Centralized and distributed control method and system for aluminum oxide production plant

Abstract

The invention relates to the technical field of industrial production process control, in particular to an alumina production plant distributed control method and system.The method comprises the steps that alumina production instrument data monitored in real time is collected and stored, and first alumina production data and second alumina production data are obtained; combining the two kinds of data with alumina production reference standard data, a dynamic path algorithm and an Euclidean distance to calculate the similarity of the data, and judging whether abnormal feature analysis needs to be carried out or not through an alumina abnormal threshold value; when anomaly analysis is needed, first aluminum oxide anomaly feature data are obtained by combining a time series data anomaly recognition algorithm and time series data prediction, second aluminum oxide anomaly feature data are obtained through clustering attribution analysis, and corresponding self-adjusting measures are taken; according to the method, the DCS system can carry out more specific and more accurate alumina fault monitoring under the condition that the production process is not influenced, and the self-diagnosis range of the alumina DCS system is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial production process control, and in particular to a method and system for distributed control of an aluminum oxide production plant. Background Art

[0002] Alumina is a metal compound with high hardness, good corrosion resistance, high temperature resistance and diverse physical and chemical properties. Due to its many advantages, it is widely used in industry, medicine, electronics and other fields.

[0003] The alumina production process includes six stages: raw material preparation, dissolution, sedimentation, decomposition, roasting and evaporation. Since the production process is relatively complicated, problems in any link can easily affect the production quality of alumina. In order to solve the quality problem, CN113608560B proposes a control system for the alumina alkali solution preparation process. In order to solve the problem of high caustic soda concentration in the prepared discharge, the real-time preparation data is monitored and adjusted through a feedforward compensator and a feedback controller to improve the qualified rate of the discharge. CN112144043B proposes an alumina deposition equipment and gas supply system. Through an improved alumina deposition equipment, the argon supply in the gas supply process is improved, which effectively reduces the probability of blockage in the system during the generation process and improves the quality and production capacity of the coated sheets.

[0004] The above method improves the production quality of alumina by improving the hardware equipment in the production process, but the above method still requires a high level of technology and operational capabilities, and cannot reduce the complex operation of alumina. In order to achieve an efficient and automated production process, the current alumina production process adopts the Decentralized Cintril System (DCS) commonly used in the current industry to reduce the complex operation of the alumina process.

[0005] The DCS system is a distributed computer control system that distributes the various control functions of the computer to multiple microprocessors to achieve precise control of each step. At the same time, the DCS system uses centralized management of these microprocessors to achieve better control. It is divided into multiple levels according to the specific production process to achieve more accurate management. At the same time, there are also some studies on improving the current DCS system of alumina. For example, CN112684772B proposes a thermal balance control system and method for alumina dissolution. In view of the hysteresis problem of thermal balance adjustment of the current DCS system, the temperature data between the heaters is obtained and the fuzzy reasoning model is introduced to analyze the relationship between the data. When there is a need for adjustment, the operating system can analyze the results in real time through the model and make adjustments to solve the hysteresis problem and improve product quality.

[0006] However, due to the high complexity of the DCS system of alumina, although the system has certain self-diagnosis functions, the diagnostic range is limited. Once a system failure occurs, it is difficult to find the specific fault problem, which makes it difficult to carry out maintenance work. Regardless of the size of the fault problem, once the DCS system fails, it will have a huge impact on the entire alumina production process. However, there is currently little research on fault monitoring of alumina DCS systems.

[0007] In order to realize real-time fault monitoring of alumina DCS system, improve the accuracy of fault monitoring of alumina DSC system and expand the scope of self-diagnosis and repair, a distributed control method and system for alumina production plant are proposed. Summary of the invention

[0008] The object of the present invention is to provide a method and system for distributed control of an alumina production plant, which acquires first alumina production data and second alumina production data by collecting and storing data of alumina production instruments monitored in real time; calculates the similarity of the two data by combining alumina production reference standard data, dynamic path algorithm and Euclidean distance, and determines whether to perform abnormal feature analysis by an alumina abnormal threshold; when abnormal analysis is required, acquires first alumina abnormal feature data by combining a time series data abnormality recognition algorithm and time series data prediction, acquires second alumina abnormal feature data by cluster attribution analysis and performs corresponding self-regulation measures; the above method enables a DCS system to perform more specific and accurate alumina fault monitoring without affecting the production process, thereby improving the self-diagnosis range of the alumina DCS system.

[0009] To achieve the above object, the present invention provides the following technical solutions:

[0010] A distributed control method for an aluminum oxide production plant, comprising:

[0011] Acquire the alumina production instrument data of each stage of the alumina plant according to the DCS system, monitor the alumina production instrument data of each stage in real time, and generate the first alumina production data of each stage;

[0012] Generate corresponding stage label data and instrument numbers according to the alumina production stage and the alumina production stage instrument, generate a corresponding database according to the instrument number and sort the alumina production instrument data according to the collection time T;

[0013] After the instrument production is completed, the alumina production instrument data is stored and the corresponding stage label data is marked to obtain the first alumina production data;

[0014] The alumina production instrument data are power data, instrument speed data, temperature data, liquid level data, humidity data, gas content data, pressure data and communication flow data of the alumina production instruments at each stage of the alumina production stage;

[0015] Obtaining second alumina production data according to the interval data of the first alumina production data in each stage; performing an abnormal comparison between the first alumina production data and the second alumina production data;

[0016] Further, the aluminum oxide abnormality threshold includes a first aluminum oxide reference threshold and a second aluminum oxide reference threshold interval;

[0017] Obtaining a first alumina similarity through the first alumina production data and alumina production reference standard data;

[0018] If the first aluminum oxide similarity is greater than the first aluminum oxide reference threshold, there is no abnormality; if the first aluminum oxide similarity is less than the first aluminum oxide reference threshold, an abnormality occurs;

[0019] If the second aluminum oxide production data is within the second aluminum oxide reference threshold interval, then there is no abnormality; if the second aluminum oxide production data is outside the second aluminum oxide reference threshold interval, then an abnormality occurs;

[0020] If the first aluminum oxide production data and / or the second aluminum oxide production data are missing, it is considered that an abnormality has occurred, and it is determined that the aluminum oxide abnormality characteristic analysis is not required, and the second aluminum oxide abnormality characteristic data is output;

[0021] Further, the first alumina similarity is obtained by comparing and calculating the alumina time series similarity, and the alumina time series similarity comparison calculation is to call the alumina production reference standard data of the corresponding stage according to the stage label data of the first alumina production data; take n as the vertical axis and m as the horizontal axis to establish an n*m mapping grid;

[0022] The first alumina production data is mapped along the direction of the longitudinal axis; the alumina production reference standard data is mapped along the horizontal axis, and the first alumina production data and the alumina production reference standard data are mapped to the n*m ​​mapping grid through dynamic path calculation to obtain a first alumina similarity path;

[0023] The alumina production reference standard data is mapped along the longitudinal and transverse directions, and mapped onto the n*m ​​mapping grid through dynamic path calculation to obtain a second alumina similar path;

[0024] The calculation of the first aluminum oxide similarity path and the second aluminum oxide similarity path is as follows:

[0025] G(i k ,j k )=min[D{i k ,j k}];

[0026]

[0027] Among them, G(i k ,j k ) is the position (i k ,j k ) is the total path distance calculated by the point, min[] is the minimum value, D{i k ,j k} is the position (i k ,j k ) point calculated from the current path distance, G(i k -1,j k ) is (i k -1,j k ) points to calculate the total path distance, G(i k -1,j k -1) is (i k -1,j k -1) points, G(i k ,j k -1) is (i k ,j k -1) points to calculate the total path distance, R(i k ,j k ) is the previous point arrival (i k ,j k ) location;

[0028] The first aluminum oxide similarity path and the second aluminum oxide similarity path are mapped to the same n*m ​​mapping grid, and the data of the first aluminum oxide similarity path and the second aluminum oxide similarity path are standardized and calculated, and then the Euclidean distance is calculated to obtain the first aluminum oxide similarity; the calculation is as follows:

[0029]

[0030] Among them, κ D is the first alumina similarity, is the coordinate of the dth point of the similarity path of the first alumina after normalization calculation, is the coordinate of the dth point of the similarity path of the second aluminum oxide after normalization calculation, and M is the total number of point coordinates;

[0031] If the abnormal comparison result meets the aluminum oxide abnormal threshold, the DCS system continues monitoring and analysis; if the abnormal comparison result does not meet the aluminum oxide abnormal threshold, it is determined whether an aluminum oxide abnormal characteristic analysis is required; if necessary, the first aluminum oxide production data and / or the second aluminum oxide production data are subjected to abnormal characteristic analysis to obtain first aluminum oxide abnormal characteristic data; if not necessary, the second aluminum oxide abnormal characteristic data is directly generated;

[0032] Further, the first alumina abnormal characteristic data is obtained through an alumina production abnormal model; the alumina production abnormal model includes an alumina production data input unit, an alumina production characteristic extraction unit, an alumina production abnormality identification unit, an alumina production abnormality prediction and evaluation unit and an alumina abnormality data output unit;

[0033] The alumina production data input unit performs production time series data preprocessing on the first alumina production data to generate alumina production input data;

[0034] The alumina production feature extraction unit extracts the features of the alumina production input data to generate alumina production feature data; the alumina production feature data is input into the alumina production anomaly identification unit, and compared with the alumina reference anomaly data to generate alumina abnormal morphology data and alumina abnormal data;

[0035] The alumina production abnormality prediction and evaluation unit predicts the second alumina production data through the alumina production input data to generate second alumina prediction data; compares and evaluates the second alumina prediction data and the second alumina production data to generate alumina prediction and evaluation data;

[0036] The alumina production feature extraction unit and the alumina production abnormality prediction and evaluation unit both adopt a multi-head attention mechanism and LSTM acquisition, which are specifically expressed as follows:

[0037]

[0038] Among them, output is output, input is input, Mul-Attention() is the multi-head attention mechanism, mul is the data volume of the multi-head attention mechanism, Pea() is feature splicing, and input is the alumina production input data;

[0039] The output of the alumina production feature extraction unit is the extracted time series feature, and the output of the alumina production abnormality prediction and evaluation unit is the value of the second alumina prediction data;

[0040] The alumina production feature extraction unit calculates and obtains feature similarity based on the time series feature and the alumina reference abnormal data, and the similarity calculation is the same as the calculation method of the first alumina similar path and the second alumina similar path;

[0041] The alumina production abnormality prediction and evaluation unit obtains a plurality of the second alumina prediction data through multiple predictions for comparative evaluation. The specific evaluation calculation is as follows:

[0042]

[0043] Among them, Score is the comparison evaluation result, θ s For the second alumina production data, α min and α max is the weight, λ ps is the second aluminum oxide prediction data, other represents other judgment situations, 1 represents similarity, 0 represents dissimilarity, maxc() is to obtain the value with the highest frequency, is N second aluminum oxide prediction data, N is the total number of the second aluminum oxide prediction data, maxf() is the number with the highest frequency of occurrence, The predicted data for the cth second aluminum oxide;

[0044] The alumina abnormal data output unit outputs corresponding data according to the generation results of the alumina abnormal morphological data, the alumina abnormal data and the alumina prediction and evaluation data;

[0045] According to the corresponding stage of the first aluminum oxide abnormal characteristic data, performing abnormal fault attribution analysis of the corresponding stage on the first aluminum oxide abnormal characteristic data to obtain the second aluminum oxide abnormal characteristic data;

[0046] Further, the abnormal morphological data of aluminum oxide and the stage label data of the abnormal aluminum oxide data are obtained; and the corresponding clustering model of aluminum oxide production instrument problems is called according to the stage label data;

[0047] A first clustering question label is obtained based on the abnormal aluminum oxide morphology data and the mean of the abnormal aluminum oxide data; a second clustering question label is obtained based on the abnormal aluminum oxide morphology data and the duration of the abnormal aluminum oxide data; a third clustering question label is obtained based on the abnormal aluminum oxide morphology data and the occurrence time of the abnormal aluminum oxide data;

[0048] analyzing a subset of the first clustered question labels, the second clustered question labels, and the third clustered question labels;

[0049] If there is a unique subset, the alumina production clustering problem label is obtained according to the unique subset; if there are multiple subsets, the alumina production clustering problem label is obtained according to the multiple subsets; if there is no subset, the alumina production clustering problem label is generated according to the first clustering problem label, the second clustering problem label, the third clustering problem label and the result that no subset exists;

[0050] The alumina production clustering problem label is the second alumina abnormal feature data;

[0051] Alumina self-adjustment judgment is performed through the second alumina abnormal characteristic data. If self-adjustment is possible, the DCS system takes corresponding alumina self-adjustment measures for the alumina production instrument data according to the second alumina abnormal characteristic data; if self-adjustment is not possible, the second alumina abnormal characteristic data is sent to relevant maintenance personnel for adjustment;

[0052] The aluminum oxide self-adjustment judgment is that if the second aluminum oxide abnormal characteristic data has a unique subset, it is judged that self-adjustment can be performed, and the corresponding aluminum oxide self-adjustment scheme is called based on the unique subset; if the second aluminum oxide abnormal characteristic data does not have a unique subset, it is judged that self-adjustment cannot be performed;

[0053] The present invention also provides an alumina production plant distributed control system, including an alumina instrument management module, an alumina instrument abnormality comparison module, an alumina abnormality judgment and model analysis judgment module, an alumina attribution analysis module and an alumina instrument self-adjustment module, specifically:

[0054] Alumina instrument management module, the DCS system obtains the alumina production instrument data of each stage of the alumina plant, monitors the alumina production instrument data of each stage in real time, and generates the first alumina production data of each stage;

[0055] an alumina instrument abnormality comparison module, which obtains the second alumina production data according to the interval data of the first alumina production data in each stage; and performs abnormality comparison between the first alumina production data and the second alumina production data;

[0056] Further, calling the alumina production reference standard data of the corresponding stage according to the stage label data of the first alumina production data; taking n as the vertical axis and m as the horizontal axis, establishing an n*m mapping grid;

[0057] The first alumina production data is mapped along the direction of the longitudinal axis; the alumina production reference standard data is mapped along the horizontal axis, and the first alumina production data and the alumina production reference standard data are mapped to the n*m ​​mapping grid through dynamic path calculation to obtain a first alumina similarity path;

[0058] The alumina production reference standard data is mapped along the longitudinal and transverse directions, and mapped onto the n*m ​​mapping grid through dynamic path calculation to obtain a second alumina similar path;

[0059] Mapping the first aluminum oxide similarity path and the second aluminum oxide similarity path to the same n*m ​​mapping grid, and normalizing and calculating the data of the first aluminum oxide similarity path and the second aluminum oxide similarity path, and then calculating the Euclidean distance to obtain the first aluminum oxide similarity;

[0060] Alumina abnormality judgment and model analysis judgment module, if the abnormal comparison result meets the aluminum oxide abnormality threshold, the DCS system continues to monitor and analyze; if the abnormal comparison result does not meet the aluminum oxide abnormality threshold, it is determined whether aluminum oxide abnormality feature analysis is required; if necessary, the first aluminum oxide production data and / or the second aluminum oxide production data are subjected to abnormal feature analysis to obtain first aluminum oxide abnormal feature data; if not necessary, the second aluminum oxide abnormal feature data is directly generated;

[0061] Further, the first alumina abnormal characteristic data is obtained through an alumina production abnormal model; the alumina production abnormal model includes an alumina production data input unit, an alumina production characteristic extraction unit, an alumina production abnormality identification unit, an alumina production abnormality prediction and evaluation unit and an alumina abnormality data output unit;

[0062] The alumina production data input unit performs production time series data preprocessing on the first alumina production data to generate alumina production input data;

[0063] The alumina production feature extraction unit extracts the features of the alumina production input data to generate alumina production feature data; the alumina production feature data is input into the alumina production anomaly identification unit, and compared with the alumina reference anomaly data to generate alumina abnormal morphology data and alumina abnormal data;

[0064] The alumina production abnormality prediction and evaluation unit predicts the second alumina production data through the alumina production input data to generate second alumina prediction data; compares and evaluates the second alumina prediction data and the second alumina production data to generate alumina prediction and evaluation data;

[0065] The alumina abnormal data output unit outputs corresponding data according to the generation results of the alumina abnormal morphological data, the alumina abnormal data and the alumina prediction and evaluation data;

[0066] an aluminum oxide attribution analysis module, which performs an abnormal fault attribution analysis of the first aluminum oxide abnormal characteristic data at the corresponding stage according to the corresponding stage of the first aluminum oxide abnormal characteristic data, and obtains the second aluminum oxide abnormal characteristic data;

[0067] The alumina instrument self-adjustment module makes a judgment on alumina self-adjustment through the second alumina abnormal characteristic data. If self-adjustment is possible, the DCS system takes corresponding alumina self-adjustment measures for the alumina production instrument data based on the second alumina abnormal characteristic data; if self-adjustment is not possible, the second alumina abnormal characteristic data is sent to relevant maintenance personnel for adjustment.

[0068] Compared with the prior art, the present invention has the following beneficial effects:

[0069] 1. In order to obtain specific fault problems in the alumina production process, the production instrument data of alumina at each stage is collected in real time, and the similarity path of the alumina instrument data is obtained by mapping it with the alumina reference standard data. The similarity path is used to calculate whether the alumina production instrument data is lower than the threshold as the judgment standard for further abnormal analysis. By performing abnormal analysis on the abnormal data at each stage, the fault problem can be checked without affecting the production process. At the same time, it can also provide data support and basis for the subsequent DCS system specific fault problem monitoring and analysis, improve the accuracy of fault analysis, and increase the fault monitoring range of the DCS system.

[0070] 2. When abnormalities are found in instrument data during production, separate fault judgments are made based on the specific abnormal data. The time series characteristics of the instrument data are obtained through the alumina production abnormality model, and the specific abnormality type is obtained based on the abnormality recognition algorithm for further judgment. At the same time, in order to prevent instrument failures from causing connection problems between instruments in production, a prediction model is added to the model to predict the connection interval through instrument data to improve the accuracy of the alumina production abnormality model in analyzing instrument failure problems. At the same time, the combination of multi-model analysis can improve the scope of DCS system fault monitoring, so as to facilitate more specific attribution analysis.

[0071] 3. After analyzing the problem of alumina production abnormality model, in order to conduct further detailed analysis through these abnormal data and obtain the cause of the specific problem, the corresponding clustering algorithm of each stage is called to obtain the cause of the problem. In order to improve the accuracy of problem analysis, three data types are clustered to generate specific problems more accurately. After obtaining the specific problem, the DCS system can automatically adjust the self-adjustable part. When the system cannot adjust itself, the answer to the system analysis will be sent to relevant personnel for data reference. This method can enable the DCS system to perform more accurate and wider-range fault monitoring, and can analyze problems by itself, improving the accuracy and range of DCS system fault self-diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] Figure 1 It is a schematic diagram of the method flow of the present invention;

[0073] Figure 2 A schematic diagram of a first aluminum oxide similarity path and a second aluminum oxide similarity path of the present invention;

[0074] Figure 3 It is a schematic diagram of the alumina production abnormal model flow chart of the present invention;

[0075] Figure 4 It is a schematic diagram of the system flow of the present invention;

[0076] Figure 5 It is a schematic diagram of the structure of the aluminum oxide distributed control system of the present invention;

[0077] Figure 6 It is a schematic diagram of the software structure of the aluminum oxide distributed control system of the present invention. DETAILED DESCRIPTION

[0078] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0079] The DCS system can achieve accurate management and control of each step of alumina production through multi-level and multi-number microprocessors, and can achieve more detailed regulation and management by splitting the huge production system to improve product quality; however, the DCS system of alumina is currently limited in its analysis of faults. When a fault occurs, it is impossible to accurately obtain the location of the fault for further diagnosis and repair, and the occurrence of the fault will have a huge impact on the quality and quantity of the entire alumina product. In order to improve the fault monitoring accuracy of the DCS system of alumina and expand the repair scope of self-diagnosis, the present invention provides a distributed control method and system for an alumina production plant, and the technical solution is as follows:

[0080] Reference Figure 1 As shown, the alumina production instrument data of each stage of the alumina plant is obtained according to the DCS system, the alumina production instrument data of each stage is monitored in real time, and the first alumina production data of each stage is generated;

[0081] Obtaining second alumina production data according to the interval data of the first alumina production data in each stage; performing an abnormal comparison between the first alumina production data and the second alumina production data;

[0082] If the abnormal comparison result meets the aluminum oxide abnormal threshold, the DCS system continues monitoring and analysis; if the abnormal comparison result does not meet the aluminum oxide abnormal threshold, it is determined whether an aluminum oxide abnormal characteristic analysis is required; if necessary, the first aluminum oxide production data and / or the second aluminum oxide production data are subjected to abnormal characteristic analysis to obtain first aluminum oxide abnormal characteristic data; if not necessary, the second aluminum oxide abnormal characteristic data is directly generated;

[0083] According to the corresponding stage of the first aluminum oxide abnormal characteristic data, performing abnormal fault attribution analysis of the corresponding stage on the first aluminum oxide abnormal characteristic data to obtain the second aluminum oxide abnormal characteristic data;

[0084] Alumina self-regulation is judged through the second alumina abnormal characteristic data. If self-regulation is possible, the DCS system takes corresponding alumina self-regulation measures for the alumina production instrument data based on the second alumina abnormal characteristic data. If self-regulation is not possible, the second alumina abnormal characteristic data is sent to relevant maintenance personnel for adjustment.

[0085] The present invention obtains the data of alumina production instruments in production in real time based on the DCS system, reasonably manages and stores the data through the DCS system, generates the first alumina production data, and facilitates the subsequent production data analysis to improve the accuracy of the subsequent analysis;

[0086] Considering that the production of alumina is a detailed and precise process, the connection between the production work of each instrument is also the key to fault analysis. For this reason, the instrument connection time in each stage of the first alumina production data is recorded to obtain the second alumina production data. Based on the normal production data and connection data, the reference data without faults is compared to accurately determine whether a fault abnormality has occurred, reduce the calculation burden of the DCS system, and improve the fault analysis scope and efficiency of the DCS system. At the same time, further splitting the fault analysis problem steps can also improve the accuracy of the analysis;

[0087] In order to accurately analyze the problems, the DCS system is judged whether to conduct further abnormal analysis based on the alumina abnormal threshold. If there is abnormal data with obvious fault characteristics, the data is directly extracted for attribution analysis. If the fault characteristics are not obvious, the abnormal characteristics are analyzed, and further attribution analysis is performed based on the analysis results. This method can classify the abnormal characteristic analysis work, save the DCS system analysis process computing resources, and conduct further accurate analysis of each data to realize subsequent fault handling work and improve the accuracy of the abnormal characteristics of the entire fault analysis.

[0088] Based on the attribution analysis of the extracted abnormal feature data, the system can extract the abnormal features accurately, analyze the causes of the abnormal faults based on these abnormal features, and generate corresponding fault self-adjustment and repair plans based on the causes to ensure that the DCS system can perform more accurate self-repair work and improve the accuracy of the repair plan. At the same time, in the case that the DCS system cannot self-adjust, the accurate analysis results can be sent to relevant personnel for repair, so as to improve the efficiency of the relevant personnel's repair and the accuracy of fault repair.

[0089] Embodiment 1

[0090] For specific description, the present invention is described in conjunction with the following embodiments, a distributed control method for an aluminum oxide production plant, comprising:

[0091] Obtain the alumina production instrument data of each stage of the alumina plant according to the DCS system, monitor the alumina production instrument data of each stage in real time, and generate the first alumina production data of each stage;

[0092] Generate corresponding stage label data and instrument numbers according to the alumina production stage and the alumina production stage instrument, generate a corresponding database according to the instrument number and sort the alumina production instrument data according to the acquisition time T (for example, but not limited to 1 second, 2 seconds, 3 seconds);

[0093] After the instrument production is completed, the alumina production instrument data is stored and the corresponding stage label data is marked to obtain the first alumina production data; the alumina production instrument data is the power data, instrument speed data, temperature data, liquid level data, humidity data, gas content data, pressure data and communication flow data of the alumina production instruments at each stage in the alumina production stage; alumina production instruments such as alumina ore crusher, alumina ball mill, alumina rotary kiln used in the roasting stage, etc.;

[0094] The specific stages include instrument data of the raw material preparation stage, dissolution stage, sedimentation stage, decomposition stage, roasting stage and evaporation stage of alumina; power data, instrument speed data, temperature data, liquid level data, humidity data, gas content data, pressure data and communication flow data are uniformly input into the data acquisition system based on the measurement system and monitoring system in the DCS system, and the corresponding data of the six stages are collected in combination with the original collected data of the data acquisition system;

[0095] Based on the distributed storage logic of the DCS system, the collected data is stored according to different stages and steps, which is convenient for subsequent management and call, and also convenient for adaptive real-time collection of data collection methods at different stages; at the same time, by distributing the data, it can also prevent the subsequent analysis from being too complicated due to excessive data, which reduces the accuracy of subsequent data analysis of the DCS system;

[0096] Obtaining second alumina production data based on interval data of the first alumina production data of each stage; performing an abnormal comparison between the first alumina production data and the second alumina production data;

[0097] The second alumina production data is the interval time between the connection of each instrument to the production work. Since the alumina production process is a relatively precise process, the time mutation between the work connection between the relevant production instruments in the production process is also an important part of analyzing whether a hardware failure occurs;

[0098] The interval time is mainly to obtain the data volume of the current stage through the stage label data and combine it with the start time of recording to calculate the corresponding collection timestamp, and obtain the interval time according to the collection timestamps of different stages to generate the second alumina production data; in addition to the instrument data, time dimension data is also added. Considering the precision of alumina production compared with other industrial production, by obtaining the interval time data, the normal or abnormal measurement range of production instruments in each stage of the DCS system can be further obtained, thereby improving the accuracy of subsequent analysis;

[0099] Further, the aluminum oxide abnormality threshold includes a first aluminum oxide reference threshold and a second aluminum oxide reference threshold interval;

[0100] Obtaining the first alumina similarity through the first alumina production data and the alumina production reference standard data;

[0101] If the first aluminum oxide similarity is greater than the first aluminum oxide reference threshold, there is no abnormality; if the first aluminum oxide similarity is less than the first aluminum oxide reference threshold, an abnormality occurs;

[0102] If the second aluminum oxide production data is within the second aluminum oxide reference threshold interval, then there is no abnormality; if the second aluminum oxide production data is outside the second aluminum oxide reference threshold interval, then there is an abnormality;

[0103] If the first aluminum oxide production data and / or the second aluminum oxide production data are missing, it is considered that an abnormality has occurred, and it is determined that the aluminum oxide abnormality characteristic analysis is not required, and the second aluminum oxide abnormality characteristic data is output;

[0104] The first aluminum oxide reference threshold is a percentage ratio of similarity, and the second aluminum oxide reference threshold interval is two fixed value intervals. In the specific implementation process, due to the different production processes of each part, in one embodiment, the first aluminum oxide reference threshold and the second aluminum oxide reference threshold interval of each stage are combined with the parameters of the actual production instrument to adopt the same numerical standard. In another embodiment, the first aluminum oxide reference threshold and the second aluminum oxide reference threshold interval of the raw material preparation stage, the dissolution stage, the sedimentation stage, the decomposition stage, the roasting stage and the evaporation stage can be combined with expert experience and the parameters of the actual production instrument to adopt different numerical standards for the six stages;

[0105] The parameters of the actual production instrument are taken as an example of the rotary kiln in the roasting stage. This table only shows part of the data. The parameters are shown in Table 1:

[0106] Table 1 Related parameters of different types of rotary kilns

[0107]

[0108] For missing data, the present invention assumes that if missing data occurs for more than 5 consecutive collection time points, it is considered an abnormality. In the specific implementation process, the judgment standard for missing data can be modified according to the actual situation;

[0109] In order to facilitate subsequent analysis and reduce the subsequent computing resource occupation of the DCS system, after obtaining real-time alumina production instrument data, a comparison is made based on the reference threshold of alumina to distinguish whether there is an abnormality. This method can filter and process the data before analysis, improve the accuracy of subsequent model feature extraction and attribution analysis, and at the same time, some more obvious faults can be screened in advance at this stage to further improve the screening of different fault problems and improve the recognition and processing accuracy of the entire fault process of the DCS system.

[0110] Further, the first alumina similarity is obtained by comparing and calculating the alumina time series similarity. The alumina time series similarity comparison calculation is to call the alumina production reference standard data of the corresponding stage according to the stage label data of the first alumina production data; using n as the vertical axis and m as the horizontal axis, an n*m mapping grid is established;

[0111] The first alumina production data is mapped along the vertical axis; the alumina production reference standard data is mapped along the horizontal axis, and the first alumina production data and the alumina production reference standard data are mapped to an n*m mapping grid through dynamic path calculation to obtain the first alumina similar path; the alumina production reference standard data will call the alumina production reference standard data of one of the raw material preparation stage, dissolution stage, sedimentation stage, decomposition stage, roasting stage and evaporation stage through the stage label data corresponding to the data for comparison;

[0112] The reference standard data of alumina production is mapped along the vertical and horizontal axes, and mapped onto an n*m mapping grid through dynamic path calculation to obtain a second alumina similarity path;

[0113] Alumina production reference standard data is obtained based on historical data collected from a large number of production instruments combined with expert experience;

[0114] The calculations for the first alumina similarity path and the second alumina similarity path are as follows:

[0115] G(i k ,j k )=min[D{i k ,j k}];

[0116]

[0117] Among them, G(i k ,j k ) is the position (i k ,j k ) is the total path distance calculated by the point, min[] is the minimum value, D{i k ,j k} is the position (i k ,j k ) point calculated from the current path distance, G(i k -1,j k ) is (i k -1,j k ) points to calculate the total path distance, G(i k -1,j k -1) is (i k -1,j k-1) points, G(i k ,j k -1) is (i k ,j k -1) points to calculate the total path distance, R(i k ,j k ) is the previous point arrival (i k ,j k ) location;

[0118] Specifically, the path calculation reference of the first aluminum oxide similar path and the second aluminum oxide similar path is Figure 2 As shown, Figure 2 The numbers in each table are path weights. The formula G(i k -1,j k )+R(i k ,j k ) calculates the longitudinal distance of the path, that is, path A in the figure; formula G(i k ,j k -1)+R(i k ,j k ) calculates the lateral distance of the path, that is, path C in the figure; formula G(i k -1,j k -1)+2R(i k ,j k ) represents the oblique distance of the path, i.e. path B in the figure;

[0119] The first aluminum oxide similarity path and the second aluminum oxide similarity path are mapped to the same n*m ​​mapping grid, and the data of the first aluminum oxide similarity path and the second aluminum oxide similarity path are standardized and calculated, and then the Euclidean distance is calculated to obtain the first aluminum oxide similarity; the calculation is as follows:

[0120]

[0121] Among them, κ D is the first alumina similarity, is the coordinate of the dth point of the similarity path of the first alumina after normalization calculation, is the coordinate of the dth point of the similarity path of the second aluminum oxide after normalization calculation, and M is the total number of point coordinates;

[0122] In the similarity comparison stage, since the collected alumina production instrument data is time series data, in order to improve the similarity of time series data calculation, the similarity calculation is performed on the alumina production reference standard data without abnormalities and the first alumina production data in the data comparison stage, and the similarity tracking path of the two time series data is obtained by combining the dynamic tracking algorithm. At the same time, in order to further improve the accuracy of the similarity calculation, the similarity tracking path without abnormalities is obtained based on the two alumina production reference standard data without abnormalities. By converting the time series data for comparative calculation, the accuracy of the DCS system similarity analysis can be significantly improved, which is convenient for the feature extraction of the subsequent model and improves the accuracy of fault analysis.

[0123] If the abnormal comparison result meets the alumina abnormal threshold, the DCS system continues monitoring and analysis; if the abnormal comparison result does not meet the alumina abnormal threshold, it is determined whether an alumina abnormal characteristic analysis is required; if necessary, the first alumina production data and / or the second alumina production data are subjected to an abnormal characteristic analysis to obtain the first alumina abnormal characteristic data; if not necessary, the second alumina abnormal characteristic data is directly generated;

[0124] The second alumina abnormal characteristic data directly generated mainly includes some more significant abnormal characteristics, such as a large amount of data missing of liquid level data, serious data jump of power data, long-term sharp rise or fall of temperature data or humidity data, etc. in the alumina production process; these data do not need to be further analyzed because of their obvious characteristics, and can be directly screened by the system through traditional statistical methods such as triple error method to obtain the corresponding abnormal data;

[0125] Furthermore, the first alumina abnormal characteristic data is obtained through an alumina production abnormal model; the alumina production abnormal model includes an alumina production data input unit, an alumina production characteristic extraction unit, an alumina production abnormality identification unit, an alumina production abnormality prediction and evaluation unit and an alumina abnormality data output unit; the process refers to Figure 3 As shown;

[0126] The alumina production data input unit performs production time series data preprocessing on the first alumina production data to generate alumina production input data; the time series data preprocessing mainly includes time series data noise reduction and normalization;

[0127] The alumina production feature extraction unit extracts the features of the alumina production input data to generate alumina production feature data; the alumina production feature data is input into the alumina production anomaly identification unit, and compared with the alumina reference anomaly data to generate alumina abnormal morphology data and alumina abnormal data;

[0128] The alumina production abnormality prediction and evaluation unit predicts the second alumina production data through the alumina production input data to generate the second alumina prediction data; compares and evaluates the second alumina prediction data and the second alumina production data to generate the alumina prediction and evaluation data;

[0129] The alumina production feature extraction unit and the alumina production abnormality prediction and evaluation unit both adopt the multi-head attention mechanism and LSTM acquisition, which are specifically expressed as follows:

[0130]

[0131] Among them, output is output, input is input, Mul-Attention() is the multi-head attention mechanism, mul is the amount of multi-head attention mechanism data, Pea() is feature concatenation, and input is the input data of alumina production;

[0132] Among them, the number of heads of the multi-head attention mechanism of the alumina production feature extraction unit is 16, and the number of LSTM units is 6; the number of heads of the multi-head attention mechanism of the alumina production anomaly prediction and evaluation unit is 8, and the number of LSTM units is 12;

[0133] The multi-head attention mechanism is calculated as follows:

[0134]

[0135] in, is the i-node feature, σ is the activation function, W is the weight vector, is the j-node feature, α ij is the feature weight coefficient, exp() is an exponential function with e as the base, e ij and e ik is the node relevance, N i is the number of nodes;

[0136] The alumina production feature extraction unit and the alumina production anomaly prediction and evaluation unit both use a multi-head attention mechanism and LSTM model to train and test with different historical data;

[0137] The output of the alumina production feature extraction unit is the extracted time series feature, and the output of the alumina production abnormality prediction and evaluation unit is the value of the second alumina prediction data;

[0138] The alumina production feature extraction unit calculates and obtains feature similarity based on the time series feature and the alumina reference abnormal data. The similarity calculation method is the same as that of the first alumina similar path and the second alumina similar path. The alumina reference abnormal data is also selected through a large amount of historical data combined with expert experience.

[0139] The alumina production abnormality prediction and evaluation unit obtains multiple second alumina prediction data through multiple predictions for comparative evaluation. The specific evaluation calculation is as follows:

[0140]

[0141] Among them, Score is the comparison evaluation result, θ s is the second alumina production data, α min and α max is the weight, λ ps is the second aluminum oxide prediction data, other represents other judgment situations, 1 represents similarity, 0 represents dissimilarity, and maxc() is used to obtain the value with the highest frequency. is N second aluminum oxide prediction data, N is the total number of second aluminum oxide prediction data, maxf() is the number with the highest frequency, Predicted data for the cth second aluminum oxide;

[0142] For less obvious fault abnormal data, the features in the first alumina abnormal feature data are extracted through the alumina production abnormal model. Matching the corresponding abnormal data can accurately obtain the abnormal types corresponding to fault data of different forms. At the same time, considering that the connection may be slow due to instrument failure during the instrument connection process, but it cannot be accurately reflected in the data, resulting in fault omissions, the second alumina prediction data is predicted by the prediction model and compared with the actual second alumina production data for evaluation, which can more accurately discover some hidden problems, so as to facilitate immediate discovery and processing, and improve the comprehensiveness and accuracy of DCS system fault analysis;

[0143] The alumina abnormal data output unit outputs corresponding data according to the generation results of the alumina abnormal morphological data, the alumina abnormal data and the alumina prediction evaluation data;

[0144] Among them, the alumina abnormal morphological data outputs the corresponding labels according to the identified morphological features. For example, step abnormality is marked as 1, sudden jump is marked as 2, and violent fluctuation is marked as 3. The number and type of annotations can be set according to the specific implementation situation. The alumina abnormal data mainly outputs the corresponding original data and the start time and end time of the corresponding abnormality. The alumina prediction and evaluation data outputs the corresponding 1 or 0. If it is 0, the corresponding original data is output and the start time and end time of the corresponding abnormality are output.

[0145] The present invention tests the similarity calculation accuracy and prediction accuracy through 50 sets of simulated data, and also compares with other models. The difference with the actual result of the simulated data is within 1%, and the model calculation is correct. The present invention conducts 3 tests and selects the average as the final accuracy. The specific comparison results are shown in Table 2 and Table 3:

[0146] Table 2 Calculation results of accuracy of model alumina feature similarity comparison

[0147] Model One test accuracy Second test accuracy Three test accuracy Final accuracy Model of the present invention 90% (45 groups correct) 90% (45 groups correct) 94% (47 groups correct) 91.33％ Transformer+LSTM 84% (42 groups correct) 96% (48 groups correct) 86% (43 groups correct) 88.67％ Attention+GRU 88% (44 groups correct) 92% (46 correct answers) 90% (45 groups correct) 90.00％

[0148] Table 3 Calculation results of the prediction accuracy of the model alumina characteristics

[0149] Model One test accuracy Second test accuracy Three test accuracy Final accuracy Model of the present invention 88% (44 groups correct) 84% (42 groups correct) 84% (42 groups correct) 85.33％ Transformer+LSTM 90% (45 groups correct) 74% (37 groups correct) 80% (40 groups correct) 81.33％ Attention+GRU 76% (38 groups correct) 76% (38 groups correct) 78% (39 groups correct) 76.33％

[0150] According to the corresponding stage of the first aluminum oxide abnormal characteristic data, the first aluminum oxide abnormal characteristic data is subjected to abnormal fault attribution analysis of the corresponding stage to obtain the second aluminum oxide abnormal characteristic data;

[0151] Furthermore, the abnormal morphological data of alumina and the stage label data of the abnormal alumina data are obtained; the corresponding clustering model of the alumina production instrument problem is called according to the stage label data; the clustering model of the alumina production instrument problem is obtained based on a large amount of historical data of each stage; the k-means model is mainly used for clustering; the number of cluster centers k is set according to the data statistics of the historical data and expert experience;

[0152] The first clustering question label is obtained based on the abnormal morphology data of aluminum oxide and the mean value of the abnormal morphology data of aluminum oxide; the second clustering question label is obtained based on the abnormal morphology data of aluminum oxide and the duration of the abnormal morphology data of aluminum oxide; the third clustering question label is obtained based on the abnormal morphology data of aluminum oxide and the occurrence time of the abnormal morphology data of aluminum oxide;

[0153] The first clustering problem label is to use the label of the abnormal morphology data of aluminum oxide as the vertical axis coordinate, and the data mean of the acquired abnormal data as the horizontal axis coordinate to generate the corresponding coordinate data for clustering calculation;

[0154] The second clustering problem label is to use the label of the abnormal morphological data of aluminum oxide as the vertical axis coordinate, and the duration of the abnormal data of aluminum oxide calculated by the start time and end time of the abnormal data as the horizontal axis coordinate to generate the corresponding coordinate data for clustering calculation;

[0155] The third clustering problem label is to use the label of the abnormal morphological data of aluminum oxide as the vertical axis coordinate, and the start time of the abnormal data as the horizontal axis coordinate to generate the corresponding coordinate data for clustering calculation;

[0156] Analyze the subsets of the first clustering problem label, the second clustering problem label, and the third clustering problem label; if there is a unique subset, obtain the alumina production clustering problem label based on the unique subset; if there are multiple subsets, obtain the alumina production clustering problem label based on the multiple subsets; if there is no subset, generate the alumina production clustering problem label based on the first clustering problem label, the second clustering problem label, the third clustering problem label, and the result that no subset exists;

[0157] The first cluster problem label, the second cluster problem label and the third cluster problem label include hardware failure problems and software failure problems. Hardware failure problems include abnormal temperature sensors used for data collection, insufficient power supply of rotary kiln motors, insufficient pressure of alumina crushers, etc. Software problems include data storage software failure and inability to store data. Alumina DCS system software diagram reference Figure 6 As shown;

[0158] Since the DCS system itself has a large computational load, the attribution analysis is separated from the model to reduce the computational load of the DCS system analysis. At the same time, by making the abnormal analysis step and the attribution step independent, the accuracy of the problem analysis in each step can be improved to ensure the accuracy of the attribution analysis. At the same time, in order to ensure that the problem can be accurately obtained, three types of data are used for cluster analysis to improve the accuracy of the relevant problem and self-adjustment solution data during the attribution analysis process, ensuring that the DCS system can accurately solve the fault problem and improve the accuracy of the self-diagnosis adjustment.

[0159] The alumina production clustering problem label is the second alumina abnormal feature data;

[0160] The alumina self-adjustment judgment is made through the second alumina abnormal characteristic data. If self-adjustment is possible, the DCS system takes corresponding alumina self-adjustment measures for the alumina production instrument data according to the second alumina abnormal characteristic data; if self-adjustment is not possible, the second alumina abnormal characteristic data is sent to the relevant maintenance personnel for adjustment;

[0161] In a specific implementation, if there is only one subset and the DCS system can self-regulate the relevant hardware parameters or software parameters, the DCS system will adjust it. If there are multiple subsets or no subsets, the DCS system will send all analyzed hardware problems and / or software problems to relevant maintenance personnel for analysis and repair.

[0162] In another implementation scheme, regardless of the number of subsets, the DCS system sends the analysis results to the relevant maintenance personnel. If there is only one subset, the DCS system will perform repairs after the relevant maintenance personnel agree. Otherwise, the relevant maintenance personnel directly obtain the problem for analysis and repair.

[0163] In the actual problem solving process, since there may be multiple possibilities in the attribution analysis, when these possibilities cannot be judged, the final results of the DCS system analysis are stored and sent to relevant personnel for judgment, so as to avoid the DCS system from making incorrect repairs in special circumstances, resulting in the occurrence of new faults. At the same time, it can also ensure the accuracy of the DCS system in fault self-adjustment, improve the efficiency and accuracy of fault problem solving, and ensure that the DCS system of Alumina can perform self-adjustment and repair work in an efficient and accurate context;

[0164] The self-regulation of alumina is judged as follows: if there is a unique subset of the second alumina abnormal characteristic data, it is judged that self-regulation can be performed, and the corresponding alumina self-regulation scheme is called based on the unique subset; if there is no unique subset of the second alumina abnormal characteristic data, it is judged that self-regulation cannot be performed.

[0165] Embodiment 2

[0166] The present invention also provides a distributed control system for an alumina production plant, which is optimized based on the overall process of the DCS system. Through the computing resources and storage resources of the existing DCS system, an algorithm is added to perform real-time monitoring and real-time analysis of faults in each part. At the same time, the distributed concept of the DCS system is also inherited. Different stages, different functions and different faults are split according to steps, and accurate problem analysis is performed while ensuring that the DCS system can be monitored in real time, so as to minimize the computing resource usage. At the same time, the problems that arise can also be quickly analyzed and calculated, thereby improving the robustness of the overall system and the accuracy of fault analysis of each problem. The system includes an alumina instrument management module, an alumina instrument abnormality comparison module, an alumina abnormality judgment and model analysis judgment module, an alumina attribution analysis module and an alumina instrument self-adjustment module. Figure 4 and Figure 5 As shown, Figure 5 All structures in the are connected through a network. The specific process is as follows:

[0167] Alumina instrument management module, DCS system through Figure 5 The data acquisition unit in the data acquisition station obtains the alumina production instrument data at each stage of the alumina plant. Figure 5 The data collection station in the system monitors the data of the alumina production instruments at each stage in real time and generates the first alumina production data at each stage;

[0168] Alumina instrument abnormal comparison module, Figure 5 The data computing station obtains the second alumina production data according to the interval data of the first alumina production data of each stage transmitted by the data acquisition station through the communication interface; the data computing station performs an abnormal comparison between the first alumina production data and the second alumina production data transmission;

[0169] Further, Figure 5 The data calculation station in the process calls the reference standard data of alumina production of the corresponding stage according to the stage label data of the first alumina production data; takes n as the vertical axis and m as the horizontal axis to establish an n*m mapping grid;

[0170] The first alumina production data is mapped along the vertical axis; the alumina production reference standard data is mapped along the horizontal axis, and the first alumina production data and the alumina production reference standard data are mapped to an n*m mapping grid through dynamic path calculation to obtain the first alumina similarity path;

[0171] The reference standard data of alumina production is mapped along the vertical and horizontal axes, and mapped onto an n*m mapping grid through dynamic path calculation to obtain a second alumina similarity path;

[0172] The first aluminum oxide similarity path and the second aluminum oxide similarity path are mapped to the same n*m ​​mapping grid, and the data of the first aluminum oxide similarity path and the second aluminum oxide similarity path are standardized and calculated, and then the Euclidean distance is calculated to obtain the first aluminum oxide similarity;

[0173] By adding corresponding fault analysis and judgment models to the DCS system according to each stage, relevant data can be judged without affecting the overall production process of alumina. At the same time, the distributed processing method based on the DCS system can also improve the accuracy of the analysis, ensuring that the DCS system can accurately detect faults and avoid repeated data analysis occupying a large amount of computing resources;

[0174] Alumina abnormality judgment and model analysis judgment module, if Figure 5 If the abnormal comparison result of the data calculation station in the data collection station meets the alumina abnormal threshold, the data collection station continues to monitor and analyze; if the abnormal comparison result does not meet the alumina abnormal threshold, the data calculation station determines whether alumina abnormal feature analysis is required; if necessary, the first alumina production data and / or the second alumina production data are transmitted to the data collection station through the network. Figure 5 The central computer 2 in the data processing station performs abnormal characteristic analysis to obtain the first aluminum oxide abnormal characteristic data; if not required, the data processing station generates the second aluminum oxide abnormal characteristic data and transmits it to the Figure 5 The central computer 1 in;

[0175] Furthermore, the first alumina abnormal characteristic data is obtained by Figure 5 The alumina production abnormality model stored in the central computer 2 is obtained; the alumina production abnormality model includes an alumina production data input unit, an alumina production feature extraction unit, an alumina production abnormality identification unit, an alumina production abnormality prediction and evaluation unit and an alumina abnormality data output unit;

[0176] The alumina production data input unit performs production time series data preprocessing on the first alumina production data to generate alumina production input data;

[0177] The alumina production feature extraction unit extracts the features of the alumina production input data to generate alumina production feature data; the alumina production feature data is input into the alumina production anomaly identification unit, and compared with the alumina reference anomaly data to generate alumina abnormal morphology data and alumina abnormal data;

[0178] The alumina production abnormality prediction and evaluation unit predicts the second alumina production data through the alumina production input data to generate the second alumina prediction data; compares and evaluates the second alumina prediction data and the second alumina production data to generate the alumina prediction and evaluation data;

[0179] The alumina abnormal data output unit outputs corresponding data according to the generation results of the alumina abnormal morphological data, the alumina abnormal data and the alumina prediction evaluation data;

[0180] In a specific embodiment, it can be combined with Figure 5 The actual computer resources of the central computer 2 in the embodiment are allocated. If the computer resources are limited, in the first embodiment, the abnormality analysis can be performed first and then the prediction; in the second embodiment, the prediction can be performed first and then the abnormality analysis; if the computer resources are sufficient, in the third embodiment, the abnormality analysis and the prediction can be performed simultaneously;

[0181] By making the alumina production abnormal model independent, it is possible to perform analysis when abnormal faults do occur, which can significantly improve the computing resource usage of the DCS system due to the excessive size of the model. At the same time, it can also give the model sufficient time for analysis to avoid errors caused by model analysis failures due to the simultaneous analysis of a large amount of data, improve the accuracy of model analysis, and ensure that the alumina DCS system can accurately analyze fault problems.

[0182] Alumina attribution analysis module, based on the corresponding stage of the first alumina abnormal characteristic data, Figure 5 The central computer 1 performs an abnormal fault attribution analysis of the first aluminum oxide abnormal characteristic data at a corresponding stage to obtain the second aluminum oxide abnormal characteristic data;

[0183] Alumina instrument self-regulating module, Figure 5 The central computer 1 in the system makes a self-adjustment judgment of aluminum oxide through the second aluminum oxide abnormal characteristic data. If self-adjustment is possible, the central computer 1 makes a self-adjustment judgment of aluminum oxide based on the second aluminum oxide abnormal characteristic data. Figure 5 The central computer 1 in the Figure 5 The operator workstation of the corresponding stage in Figure 5Send instructions to the engineer workstation in Figure 5 The corresponding instrument control station sends instructions through the communication interface based on the data of the alumina production instrument. Figure 5 The corresponding control unit takes corresponding alumina self-regulation measures; Figure 5 The control unit in the system is connected to the crusher and ball mill, etc., which record the instrument data of the raw material preparation stage, dissolution stage, sedimentation stage, decomposition stage, roasting stage and evaporation stage of alumina, and can adjust the adjustable parameters of each production instrument; if it cannot be adjusted by itself, then Figure 5 The central computer 1 in the system sends the second alumina abnormal characteristic data to Figure 5 Operator workstation in Figure 5 The relevant maintenance personnel of the engineer workstation in the system will make adjustments.

[0184] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A distributed control method for an aluminum oxide production plant, characterized in that: include: Acquire the alumina production instrument data of each stage of the alumina plant according to the DCS system, monitor the alumina production instrument data of each stage in real time, and generate the first alumina production data of each stage; Obtaining second alumina production data according to the interval data of the first alumina production data in each stage; performing an abnormal comparison between the first alumina production data and the second alumina production data; If the abnormal comparison result meets the aluminum oxide abnormal threshold, the DCS system continues monitoring and analysis; if the abnormal comparison result does not meet the aluminum oxide abnormal threshold, it is determined whether aluminum oxide abnormal feature analysis is required; If necessary, performing abnormal characteristic analysis on the first aluminum oxide production data and / or the second aluminum oxide production data to obtain first aluminum oxide abnormal characteristic data; If not required, the second alumina abnormal characteristic data is directly generated; According to the corresponding stage of the first aluminum oxide abnormal characteristic data, performing abnormal fault attribution analysis of the corresponding stage on the first aluminum oxide abnormal characteristic data to obtain the second aluminum oxide abnormal characteristic data; Alumina self-regulation is judged through the second alumina abnormal characteristic data. If self-regulation is possible, the DCS system takes corresponding alumina self-regulation measures for the alumina production instrument data based on the second alumina abnormal characteristic data. If self-regulation is not possible, the second alumina abnormal characteristic data is sent to relevant maintenance personnel for adjustment.

2. A distributed control method for an aluminum oxide production plant according to claim 1, characterized in that Generate corresponding stage label data and instrument numbers according to the alumina production stages and alumina production stage instruments, generate a corresponding database according to the instrument numbers and sort the alumina production instrument data according to the acquisition time T, and after the instrument production is completed, store the alumina production instrument data and mark the corresponding stage label data to obtain the first alumina production data.

3. The distributed control method for an aluminum oxide production plant according to claim 1, characterized in that: The abnormal comparison of the alumina abnormal threshold value includes: The alumina abnormality threshold includes a first alumina reference threshold and a second alumina reference threshold interval; the first alumina similarity is obtained through the first alumina production data and the alumina production reference standard data; if the first alumina similarity is greater than the first alumina reference threshold, there is no abnormality; if the first alumina similarity is less than the first alumina reference threshold, an abnormality occurs; if the second alumina production data is within the second alumina reference threshold interval, there is no abnormality; if the second alumina production data is outside the second alumina reference threshold interval, an abnormality occurs; if the first alumina production data and / or the second alumina production data have data missing, an abnormality occurs, it is determined that the alumina abnormality feature analysis is not required, and the second alumina abnormality feature data is output.

4. A distributed control method for an aluminum oxide production plant according to claim 3, characterized in that The first aluminum oxide similarity is obtained by comparing and calculating the aluminum oxide time series similarity, and the aluminum oxide time series similarity comparison calculation includes: According to the stage label data of the first alumina production data, the reference standard data of alumina production of the corresponding stage is called; with n as the vertical axis and m as the horizontal axis, an n*m mapping grid is established; The first alumina production data is mapped along the direction of the longitudinal axis; the alumina production reference standard data is mapped along the horizontal axis, and the first alumina production data and the alumina production reference standard data are mapped to the n*m ​​mapping grid through dynamic path calculation to obtain a first alumina similarity path; The alumina production reference standard data is mapped along the vertical and horizontal axes, and mapped to the n*m ​​mapping grid through dynamic path calculation to obtain the second alumina similarity path; the first alumina similarity path and the second alumina similarity path are mapped to the same n*m ​​mapping grid, and the data of the first alumina similarity path and the second alumina similarity path are standardized and calculated, and then the Euclidean distance is calculated to obtain the first alumina similarity.

5. The distributed control method for an aluminum oxide production plant according to claim 1, characterized in that: The first alumina abnormal characteristic data acquisition includes: The first alumina abnormal characteristic data is obtained through an alumina production abnormal model; the alumina production abnormal model includes an alumina production data input unit, an alumina production characteristic extraction unit, an alumina production abnormality identification unit, an alumina production abnormality prediction and evaluation unit, and an alumina abnormality data output unit; The alumina production data input unit performs production time series data preprocessing on the first alumina production data to generate alumina production input data; the alumina production feature extraction unit extracts the features of the alumina production input data to generate alumina production feature data; the alumina production feature data is input into the alumina production anomaly identification unit, and compared with the alumina reference anomaly data to generate alumina abnormal morphology data and alumina abnormal data; The alumina production abnormality prediction and evaluation unit predicts the second alumina production data through the alumina production input data to generate second alumina prediction data; compares and evaluates the second alumina prediction data and the second alumina production data to generate alumina prediction and evaluation data; the alumina abnormality data output unit outputs corresponding data based on the generation results of the alumina abnormal morphological data, the alumina abnormality data and the alumina prediction and evaluation data.

6. The distributed control method for an aluminum oxide production plant according to claim 1, characterized in that: Performing an abnormal fault attribution analysis on the first aluminum oxide abnormal characteristic data in the corresponding stage to obtain the second aluminum oxide abnormal characteristic data includes: Acquire the alumina abnormal morphology data and the stage label data of the alumina abnormal data; call the corresponding alumina production instrument problem clustering model according to the stage label data; A first clustering question label is obtained based on the abnormal aluminum oxide morphology data and the mean of the abnormal aluminum oxide data; a second clustering question label is obtained based on the abnormal aluminum oxide morphology data and the duration of the abnormal aluminum oxide data; a third clustering question label is obtained based on the abnormal aluminum oxide morphology data and the occurrence time of the abnormal aluminum oxide data; analyzing a subset of the first clustered question labels, the second clustered question labels, and the third clustered question labels; If there is a unique subset, the alumina production clustering problem label is obtained based on the unique subset; if there are multiple subsets, the alumina production clustering problem label is obtained based on the multiple subsets; if there is no subset, the alumina production clustering problem label is generated based on the first clustering problem label, the second clustering problem label, the third clustering problem label and the result that no subset exists; the alumina production clustering problem label is the second alumina abnormal feature data.

7. The distributed control method for an aluminum oxide production plant according to claim 1, characterized in that: The aluminum oxide self-adjustment judgment is that if the second aluminum oxide abnormal characteristic data has a unique subset, it is judged that self-adjustment can be performed, and the corresponding aluminum oxide self-adjustment scheme is called based on the unique subset; If there is no unique subset of the second aluminum oxide abnormal characteristic data, it is determined that self-adjustment is not possible.

8. A distributed control system for an aluminum oxide production plant, characterized in that: Alumina instrument management module, the DCS system obtains the alumina production instrument data of each stage of the alumina plant, monitors the alumina production instrument data of each stage in real time, and generates the first alumina production data of each stage; an alumina instrument abnormality comparison module, which obtains the second alumina production data according to the interval data of the first alumina production data in each stage; and performs abnormality comparison between the first alumina production data and the second alumina production data; Alumina abnormality judgment and model analysis judgment module, if the abnormal comparison result meets the aluminum oxide abnormality threshold, the DCS system continues to monitor and analyze; if the abnormal comparison result does not meet the aluminum oxide abnormality threshold, it is determined whether aluminum oxide abnormality feature analysis is required; If necessary, performing abnormal characteristic analysis on the first aluminum oxide production data and / or the second aluminum oxide production data to obtain first aluminum oxide abnormal characteristic data; If not required, the second alumina abnormal characteristic data is directly generated; an aluminum oxide attribution analysis module, which performs an abnormal fault attribution analysis of the first aluminum oxide abnormal characteristic data at the corresponding stage according to the corresponding stage of the first aluminum oxide abnormal characteristic data, and obtains the second aluminum oxide abnormal characteristic data; The alumina instrument self-adjustment module makes a judgment on alumina self-adjustment through the second alumina abnormal characteristic data. If self-adjustment is possible, the DCS system takes corresponding alumina self-adjustment measures for the alumina production instrument data based on the second alumina abnormal characteristic data; if self-adjustment is not possible, the second alumina abnormal characteristic data is sent to relevant maintenance personnel for adjustment.

9. A distributed control system for an aluminum oxide production plant according to claim 8, characterized in that The alumina instrument abnormality comparison module determines that based on the dynamic path algorithm, the first alumina similarity path and the second alumina similarity path are mapped to an n*m mapping network, the Euclidean distance between the first alumina similarity path and the second alumina similarity path is calculated, and the first alumina similarity is obtained for judgment.

10. The aluminum oxide production plant distributed control system according to claim 8, characterized in that The model analysis of the alumina anomaly judgment and model analysis judgment module is obtained based on the alumina production anomaly model.

Images