in the embodiments of this application can achieve the following beneficial effects: The method provided in this application provides the following steps: A first set of operating parameters, a first set of acoustic emission signals, and a first set of infrared thermal images of a purification system can be obtained. These parameters are then projected to obtain first feature data. Based on a constructed multidimensional probability distribution model, the first feature data is filtered to obtain a first distribution region. Within this first distribution region, a regional health index is determined based on a second set of operating parameters, a second set of acoustic emission signals, and a second set of infrared thermal images. Based on the regional health index, a control strategy for the first distribution region is determined, and control commands for the first distribution region are generated to control the purification system. This method integrates the multidimensional features of the first operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal images, combined with a constructed multidimensional probability distribution model, to filter the first feature data globally, i.e., to perform a first-stage health status assessment of the purification system. Then, based on the results of the first-stage health status assessment, i.e., the determined first distribution region, a local multidimensional performance assessment is performed on that region to obtain a regional health index. Finally, a control strategy is determined based on the determined regional health index, and control commands are generated. The above scheme, which combines global screening and local evaluation to determine the health index, can improve the problem that existing technologies only focus on operating parameters and fail to effectively integrate multi-source information with different physical characteristics, thus failing to achieve a comprehensive and accurate multi-dimensional evaluation of the purification system. This improves the purification performance of the circulating fluidized bed flue gas purification system. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A schematic diagram illustrating the implementation process of a collaborative control method for circulating fluidized bed flue gas purification provided in an embodiment of this application; Figure 2 This application provides a schematic diagram of the specific structure of a collaborative control system for circulating fluidized bed flue gas purification. Detailed Implementation To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] As will be known to those skilled in the art, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0021] The terms "first," "second," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such terms can be used interchangeably where appropriate; this is merely a way of distinguishing objects with the same attributes in the embodiments of this application.

[0022] The technical solutions provided by the various embodiments of this application are described in detail below with reference to the accompanying drawings.

[0023] Example 1 To address the lack of comprehensiveness and accuracy in the status assessment of circulating fluidized bed flue gas purification systems in existing technologies, this application provides a collaborative control system method for circulating fluidized bed flue gas purification.

[0024] Specifically, the implementation flow of the method provided in this application embodiment is as follows: Figure 1 As shown, it includes the following steps: Step 11: Obtain the first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal images of the purification system.

[0025] In this embodiment, the first set of operating parameters includes: reaction tower differential pressure, inlet and outlet temperatures, flue gas flow rate, absorbent feed rate, and process water flow rate. The first set of acoustic emission signals includes: high-frequency acoustic emission sensors installed at key locations such as the lower conical section, venturi tube region, and top of the reaction tower to collect high-frequency stress wave signals generated by particle collisions and airflow shearing within the fluidized bed. The first set of infrared thermal imaging data consists of periodically scanning the outer wall of the reaction tower with an infrared thermal imager to obtain a temperature field distribution map of the tower wall and extracting the highest, lowest, and average temperatures of characteristic areas. The acquisition points of the acquisition devices corresponding to the first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal imaging data are arranged as follows: First set of operating parameters collected: Reactor tower inlet: flue gas flow meter, temperature sensor, pressure sensor; Middle of the reaction tower: Differential pressure transmitter; Reactor tower outlet: temperature sensor, SO2 online analyzer, HCl online analyzer; Absorbent system: screw feeder speed feedback, silo level gauge; Process water system: flow meter, pressure sensor.

[0026] First set of transmitted signals: Conical section of the reaction tower: 2 sensors, arranged symmetrically at 180°, 1.5 meters from the bottom of the tower; Venturi region: 1 sensor, installed 0.5 meters above the Venturi tube; Top section of the reaction tower: 1 sensor, installed 2 meters down from the top of the tower.

[0027] First infrared thermal image dataset: Two thermal imagers were installed 3-5 meters outside the reaction tower, covering the upper and lower parts of the tower respectively.

[0028] Step 12: Project the first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal images to obtain the first feature data.

[0029] Specifically, step 12 is performed based on a projection matrix that has already been trained using historical health data (e.g., obtained through PCA principal component analysis). This projection matrix is ​​used to project real-time acquired multimodal data (operational parameters, acoustic emission signals, infrared thermography) from a high-dimensional space to a low-dimensional feature space for subsequent health status assessment.

[0030] The specific steps are as follows: Data preprocessing: The first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal images acquired in real time are preprocessed in the same way as the historical training data, including data cleaning and normalization.

[0031] Feature extraction: Features are extracted from each dataset to form a feature vector. For example: Operating parameter set: This might include reaction tower pressure difference, inlet and outlet temperatures, flue gas flow rate, absorbent feed rate, etc., which are directly used as features. Acoustic emission signal set: Time-domain features (such as root mean square value, peak value, amplitude, etc.) and frequency-domain features (such as center frequency, band energy, etc.) are extracted from the acoustic emission signals. Infrared thermal image dataset: Statistical features (such as average temperature, maximum temperature, temperature distribution standard deviation, etc.) and texture features (such as gray-level co-occurrence matrix features, etc.) are extracted from infrared images.

[0032] Feature fusion: The three feature vectors above are concatenated into a high-dimensional feature vector.

[0033] Projection processing: The projection matrix obtained by training with historical data is used to project the high-dimensional feature vectors into a low-dimensional space to obtain the first feature data.

[0034] The projection matrix is ​​obtained from historical health data, including: a. collecting a large amount of multimodal data under historical health states, which is then preprocessed and feature extracted to form a high-dimensional feature vector set. b. analyzing this historical feature vector set to obtain the projection matrix, and selecting the top k principal components whose cumulative contribution rate reaches a certain requirement (e.g., 85%). Therefore, the projection processing of real-time data involves multiplying the real-time feature vectors by the projection matrix to obtain the dimensionality-reduced feature data (i.e., the first feature data).

[0035] After obtaining the first feature data, a pre-constructed multidimensional probability distribution model (e.g., a Gaussian distribution constructed in the principal component space based on historical health data) will be used to calculate the degree of deviation between the current state and the health state, thereby conducting a health status assessment.

[0036] In this embodiment of the application, the specific steps for constructing a multidimensional probability distribution model include: Acquire the third set of operating parameters, the third set of acoustic emission signals, and the third set of infrared thermal images of the purification system in a healthy operating state, and construct a health status data matrix; The health status data matrix is ​​standardized to obtain the covariance matrix, and the covariance matrix is ​​decomposed to obtain the projection matrix. By using a projection matrix, the health status data matrix is ​​projected onto a low-dimensional health feature space to obtain the projected second feature data. Based on the second feature data, a multidimensional probability distribution model is obtained; the multidimensional probability distribution model is used to describe the health status of the purification system.

[0037] Specifically, the third operating parameter set, the third acoustic emission signal set, and the third infrared thermal image set are multimodal data of the system under healthy operating conditions, corresponding to the operating parameter set, acoustic emission signal set, and infrared thermal image set, respectively. Features are extracted from the collected third operating parameter set, third acoustic emission signal set, and third infrared thermal image set. The feature vectors corresponding to each dataset are then concatenated to obtain a high-dimensional feature vector. The feature vectors from multiple time points are combined into a matrix, with each row representing a sample and each column representing a feature. Each column of the matrix is ​​standardized so that the mean of each feature is 0 and the variance is 1, resulting in a standardized matrix. The covariance matrix is ​​calculated from the standardized matrix. Eigenvalue decomposition or singular value decomposition is performed on the covariance matrix. Based on the preset cumulative variance contribution rate, the first k principal components of the left singular vector matrix of the covariance matrix are selected to form a projection matrix. Based on this projection matrix, the standardized healthy sample data matrix is ​​projected onto the principal component space to obtain the dimensionality-reduced second feature data.

[0038] In the health feature space corresponding to the k-dimensional second feature data, the distribution of health status is described by a multidimensional normal distribution. Its probability density function is:

[0039] Wherein, the mean vector μ is the mean of each column of the second feature data (a k-dimensional vector, ideally close to a zero vector), and the covariance matrix Σ is the covariance matrix of the second feature data (k × k dimensions).

[0040] Step 13: Based on the constructed multidimensional probability distribution model, the first feature data is filtered to obtain the first distribution region.

[0041] In this embodiment of the application, step 13 includes: Calculate the Mahalanobis distance of the first feature data in the multidimensional probability distribution model and the health probability value belonging to the historical health state; Based on the Mahalanobis distance and health probability value, the first feature data is filtered to determine the first distribution area.

[0042] Specifically, the formula for calculating the Mahalanobis distance of the first feature data in the multidimensional probability distribution model includes:

[0043] in, Let be the Maharanobis distance, which follows a chi-square distribution with n degrees of freedom. For the first feature data, The average of all the second feature data. Let be the covariance matrix of the second feature data.

[0044] Accordingly, the formula for calculating the health probability of the first feature data in the multidimensional probability distribution model includes:

[0045] in, It is the cumulative distribution function of the chi-square distribution.

[0046] Based on the determined different health probability values, the health probability value interval to which the first feature data belongs can be identified, and the health probability value interval to which it belongs can be determined as the first distribution area.

[0047] This application divides regions based on health probability values, for example: Zone 1: Excellent zone (H>= 0.85); Zone 2: Good zone (0.70<= H<0.85); Zone 3: Warning zone (0.50<= H<0.70); Zone 4: Risk zone (0.30<= H<0.50); ​​Zone 5: Danger zone (H<0.30).

[0048] Step 14: In the first distribution area, based on the second set of operating parameters, the second set of acoustic emission signals, and the second set of infrared thermal images, determine the regional health index corresponding to the first distribution area.

[0049] In this embodiment of the application, step 14 includes: The second set of operating parameters includes the calcium-to-acid ratio and deacidification efficiency; the second set of acoustic emission signals includes the reaction tower pressure difference; the second set of infrared thermal imaging data includes the reaction tower outlet temperature; in the first distribution area, based on the second set of operating parameters, the second set of acoustic emission signals, and the second set of infrared thermal imaging data, the regional health index corresponding to the first distribution area is determined, including: Based on the calcium-acid ratio and deacidification efficiency, a normal distribution was used to evaluate and determine the deacidification efficiency sub-evaluation value. Based on the pressure difference in the reaction tower, a normal distribution assessment was used to determine the fluidization stability sub-assessment value; Based on the reaction tower outlet temperature, a normal distribution assessment is used to determine the temperature field suitability sub-assessment value. The regional health index corresponding to the first distribution region is obtained by using the offset weight, deacidification efficiency sub-evaluation value, fluidization stability sub-evaluation value, and temperature field suitability sub-evaluation value corresponding to the first distribution region.

[0050] Specifically, the first set of operating parameters includes the second set of operating parameters, the first set of acoustic emission signals includes the second set of acoustic emission signals, and the first set of infrared thermal images includes the second set of infrared thermal images. In step 14, some features are selected from the first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal images to serve as the second set of operating parameters, the second set of acoustic emission signals, and the second set of infrared thermal images, respectively, and are used to calculate the regional health index.

[0051] Based on the above embodiments, the regional health index is a further subdivision of the health index of the determined first distribution area. The determined first distribution area can determine the health status of the purification system at a global level, while the determined regional health index quantifies the health status of key monitoring points at a local level. Through health status assessments at different levels, the comprehensiveness of the purification system's supervision can be improved, ensuring the accuracy of the assessment results.

[0052] In this embodiment of the application, determining the regional health index corresponding to the first distribution area specifically includes: Step 1401: Determine the sub-evaluation values ​​for deacidification efficiency, fluidization stability, and temperature field suitability.

[0053] (1) Determine the sub-evaluation value of deacidification efficiency Specifically, the deacidification efficiency sub-evaluation value is determined by weighting the calcium-to-acid ratio score and the deacidification efficiency score.

[0054] Accordingly, the calcium-to-acid ratio scoring formula is as follows:

[0055] in, The calcium-to-acid ratio is scored. It is a normal distribution function. This is the actual calcium-acid ratio. To set the optimal calcium-to-acid ratio, This represents the standard deviation of the calcium-acid ratio.

[0056] The formula for scoring deacidification efficiency is as follows:

[0057] in, To score the deacidification efficiency, It is a normal distribution function. For actual deacidification efficiency, To achieve the optimal deacidification efficiency, The standard deviation of deacidification efficiency.

[0058] The formula for evaluating deacidification efficiency is as follows:

[0059] in, This is a sub-evaluation value for deacidification efficiency. and As the weight, satisfying .

[0060] (2) Determine the fluidization stability sub-evaluation value Specifically, the fluidization stability sub-evaluation value is determined by the pressure difference in the reaction tower.

[0061] Accordingly, the formula for the fluidization stability sub-evaluation value is as follows:

[0062] in, This is the sub-evaluation value for fluidization stability. It is a normal distribution function. This represents the actual pressure difference in the reaction tower. Set the optimal reaction tower pressure differential. This represents the standard deviation of the pressure difference in the reaction tower.

[0063] (3) Determine the sub-evaluation value of temperature field suitability Specifically, the temperature field suitability assessment value is determined by the reaction tower outlet temperature.

[0064] Accordingly, the formula for the sub-evaluation value of the temperature field fitness is as follows:

[0065] in, This is a sub-evaluation value for the suitability of the temperature field. It is a normal distribution function. This represents the actual temperature difference at the outlet of the reaction tower. The optimal reaction tower outlet temperature was set. This represents the standard deviation of the reaction tower outlet temperature.

[0066] Step 1402: Obtain the regional health index corresponding to the first distribution region by using the offset weight, deacidification efficiency sub-evaluation value, fluidization stability sub-evaluation value and temperature field suitability sub-evaluation value corresponding to the first distribution region.

[0067] Specifically, when the system is in a healthy region (e.g., excellent or good), because the system is stable, the sub-evaluation value better reflects the current state, and we can trust the sub-evaluation value more. Accordingly, w can be set larger, close to 1. When the system is in an unhealthy region (e.g., risk or danger), because the system is unstable, the sub-evaluation value may be unreliable, so we trust the baseline health probability more. Accordingly, w is set smaller, close to 0. The offset weight w ranges from 0 to 1.

[0068] A mapping relationship can be preset in the control system. For example: Area: Excellent area: w = 0.8; Good area: w = 0.7; Warning area: w = 0.5; Risk area: w = 0.3; Dangerous area: w = 0.1.

[0069] In this embodiment, firstly, based on the above mapping relationship, the offset weight corresponding to the first distribution area is determined. Then, the health probability value corresponding to the first distribution area is used as the basic health probability. Combined with the deacidification efficiency sub-evaluation value, fluidization stability sub-evaluation value and temperature field suitability sub-evaluation value, a comprehensive sub-evaluation value is determined to obtain the regional health index corresponding to the first distribution area.

[0070] Accordingly, the formula for the comprehensive sub-evaluation value is as follows:

[0071] The formula for the regional health index is as follows:

[0072] Step 15: Based on the regional health index, determine the control strategy for the first distribution area and generate control instructions for the first distribution area to control the purification system.

[0073] Specifically, the control strategy mainly involves coordinated control across three dimensions: absorbent feed optimization, process water spray optimization, and induced draft fan operation optimization. Below is a set of examples of the generated control strategies.

[0074] When the first distribution region is an excellent region, the absorbent control strategy includes: Control mode: Economic optimization mode, Target calcium-to-acid ratio: 1.05-1.15 Adjustment method: Gradual fine-tuning. Adjustment range: ±2-5%, Adjust the frequency: every 5-10 minutes. Optimization algorithm: Gradient descent method to find the point of minimum cost. Safety constraints: {SO2 emission limits: Ensure compliance with emission standards,} Lower limit of deacidification efficiency: Maintain ≥98%, Maximum adjustment rate: to prevent bed disturbance. Process water control strategies include: Control mode: Precise temperature control Target temperature range: 145-155°C Atomization quality: Striving for optimal atomization effect. Water volume regulation: Minimum spray volume based on heat balance calculations. Measures to prevent damp walls: Strengthen monitoring of tower wall temperature. Optimization objective: Minimize water volume while preventing wall wetting. The induced draft fan control strategy includes: Control mode: Differential pressure stabilization control Target pressure differential: Maintain the design optimum value. Regulation strategy: Feedforward + feedback composite control. Optimization objective: To reduce the rotational speed while maintaining stable fluidization. Anti-collapse measures: Set a pressure differential safety margin. The method provided in this application provides the following steps: A first set of operating parameters, a first set of acoustic emission signals, and a first set of infrared thermal images of a purification system can be obtained. These parameters are then projected to obtain first feature data. Based on a constructed multidimensional probability distribution model, the first feature data is filtered to obtain a first distribution region. Within this first distribution region, a regional health index is determined based on a second set of operating parameters, a second set of acoustic emission signals, and a second set of infrared thermal images. Based on the regional health index, a control strategy for the first distribution region is determined, and control commands for the first distribution region are generated to control the purification system. This method integrates the multidimensional features of the first operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal images, combined with a constructed multidimensional probability distribution model, to filter the first feature data globally, i.e., to perform a first-stage health status assessment of the purification system. Then, based on the results of the first-stage health status assessment, i.e., the determined first distribution region, a local multidimensional performance assessment is performed on that region to obtain a regional health index. Finally, a control strategy is determined based on the determined regional health index, and control commands are generated. The above scheme, which combines global screening and local evaluation to determine the health index, can improve the problem that existing technologies only focus on operating parameters and fail to effectively integrate multi-source information with different physical characteristics, thus failing to achieve a comprehensive and accurate multi-dimensional evaluation of the purification system. This improves the purification performance of the circulating fluidized bed flue gas purification system.

[0075] Example 2 To address the lack of comprehensiveness and accuracy in the status assessment of circulating fluidized bed flue gas purification systems in existing technologies, this application provides a collaborative control system for circulating fluidized bed flue gas purification. A schematic diagram of the specific structure of this control system is shown below. Figure 2 As shown, it includes a data acquisition unit 21, a feature processing unit 22, a filtering unit 23, a determination unit 24, and a control command generation unit 25. The functions of each unit are as follows: Data acquisition unit 21 is used to acquire the first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal image data of the purification system; Feature processing unit 22 is used to perform projection processing on the first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal image data to obtain first feature data; The filtering unit 23 is used to filter the first feature data according to the constructed multidimensional probability distribution model to obtain the first distribution area; The determining unit 24 is used to determine the regional health index corresponding to the first distribution area based on the second set of operating parameters, the second set of acoustic emission signals, and the second set of infrared thermal images in the first distribution area. The control command generation unit 25 is used to determine the control strategy of the first distribution area based on the regional health index, and generate control commands for the first distribution area to control the purification system.

[0076] Optionally, the system also includes a building unit 20, accordingly, The data acquisition unit 21 is also used to acquire the third set of operating parameters, the third set of acoustic emission signals, and the third set of infrared thermal images of the purification system in a healthy operating state, and to construct a health status data matrix. The feature processing unit 22 is also used to standardize the health status data matrix to obtain a covariance matrix, and decompose the covariance matrix to obtain a projection matrix; through the projection matrix, the health status data matrix is ​​projected onto a low-dimensional health feature space to obtain the projected second feature data. The construction unit 20 is used to obtain a multidimensional probability distribution model based on the second feature data; wherein the multidimensional probability distribution model is used to describe the health status of the purification system.

[0077] Optionally, the filtering unit 23 is specifically used for: Calculate the Mahalanobis distance of the first feature data in the multidimensional probability distribution model and the health probability value belonging to the historical health state; Based on the Mahalanobis distance and health probability value, the first feature data is filtered to determine the first distribution area.

[0078] Optionally, the second set of operating parameters includes the calcium-to-acid ratio and the deacidification efficiency value; the second set of acoustic emission signals includes the reaction tower pressure difference; the second set of infrared thermal imaging data includes the reaction tower outlet temperature; the determination unit 24 is specifically used for: Based on the calcium-acid ratio and deacidification efficiency, a normal distribution was used to evaluate and determine the deacidification efficiency sub-evaluation value. Based on the pressure difference in the reaction tower, a normal distribution assessment was used to determine the fluidization stability sub-assessment value; Based on the reaction tower outlet temperature, a normal distribution assessment is used to determine the temperature field suitability sub-assessment value. The regional health index corresponding to the first distribution region is obtained by using the offset weight, deacidification efficiency sub-evaluation value, fluidization stability sub-evaluation value, and temperature field suitability sub-evaluation value corresponding to the first distribution region.

[0079] The control system provided in this application integrates multi-dimensional features from first operating parameters, a first acoustic emission signal set, and a first infrared thermal image dataset, combined with a constructed multi-dimensional probability distribution model, to globally filter the first feature data, i.e., to perform a first-stage health status assessment of the purification system. Then, based on the results of the first-stage health status assessment, i.e., the determined first distribution area, a local multi-dimensional performance assessment is performed on that area to obtain a regional health index. Finally, a control strategy is determined based on the determined regional health index, and control commands are generated. This approach, which combines global filtering and local assessment to determine the health index, overcomes the problem in existing technologies that focus only on operating parameters and fail to effectively integrate multi-source information with different physical characteristics, thus failing to achieve a comprehensive and accurate multi-dimensional assessment of the purification system. This improves the purification performance of the circulating fluidized bed flue gas purification system.

[0080] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0081] The above description is merely a specific implementation example of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0082] Secondly, those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0083] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0084] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0085] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0086] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0087] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0088] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0089] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0090] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A synergistic control method for flue gas purification in a circulating fluidized bed, characterized in that, include: Acquire the first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal images of the purification system; The first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal images are projected to obtain the first feature data. Acquire the third set of operating parameters, the third set of acoustic emission signals, and the third set of infrared thermal images of the purification system in a healthy operating state, and construct a health status data matrix; The health status data matrix is ​​standardized to obtain a covariance matrix, and the covariance matrix is ​​decomposed to obtain a projection matrix. The health status data matrix is ​​projected onto a low-dimensional health feature space using the projection matrix to obtain the projected second feature data. Based on the second feature data, a multidimensional probability distribution model is obtained; wherein, the multidimensional probability distribution model is used to describe the health status of the purification system; Based on the constructed multidimensional probability distribution model, the first feature data is filtered to obtain the first distribution region; In the first distribution area, the regional health index corresponding to the first distribution area is determined based on the second set of operating parameters, the second set of acoustic emission signals, and the second set of infrared thermal imaging data. Based on the regional health index, a control strategy for the first distribution area is determined, and control instructions for the first distribution area are generated to control the purification system. The step of filtering the first feature data according to the constructed multidimensional probability distribution model to obtain the first distribution region includes: Calculate the Mahalanobis distance of the first feature data in the multidimensional probability distribution model and the health probability value belonging to the historical health state; Based on the Mahalanobis distance and the health probability value, the first feature data is filtered to determine the first distribution area; The calculation of the Mahalanobis distance of the first feature data in the multidimensional probability distribution model and the health probability value belonging to the historical health state includes: The formula for calculating the Mahalanobis distance of the first feature data in the multidimensional probability distribution model includes: in, Let be the Maharanobis distance, which follows a chi-square distribution with n degrees of freedom. For the first feature data, The average of all the second feature data; The calculation of the Mahalanobis distance of the first feature data in the multidimensional probability distribution model and the health probability value belonging to the historical health state includes: The formula for calculating the health probability of the first feature data in the multidimensional probability distribution model includes: in, It is the cumulative distribution function of the chi-square distribution.

2. The method as described in claim 1, characterized in that, The second set of operating parameters includes the calcium-to-acid ratio and the deacidification efficiency value; the second set of acoustic emission signals includes the reaction tower pressure difference; the second set of infrared thermal imaging data includes the reaction tower outlet temperature; in the first distribution area, based on the second set of operating parameters, the second set of acoustic emission signals, and the second set of infrared thermal imaging data, the regional health index corresponding to the first distribution area is determined, including: Based on the calcium-acid ratio and the deacidification efficiency value, a normal distribution evaluation is used to determine the deacidification efficiency sub-evaluation value. Based on the pressure difference in the reaction tower, a normal distribution assessment is used to determine the fluidization stability sub-assessment value; Based on the outlet temperature of the reaction tower, a normal distribution assessment is used to determine the temperature field suitability sub-assessment value. The regional health index corresponding to the first distribution area is obtained by using the offset weight corresponding to the first distribution area, the deacidification efficiency sub-evaluation value, the fluidization stability sub-evaluation value, and the temperature field suitability sub-evaluation value.

3. A collaborative control system for circulating fluidized bed flue gas purification, characterized in that, include: The data acquisition unit is used to acquire the first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal images of the purification system. In addition, the third set of operating parameters, the third set of acoustic emission signals, and the third set of infrared thermal images of the purification system under healthy operating conditions are obtained to construct a health status data matrix; The feature processing unit is used to perform projection processing on the first set of operating parameters, the first set of acoustic emission signals, and the first set of infrared thermal images to obtain first feature data. The health status data matrix is ​​standardized to obtain a covariance matrix, and the covariance matrix is ​​decomposed to obtain a projection matrix. The health status data matrix is ​​then projected onto a low-dimensional health feature space using the projection matrix to obtain the projected second feature data. A construction unit is used to obtain a multidimensional probability distribution model based on the second feature data; wherein the multidimensional probability distribution model is used to describe the health status of the purification system; The filtering unit is used to filter the first feature data according to the constructed multidimensional probability distribution model to obtain the first distribution region; The determining unit is used to determine the regional health index corresponding to the first distribution area based on the second set of operating parameters, the second set of acoustic emission signals, and the second set of infrared thermal imaging data in the first distribution area. The control command generation unit is used to determine the control strategy of the first distribution area based on the regional health index, and generate control commands for the first distribution area to control the purification system. The filtering unit is specifically used for: Calculate the Mahalanobis distance of the first feature data in the multidimensional probability distribution model and the health probability value belonging to the historical health state; Based on the Mahalanobis distance and the health probability value, the first feature data is filtered to determine the first distribution area.

4. The collaborative control system for circulating fluidized bed flue gas purification according to claim 3, characterized in that, The second set of operating parameters includes the calcium-to-acid ratio and the deacidification efficiency value; the second set of acoustic emission signals includes the reaction tower pressure difference; the second set of infrared thermal imaging data includes the reaction tower outlet temperature; the determining unit is specifically used for: Based on the calcium-acid ratio and the deacidification efficiency value, a normal distribution evaluation is used to determine the deacidification efficiency sub-evaluation value. Based on the pressure difference in the reaction tower, a normal distribution assessment is used to determine the fluidization stability sub-assessment value; Based on the outlet temperature of the reaction tower, a normal distribution assessment is used to determine the temperature field suitability sub-assessment value. The regional health index corresponding to the first distribution area is obtained by using the offset weight corresponding to the first distribution area, the deacidification efficiency sub-evaluation value, the fluidization stability sub-evaluation value, and the temperature field suitability sub-evaluation value.

Citation Information

Patent Citations

  • Acid regeneration flue gas purification device and method

    CN116832588A

  • Charging pile cluster edge collaborative scheduling method based on power grid time-sharing margin

    CN120638303A