A method and device for blast furnace process fault diagnosis based on dynamic canonical correlation analysis

Abstract

This invention provides a method and apparatus for blast furnace process fault diagnosis based on dynamic canonical correlation analysis, belonging to the field of blast furnace process fault diagnosis technology. The method includes: acquiring offline and online blast furnace data; performing data identification processing based on the offline and online blast furnace data using a data-driven method to obtain offline zero-mean vectors and online zero-mean vectors; performing fault verification using the SD-CCA method based on a preset fault threshold and the offline and online zero-mean vectors to obtain SD-CCA-based residual vectors and fault detection results; when the fault detection result indicates a fault, performing fault retrieval based on the SD-CCA-based residual vectors and obtaining the fault type based on a fault database. This invention is an accurate and efficient method for blast furnace process fault diagnosis based on dynamic canonical correlation analysis.

Description

Technical Field

[0001] This invention relates to the field of blast furnace process fault diagnosis technology, and in particular to a method and apparatus for blast furnace process fault diagnosis based on dynamic canonical correlation analysis. Background Technology

[0002] Steel is a raw material in industry, supporting industrialization and modernization. With economic and social development, the demand for high-quality steel and the requirements for production safety are constantly increasing. The steel production process can be divided into two processes: steelmaking and rolling. Steelmaking mainly transforms raw materials into molten steel, while rolling mainly processes the molten steel into billets and then into various steel products through rolling mills. The blast furnace process is the core of the steelmaking process, accounting for the largest proportion of emissions and energy consumption in the entire steel production process, and approximately 35% of the total production cost. Because it is located upstream in the production process, the quality of the produced molten iron directly affects the subsequent rolling process. Due to the crucial role of the blast furnace process in the entire production process, real-time monitoring of the blast furnace process and timely detection and identification of blast furnace fault types are particularly important. Failure to detect and handle faults in a timely manner can lead to significant losses of resources and equipment, and may even cause accidents resulting in casualties. Therefore, designing a suitable fault diagnosis algorithm for the blast furnace system to promptly detect and locate faults, predict potential risks, and optimize production parameters is of great significance for production safety and improving economic efficiency in the blast furnace ironmaking process.

[0003] The blast furnace system mainly consists of the blast furnace body, charging system, pulverized coal injection system, tapping system, hot blast system, and gas treatment system. The blast furnace process involves inputting raw materials and fuel, and then blowing air into the furnace. A series of reactions occur within the blast furnace, producing molten iron, slag, and gas as products. Molten iron and slag are products of long-term reactions, and using them to judge the blast furnace condition has a certain lag. Because pulverized coal is solid and has poor fluidity, using pulverized coal control methods in blast furnace regulation also suffers from response lag. In contrast, air flows much faster, not only reflecting the blast furnace condition more accurately but also allowing for faster control of the blast furnace condition by adjusting parameters such as air volume and temperature.

[0004] If the model of the blast furnace process is known, model-based detection methods can be used. However, the blast furnace process is a highly complex dynamic system, a complex combination of physicochemical processes, making its mechanisms difficult to analyze and difficult to detect faults using accurate mechanistic models. With the development of instrumentation science, high-quality sensors in blast furnaces have collected a large amount of process data, providing data support for data-driven fault detection methods. Data-driven fault detection methods refer to methods that utilize real-time data and techniques such as machine learning and statistical analysis to identify anomalies or faults in equipment or systems. Unlike physical model-based fault detection methods, data-driven methods do not rely on detailed physical models of the equipment, but rather predict and detect faults through changes in patterns or characteristics in historical and real-time data. This effectively solves the problem of complex blast furnace system models and the difficulty of modeling them.

[0005] Diagnostic methods based on multivariate statistical analysis are a type of data-driven approach and are now widely used in the field of fault detection. This method utilizes multivariate statistical techniques to analyze the operational data of industrial processes, identifying anomalies and detecting faults. Its core is analyzing the interrelationships and collaborative change patterns among multiple variables. Commonly used methods include Principal Component Analysis (PCA), Canonical Correlation Analysis (CCA), and Partial Least Squares Regression (PLS). CCA-based fault diagnosis methods diagnose faults by analyzing the correlation between inputs and outputs. This method is model-independent and can perform fault diagnosis based on process data. However, traditional CCA fault diagnosis methods can only handle static data, and their applicability is limited to industrial processes operating under single, stable conditions. Modern industrial processes, such as blast furnace processes, are dynamic and time-varying. During blast furnace operation, operating conditions frequently change due to adjustments in production demand, fluctuations in raw materials, and equipment aging. If only the traditional CCA fault diagnosis method is used, it is impossible to accurately capture the dynamic characteristics of the system, resulting in diagnostic lag and the inability to eliminate the influence of changes in reference variables, which seriously interferes with the judgment of blast furnace faults.

[0006] In the existing technology, there is a lack of an accurate and efficient method for blast furnace process fault diagnosis based on dynamic canonical correlation analysis. Summary of the Invention

[0007] To address the technical problems of data dynamism and time-varying nature in existing blast furnace fault diagnosis technologies, this invention provides a method and apparatus for blast furnace process fault diagnosis based on dynamic canonical correlation analysis. The technical solution is as follows:

[0008] On the one hand, a blast furnace process fault diagnosis method based on dynamic canonical correlation analysis is provided. This method is implemented by blast furnace process fault diagnosis equipment and includes:

[0009] Acquire offline and online blast furnace data; based on a data-driven method, perform data identification processing on the offline and online blast furnace data to obtain offline zero-mean vector and online zero-mean vector;

[0010] Based on a preset fault threshold, the SD-CCA method is used to perform fault verification according to the offline zero-mean vector and the online zero-mean vector, so as to obtain the residual vector based on SD-CCA and the fault detection result.

[0011] When the fault detection result is a fault, the fault type is obtained by fault retrieval based on the residual vector based on SD-CCA, according to the fault database.

[0012] On the other hand, a blast furnace process fault diagnosis device based on dynamic canonical correlation analysis is provided. This device is applied to a blast furnace process fault diagnosis method based on dynamic canonical correlation analysis. The device includes:

[0013] The blast furnace data acquisition module is used to acquire offline blast furnace data and online blast furnace data; based on a data-driven method, it performs data identification processing on the offline blast furnace data and online blast furnace data to obtain the offline zero-mean vector and the online zero-mean vector;

[0014] The fault detection module is used to perform fault verification using the SD-CCA method based on a preset fault threshold, the offline zero-mean vector, and the online zero-mean vector, to obtain the residual vector based on SD-CCA and the fault detection result.

[0015] The fault type retrieval module is used to retrieve the fault type based on the residual vector based on SD-CCA when the fault detection result is a fault.

[0016] On the other hand, a blast furnace process fault diagnosis device is provided, the blast furnace process fault diagnosis device comprising: a processor; a memory, the memory storing computer-readable instructions, which, when executed by the processor, implement any of the above-described blast furnace process fault diagnosis methods based on dynamic canonical correlation analysis.

[0017] On the other hand, a computer-readable storage medium is provided, wherein at least one instruction is stored in the storage medium, the at least one instruction being loaded and executed by a processor to implement any of the above-described methods for blast furnace process fault diagnosis based on dynamic canonical correlation analysis.

[0018] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0019] This invention proposes a fault diagnosis method for blast furnace processes based on dynamic canonical correlation analysis. It utilizes image representation methods to identify stable image representation operators using offline data, eliminating the dynamic nature of the closed-loop blast furnace system. The SD-CCA method is used for fault verification, transforming zero-mean vectors into SD-CCA residuals to achieve fault detection. This invention effectively addresses the dynamic nature of blast furnace systems. In fault detection, the proposed method achieves good fault detection performance when facing the dynamic and time-varying characteristics of blast furnace systems. For fault classification, based on stable image representation operators, an orientation angle-based fault classification method is used. A fault database is established using historical fault data. When a new fault is detected, it is compared with the vectors in the database, and the fault type is determined based on the angle. This method can accurately classify faults and identify faults in the blast furnace process, providing guidance for subsequent fault elimination. This invention is an accurate and efficient fault diagnosis method for blast furnace processes based on dynamic canonical correlation analysis. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a flowchart of a blast furnace process fault diagnosis method based on dynamic canonical correlation analysis provided by an embodiment of the present invention;

[0022] Figure 2 This is a schematic diagram of a blast furnace ironmaking process provided in an embodiment of the present invention;

[0023] Figure 3 This is a schematic diagram of a fault detection result when no fault occurs, provided by an embodiment of the present invention;

[0024] Figure 4 This is a schematic diagram of the fault detection results when a blast furnace pipeline stroke fault occurs, provided by an embodiment of the present invention;

[0025] Figure 5This is a schematic diagram of the fault detection results when a blast furnace charge suspension failure occurs, provided by an embodiment of the present invention;

[0026] Figure 6 This is a schematic diagram of the fault detection results when a blast furnace cooling failure occurs, provided by an embodiment of the present invention.

[0027] Figure 7 This is a block diagram of a blast furnace process fault diagnosis device based on dynamic canonical correlation analysis provided in an embodiment of the present invention;

[0028] Figure 8 This is a schematic diagram of the structure of a blast furnace process fault diagnosis device provided in an embodiment of the present invention. Detailed Implementation

[0029] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0030] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0031] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0032] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0033] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0034] This invention provides a blast furnace process fault diagnosis method based on dynamic canonical correlation analysis. This method can be implemented by a blast furnace process fault diagnosis device, which can be a terminal or a server. Figure 1 The flowchart shown is for a blast furnace process fault diagnosis method based on dynamic canonical correlation analysis. The processing flow of this method may include the following steps:

[0035] S1. Obtain offline and online blast furnace data; using a data-driven approach, perform data identification and processing based on the offline and online blast furnace data to obtain the offline zero-mean vector and the online zero-mean vector.

[0036] Optionally, based on a data-driven approach, data identification processing is performed on offline and online blast furnace data to obtain offline zero-mean vectors and online zero-mean vectors, including:

[0037] Based on the offline blast furnace data and the preset reference vector, a data-driven approach is used to process the data and obtain the offline zero-mean vector.

[0038] Based on the online blast furnace data and the preset reference vector, a data-driven approach is used to process the data and obtain the online zero-mean vector.

[0039] In one feasible implementation, to eliminate the influence of time-varying characteristics of the blast furnace system, the present invention utilizes offline data and a data-driven method to identify stable image representation operators. For online blast furnace data, a data-driven method is used to obtain a zero-mean vector of the online data, thus eliminating the influence of time-varying characteristics.

[0040] Blast furnace system such as Figure 2 As shown, the closed-loop system is represented by equations (1) and (2):

[0041] (1);

[0042] (2);

[0043] Where K(z) represents the controller, G(z) represents the controlled object, and w, u, and y represent the reference input, process input, and process output of the closed-loop system, respectively. In the blast furnace process, the setpoint of the blast furnace parameters is used as the reference input w, the actual measured internal state of the blast furnace is used as the process input u, and the output of the blast furnace is used as the output y. For any time k, w, u, and y can be written as... , , For a dataset of length s, the mathematical expressions are as follows: (3), (4), (5):

[0044] (3);

[0045] (4);

[0046] (5);

[0047] Optionally, based on offline blast furnace data and a preset reference vector, a data-driven approach is used to process the data to obtain an offline zero-mean vector, including:

[0048] Using blast furnace parameter setpoints as reference inputs, actual measured blast furnace internal states as process inputs, and blast furnace outputs as process outputs, a sample Hankel matrix is ​​obtained based on offline blast furnace data.

[0049] Based on the sample Hankel matrix, factorization is performed through low-rank decomposition to obtain stable image representation operators;

[0050] Based on the stable image representation operator, calculations are performed according to a preset reference vector to obtain a non-zero mean vector driven by the reference vector.

[0051] Based on the non-zero mean vector driven by the reference vector, the time-varying effects of multiple image representation operators are eliminated to obtain an offline zero mean vector.

[0052] In one feasible implementation, w, u, and y in a closed system are written in the form of a Hankel matrix, where each column or row is a temporal shift relative to the previous column or row. This structure is particularly suitable for revealing the temporal dependencies and dynamic characteristics in sequence data. Depending on the indices, Hankel matrices are divided into two forms with indices f and p, storing past and future information respectively. u and w are written in Hankel matrix form containing N samples. , , As shown in equations (6), (7), and (8):

[0053] (6);

[0054] (7);

[0055] (8);

[0056] System Input and output From the reference vector This representation is called a stable image representation, as shown in equation (9). This is called the stable image representation operator:

[0057] (9);

[0058] Will The vector is decomposed into low-rank factors as shown in equation (10):

[0059] (10);

[0060] in, and These are the lower triangular matrix and the orthogonal matrix corresponding to the low-rank decomposition, respectively. and i,j,n=1,2,3 are respectively and The block matrix.

[0061] therefore, It can be approximated by the following equation (11):

[0062] (11);

[0063] Based on the definition of image representation, the stable image representation operator Approximate the following formula (12):

[0064] (12);

[0065] After identifying the stable image representation operator, a zero-mean vector can be obtained by driving it with a reference vector. Eliminate the influence of time-varying factors on the system, offline zero-mean vector As shown in equation (13):

[0066] (13);

[0067] Offline zero-mean vector The same method can be used to obtain it, so it will not be repeated here. The method for calculating the zero average of online blast furnace data is the same as that for offline data, so it will not be repeated here.

[0068] S2. Based on the preset fault threshold, the SD-CCA method is used to verify the fault according to the offline zero-mean vector and the online zero-mean vector, so as to obtain the residual vector based on SD-CCA and the fault detection result.

[0069] Optionally, based on a preset fault threshold, and according to the offline zero-mean vector and the online zero-mean vector, the SD-CCA method is used for fault verification to obtain the SD-CCA-based residual vector and fault detection results, including:

[0070] Based on the offline zero-mean vector, the covariance matrix is ​​calculated using canonical correlation analysis for fault detection.

[0071] Singular value decomposition is performed on the covariance matrix to obtain the left singular vector matrix, the singular value matrix, and the right singular vector matrix.

[0072] The parameters based on SD-CCA are calculated using the left singular vector matrix, singular value matrix, and right singular vector matrix.

[0073] The residual vector based on SD-CCA is obtained by calculating the parameters based on SD-CCA and the online zero-mean vector.

[0074] Fault detection results are obtained by performing fault verification based on preset fault thresholds and residual vectors based on SD-CCA.

[0075] One feasible implementation method is fault detection based on stable image representation and dynamic canonical correlation analysis. According to the canonical correlation analysis fault detection method, the covariance matrix of the residuals is calculated, and the parameters are obtained through singular value decomposition to calculate the residuals based on SD-CCA. The SD-CCA-based residuals are then converted into statistical forms and compared with a threshold to achieve fault detection.

[0076] After eliminating the time-varying effects in the above steps, we obtain the vector of offline zero mean. and Based on the vectors obtained from offline data, parameters can be calculated. Referring to the steps of CCA fault detection, the following parameters are calculated: and The covariance matrices are multiplied by their singular value decomposition to obtain the left singular vector matrix. Singular Value Matrix Right singular vector matrix In canonical correlation analysis, these three parameter matrices are used to express and interpret the key linear relationship between two sets of data. The mathematical expression is as follows (14):

[0077] (14);

[0078] Where n represents the degrees of freedom, , express and The covariance and cross-covariance.

[0079] Based on equation (14), the optimal parameters can be determined as follows: , , , The calculation method is as follows (15):

[0080] (15);

[0081] Through online zero-mean vector and And with four optimal parameters, residuals based on SD-CCA can be calculated. , The vectors are as follows: (16) and (17):

[0082] (16);

[0083] (17);

[0084] The residual vector based on SD-CCA is written as The statistical measures are in the following forms (18) and (19):

[0085] (18);

[0086] (19);

[0087] in , Represents the residual vector , Covariance of vectors.

[0088] Based on the set confidence level Given the degrees of freedom n and the sample size N, the threshold can be calculated using the F-distribution. As shown in equation (20):

[0089] (20);

[0090] Statistic and threshold By making a comparison, fault detection can be performed. The mathematical expression for this process is as follows (21):

[0091] (twenty one);

[0092] Among them, the SD-CCA method refers to a method that combines signal detection threshold and canonical correlation analysis.

[0093] In one feasible implementation, Stable Image Representation Aided Dynamic CCA (SD-CCA) is a method that combines signal detection thresholding and canonical correlation analysis for channel state assessment and fault detection in dynamic environments. This method enhances the interpretability of canonical variables by introducing sparsity into feature extraction. Furthermore, by extracting sparse features during the dynamic process, it enhances the ability to explore the relationship between inputs and outputs. This method also demonstrates its superiority in fields such as industrial process monitoring.

[0094] S3. When the fault detection result is a fault, the fault type is obtained by fault retrieval based on the residual vector based on SD-CCA, according to the fault database.

[0095] Optionally, when the fault detection result is a fault, fault retrieval is performed based on the fault database and the residual vector based on SD-CCA to obtain the fault type, including:

[0096] Based on the fault database, obtain all types of fault vectors;

[0097] Based on all types of fault vectors, the vector angle dataset is obtained by calculating the residual vectors based on SD-CCA.

[0098] The minimum value of the vector angle in the dataset is selected as the fault vector angle.

[0099] Based on the fault database, the fault type is determined according to the angle between the fault vectors.

[0100] In one feasible implementation, fault separation is performed on detected faults based on stable image representation and dynamic canonical correlation analysis. This method belongs to the fault isolation method based on residual direction vectors. Historical fault data is collected to construct a fault database, where each vector represents a fault type. When a fault is detected, the fault type is determined by the angle between the fault database and the fault residual vector; if the angle with a certain vector in the fault database is the smallest, then that fault type is identified.

[0101] The detected fault residual vector Compare with each fault type, i.e., each column in fault database B. Each column in the fault database... Both represent a direction vector of a fault type. When With a certain fault type When the directional angle j is at its minimum, the newly detected fault is considered to be of this fault type. The calculation formula for this process is as follows (22):

[0102] (twenty two);

[0103] The fault database is constructed based on historical fault data; each vector in the fault database represents a fault type.

[0104] In one feasible implementation, this invention requires the prior collection of different faults, based on the SD-CCA residual vector. It can be used and Simplified representation, further written as the following formula (23):

[0105] (twenty three);

[0106] in, , .

[0107] For each fault sample , measured value It can be divided into fault items and noise terms The two parts, equation (23), can be transformed into the following equation (24):

[0108] (twenty four);

[0109] in, Let be the direction of FS at time k, assuming that the residual vectors of different faults are not collinear.

[0110] Each fault type Sum of each fault sample, This yields the direction vector for each fault type. Each fault type constitutes a fault database B, and each column in the fault database... This corresponds to the direction of each fault type. The final fault database B can be obtained as shown in equation (25):

[0111] (25)

[0112] In one feasible implementation, the present invention takes the condition of a blast furnace of Liuzhou Iron and Steel Group as an example. It sets controllable standard wind speed, oxygen enrichment flow rate, cold blast pressure, and hot blast temperature as reference inputs, and uses the measured blast kinetic energy, total pressure difference, and actual wind speed inside the blast furnace as process inputs. It uses the amount of blast furnace gas in the flank and the flank gas index as process outputs. The offline data adopts the data of its normal operation, and the blast furnace data is sampled approximately every 10 seconds.

[0113] There are three typical failure scenarios in the blast furnace process. The first failure is a pipeline travel failure, which refers to an area of ​​the blast furnace burden having exceptionally high permeability, causing the gas flow to develop abnormally in that area, similar to that in a pipeline. In this case, the blast pressure gradually decreases while the blast volume gradually increases. The second failure is a suspended burden failure, which refers to the blast furnace burden stopping its descent. In this case, the blast pressure increases, the blast volume decreases, and the permeability index drops. The third failure is a furnace cooling failure, which refers to a sharp drop in blast furnace temperature, specifically manifested as unstable blast volume and pressure, and low pressure and temperature of the gas at the furnace top. In a blast furnace at Liuzhou Iron and Steel Group, 2000, 1000, and 2500 failure samples were collected respectively.

[0114] When no fault occurs, the test results are as follows Figure 3 As shown, the statistics are mostly below the curve. The results of the three types of fault detection are as follows: Figure 4 , Figure 5 , Figure 6 As shown in the figure, the false alarm rate was set to 0.01. It can be seen that this fault detection method can detect faults in a timely manner. After detecting a fault, the residuals of 1000 detected fault samples were classified, as shown in Table 1 (Fault Classification Results Table). It can be seen that all three types of faults can be classified well, with a classification success rate of 99.8%.

[0115] Table 1

[0116]

[0117] This invention proposes a fault diagnosis method for blast furnace processes based on dynamic canonical correlation analysis. It utilizes image representation methods to identify stable image representation operators using offline data, eliminating the dynamic nature of the closed-loop blast furnace system. The SD-CCA method is used for fault verification, transforming zero-mean vectors into SD-CCA residuals to achieve fault detection. This invention effectively addresses the dynamic nature of blast furnace systems. In fault detection, the proposed method achieves good fault detection performance when facing the dynamic and time-varying characteristics of blast furnace systems. For fault classification, based on stable image representation operators, an orientation angle-based fault classification method is used. A fault database is established using historical fault data. When a new fault is detected, it is compared with the vectors in the database, and the fault type is determined based on the angle. This method can accurately classify faults and identify faults in the blast furnace process, providing guidance for subsequent fault elimination. This invention is an accurate and efficient fault diagnosis method for blast furnace processes based on dynamic canonical correlation analysis.

[0118] Figure 7 This is a block diagram illustrating a blast furnace process fault diagnosis device based on dynamic canonical correlation analysis, according to an exemplary embodiment. The device is used in a blast furnace process fault diagnosis method based on dynamic canonical correlation analysis. (Refer to...) Figure 7 The device includes a blast furnace data acquisition module 710, a fault detection module 720, and a fault type retrieval module 730. Among them:

[0119] Blast furnace data acquisition module 710 is used to acquire offline blast furnace data and online blast furnace data; based on a data-driven method, it performs data identification processing on the offline blast furnace data and online blast furnace data to obtain the offline zero-mean vector and the online zero-mean vector.

[0120] The fault detection module 720 is used to perform fault verification based on a preset fault threshold, according to the offline zero-mean vector and the online zero-mean vector, using the SD-CCA method to obtain the residual vector based on SD-CCA and the fault detection result.

[0121] The fault type retrieval module 730 is used to retrieve the fault type based on the fault database and the residual vector based on SD-CCA when the fault detection result is a fault.

[0122] Optionally, the blast furnace data acquisition module 710 is further used for:

[0123] Based on the offline blast furnace data and the preset reference vector, a data-driven approach is used to process the data and obtain the offline zero-mean vector.

[0124] Based on the online blast furnace data and the preset reference vector, a data-driven approach is used to process the data and obtain the online zero-mean vector.

[0125] Optionally, the blast furnace data acquisition module 710 is further used for:

[0126] Using blast furnace parameter setpoints as reference inputs, actual measured blast furnace internal states as process inputs, and blast furnace outputs as process outputs, a sample Hankel matrix is ​​obtained based on offline blast furnace data.

[0127] Based on the sample Hankel matrix, factorization is performed through low-rank decomposition to obtain stable image representation operators;

[0128] Based on the stable image representation operator, calculations are performed according to a preset reference vector to obtain a non-zero mean vector driven by the reference vector.

[0129] Based on the non-zero mean vector driven by the reference vector, the time-varying effects of multiple image representation operators are eliminated to obtain an offline zero mean vector.

[0130] Optionally, the fault detection module 720 is further used for

[0131] Based on the offline zero-mean vector, the covariance matrix is ​​calculated using canonical correlation analysis for fault detection.

[0132] Singular value decomposition is performed on the covariance matrix to obtain the left singular vector matrix, the singular value matrix, and the right singular vector matrix.

[0133] The parameters based on SD-CCA are calculated using the left singular vector matrix, singular value matrix, and right singular vector matrix.

[0134] The residual vector based on SD-CCA is obtained by calculating the parameters based on SD-CCA and the online zero-mean vector.

[0135] Fault detection results are obtained by performing fault verification based on preset fault thresholds and residual vectors based on SD-CCA.

[0136] Among them, the SD-CCA method refers to a method that combines signal detection threshold and canonical correlation analysis.

[0137] Optionally, the fault type retrieval module 730 is further used for:

[0138] Based on the fault database, obtain all types of fault vectors;

[0139] Based on all types of fault vectors, the vector angle dataset is obtained by calculating the residual vectors based on SD-CCA.

[0140] The minimum value of the vector angle in the dataset is selected as the fault vector angle.

[0141] Based on the fault database, the fault type is determined according to the angle between the fault vectors.

[0142] The fault database is constructed based on historical fault data; each vector in the fault database represents a fault type.

[0143] This invention proposes a fault diagnosis method for blast furnace processes based on dynamic canonical correlation analysis. It utilizes image representation methods to identify stable image representation operators using offline data, eliminating the dynamic nature of the closed-loop blast furnace system. The SD-CCA method is used for fault verification, transforming zero-mean vectors into SD-CCA residuals to achieve fault detection. This invention effectively addresses the dynamic nature of blast furnace systems. In fault detection, the proposed method achieves good fault detection performance when facing the dynamic and time-varying characteristics of blast furnace systems. For fault classification, based on stable image representation operators, an orientation angle-based fault classification method is used. A fault database is established using historical fault data. When a new fault is detected, it is compared with the vectors in the database, and the fault type is determined based on the angle. This method can accurately classify faults and identify faults in the blast furnace process, providing guidance for subsequent fault elimination. This invention is an accurate and efficient fault diagnosis method for blast furnace processes based on dynamic canonical correlation analysis.

[0144] Figure 8 This is a schematic diagram of the structure of a blast furnace process fault diagnosis device provided in an embodiment of the present invention, as shown below. Figure 8 As shown, the blast furnace process fault diagnosis equipment may include the above-mentioned Figure 7 The illustrated blast furnace process fault diagnosis device is based on dynamic canonical correlation analysis. Optionally, the blast furnace process fault diagnosis device 810 may include a first processor 2001.

[0145] Optionally, the blast furnace process fault diagnosis device 810 may also include a memory 2002 and a transceiver 2003.

[0146] The first processor 2001, memory 2002, and transceiver 2003 can be connected via a communication bus.

[0147] The following is combined with Figure 8 A detailed introduction to each component of the blast furnace process fault diagnosis equipment 810 is provided below:

[0148] The first processor 2001 is the control center of the blast furnace process fault diagnosis equipment 810. It can be a single processor or a collective term for multiple processing elements. For example, the first processor 2001 can be one or more central processing units (CPUs), application-specific integrated circuits (ASICs), or one or more integrated circuits configured to implement embodiments of the present invention, such as one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs).

[0149] Optionally, the first processor 2001 can perform various functions of the blast furnace process fault diagnosis device 810 by running or executing software programs stored in the memory 2002 and calling data stored in the memory 2002.

[0150] In a specific implementation, as one example, the first processor 2001 may include one or more CPUs, for example... Figure 8 CPU0 and CPU1 are shown in the diagram.

[0151] In a specific implementation, as one example, the blast furnace process fault diagnosis device 810 may also include multiple processors, for example... Figure 8 The first processor 2001 and the second processor 2004 are shown in the diagram. Each of these processors can be a single-core processor or a multi-core processor. Here, a processor can refer to one or more devices, circuits, and / or processing cores used to process data (such as computer program instructions).

[0152] The memory 2002 is used to store the software program that executes the present invention, and is controlled by the first processor 2001 to execute it. The specific implementation method can be referred to the above method embodiment, and will not be repeated here.

[0153] Optionally, the memory 2002 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory 2002 may be integrated with the first processor 2001 or may exist independently, and may be connected via the interface circuit of the blast furnace process fault diagnosis device 810. Figure 8 (Not shown in the image) is coupled to the first processor 2001, and this embodiment of the invention does not specifically limit this.

[0154] The transceiver 2003 is used to communicate with network devices or with terminal devices.

[0155] Alternatively, transceiver 2003 may include a receiver and a transmitter. Figure 8 (Not shown separately). The receiver is used to implement the receiving function, and the transmitter is used to implement the transmitting function.

[0156] Optionally, the transceiver 2003 can be integrated with the first processor 2001, or it can exist independently and be connected to the interface circuit of the blast furnace process fault diagnosis device 810. Figure 8 (Not shown in the image) is coupled to the first processor 2001, and this embodiment of the invention does not specifically limit this.

[0157] It should be noted that, Figure 8 The structure of the blast furnace process fault diagnosis device 810 shown in the figure does not constitute a limitation on the router. The actual knowledge structure identification device may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0158] Furthermore, the technical effect of the blast furnace process fault diagnosis equipment 810 can be referred to the technical effect of the blast furnace process fault diagnosis method based on dynamic canonical correlation analysis described in the above method embodiments, and will not be repeated here.

[0159] It should be understood that the first processor 2001 in the embodiments of the present invention may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0160] It should also be understood that the memory in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0161] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0162] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0163] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0164] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0165] 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, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0166] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0167] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0168] 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; that is, 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 according to actual needs.

[0169] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0170] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0171] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for blast furnace process fault diagnosis based on dynamic canonical correlation analysis, characterized in that, The method includes: Acquire offline and online blast furnace data; based on a data-driven method, perform data identification processing on the offline and online blast furnace data to obtain offline zero-mean vector and online zero-mean vector; The data-driven method, which performs data identification processing based on the blast furnace offline data and the blast furnace online data to obtain the offline zero-mean vector and the online zero-mean vector, includes: Based on the offline blast furnace data and the preset reference vector, a data-driven method is used to process the data and obtain the offline zero-mean vector. Based on the blast furnace online data and the preset reference vector, a data-driven method is used to process the data and obtain the online zero-mean vector. The step of processing data using a data-driven method based on the offline blast furnace data and a preset reference vector to obtain an offline zero-mean vector includes: Using the blast furnace parameter setpoints as reference inputs, the actual measured internal state of the blast furnace as process inputs, and the blast furnace output as process outputs, a sample Hankel matrix is ​​obtained based on the offline blast furnace data. Based on the sample Hankel matrix, factorization is performed through low-rank decomposition to obtain stable image representation operators; Based on the stable image representation operator, calculations are performed according to a preset reference vector to obtain a non-zero mean vector driven by the reference vector. Based on the non-zero mean vector driven by the reference vector, the time-varying effects of the multiple image representation operators are eliminated to obtain an offline zero mean vector; Based on a preset fault threshold, the SD-CCA method is used to perform fault verification according to the offline zero-mean vector and the online zero-mean vector, so as to obtain the residual vector based on SD-CCA and the fault detection result. The step of performing fault verification using the SD-CCA method based on a preset fault threshold, the offline zero-mean vector, and the online zero-mean vector to obtain the SD-CCA-based residual vector and fault detection results includes: Based on the offline zero-mean vector, the covariance matrix is ​​calculated using canonical correlation analysis for fault detection. Singular value decomposition is performed on the covariance matrix to obtain the left singular vector matrix, the singular value matrix, and the right singular vector matrix; The parameters based on SD-CCA are calculated using the left singular vector matrix, singular value matrix, and right singular vector matrix. The residual vector based on SD-CCA is calculated based on the parameters of SD-CCA and the online zero-mean vector. Fault verification is performed based on a preset fault threshold and the residual vector based on SD-CCA to obtain fault detection results; When the fault detection result is a fault, the fault type is obtained by fault retrieval based on the residual vector based on SD-CCA, according to the fault database.

2. The blast furnace process fault diagnosis method based on dynamic canonical correlation analysis according to claim 1, characterized in that, The SD-CCA method refers to a method that combines signal detection threshold and canonical correlation analysis.

3. The blast furnace process fault diagnosis method based on dynamic canonical correlation analysis according to claim 1, characterized in that, When the fault detection result is a fault, based on the fault database, fault retrieval is performed according to the residual vector based on SD-CCA to obtain the fault type, including: Based on the fault database, obtain all types of fault vectors; Based on all types of fault vectors, the vector angle dataset is obtained by calculating the residual vector based on SD-CCA. The minimum value in the vector angle dataset is selected as the fault vector angle. Based on the fault database, the fault type is determined according to the angle between the fault vectors.

4. The blast furnace process fault diagnosis method based on dynamic canonical correlation analysis according to claim 3, characterized in that, The fault database is constructed based on historical fault data; each vector in the fault database represents a fault type.

5. A blast furnace process fault diagnosis device based on dynamic canonical correlation analysis, wherein the blast furnace process fault diagnosis device based on dynamic canonical correlation analysis is used to implement the blast furnace process fault diagnosis method based on dynamic canonical correlation analysis as described in any one of claims 1-4, characterized in that, The device includes: The blast furnace data acquisition module is used to acquire offline blast furnace data and online blast furnace data; based on a data-driven method, it performs data identification processing on the offline blast furnace data and online blast furnace data to obtain the offline zero-mean vector and the online zero-mean vector; The fault detection module is used to perform fault verification using the SD-CCA method based on a preset fault threshold, the offline zero-mean vector, and the online zero-mean vector, to obtain the residual vector based on SD-CCA and the fault detection result. The fault type retrieval module is used to retrieve the fault type based on the residual vector based on SD-CCA when the fault detection result is a fault.

6. A blast furnace process fault diagnosis device, characterized in that, The blast furnace process fault diagnosis equipment includes: processor; A memory storing computer-readable instructions that, when executed by the processor, implement the method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium contains program code that can be invoked by a processor to execute the method as described in any one of claims 1 to 4.

Images