Coal gasification unit process diagnosis method and system

By building an initial model library and process diagnostic model library for coal gasification devices, and using clustering algorithms and data correction technology for online diagnosis, the refinement and persistence of coal gasification devices diagnosis in the existing technology is solved, and the model is improved. and user operation convenience, reducing energy consumption.

CN114861810BActive Publication Date: 2024-11-19CHINA PETROLEUM & CHEMICAL CORP +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210554979.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-05-20
Publication Date
2024-11-19
Estimated Expiration
2042-05-20

AI Technical Summary

Technical Problem

The existing coal gasification device process diagnostic methods rely on empirical qualitative diagnosis or process simulation software, making it difficult to achieve refined operation and continuous normalized diagnosis, and the model is poorly robust and difficult to promote and apply.

Method used

By acquiring and preprocessing the operation data of the coal gasification device, an initial model library and a process diagnostic model library are built, online diagnosis is carried out based on real-time data, and visual diagnostic results are formed using clustering algorithms and data correction technology.

Benefits of technology

The continuous normalized process diagnosis of coal gasification devices is realized, the robustness of the model is improved, the difficulty of user operation is reduced, the guidance of refined operation optimization is guided, and energy consumption is reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114861810B_ABST
    Figure CN114861810B_ABST
Patent Text Reader

Abstract

The present invention relates to a coal gasification device process diagnosis method and system. The coal gasification device process diagnosis method forms a sample database by preprocessing the coal gasification device operation data, and pre-builds a coal gasification device initialization model library based on the data in the sample database, and then develops a process diagnosis model library based on the coal gasification device initialization model, assessment technical parameters and constraint conditions, and automatically matches and calls the current operation data of the coal gasification device with the diagnosis model in the process diagnosis model library through the developed coal gasification device process diagnosis system, so as to realize online diagnosis of the coal gasification device. The coal gasification device process diagnosis model library in the method covers various working conditions of the coal gasification device, and the model has high robustness. The user interface of the coal gasification device process diagnosis system is easy to operate, which greatly reduces the difficulty of using the model. The coal gasification device can be continuously and normally diagnosed, and the coal gasification device can be guided to perform refined operation optimization.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of coal gasification devices, and in particular to a process diagnosis method and system for coal gasification devices. Background Art

[0002] Coal gasification is an important means of clean and efficient utilization of coal and is the core device of modern coal chemical industry. However, due to its characteristics of large fluctuations in raw material properties, harsh reaction conditions and complex processes, it has high energy consumption and high emissions. Compared with the energy consumption of natural gas and petrochemicals, the high energy consumption of coal chemical industry has become the main factor restricting its development. In order to reduce carbon emissions, coal gasification devices urgently need a reliable means to guide the device to optimize the process and reduce the energy consumption of the device.

[0003] At present, there are several main methods for process diagnosis of coal gasification units: one process diagnosis method is for relevant experts and technicians to judge and diagnose the operating status of the unit based on their experience, but this method has great limitations due to its heavy reliance on personnel experience, and can only perform qualitative diagnosis on the unit, but cannot perform quantitative diagnosis, making it difficult to guide the refined operation of the unit; another process diagnosis method is to use process simulation software to establish an offline model of the typical working conditions of the unit, perform simulation diagnosis on the unit to obtain the unit optimization plan to guide the optimization of the unit. However, since coal gasification is a typical process chemical industry, the production process is continuous and dynamically changing, involving complex chemical and physical changes, resulting in poor robustness of the model developed by this method, and it is impossible to perform continuous and normalized diagnosis of the unit. In addition, due to the high professionalism of process simulation software, there are technical barriers for modelers and users, making it difficult to promote and apply. Summary of the invention

[0004] The first technical problem to be solved by the present invention is to provide a process diagnosis method for a coal gasification device in view of the above-mentioned prior art.

[0005] The second technical problem to be solved by the present invention is to provide a coal gasification device process diagnosis system for realizing the above-mentioned coal gasification device process diagnosis method.

[0006] The technical solution adopted by the present invention to solve the first technical problem is: a process diagnosis method for a coal gasification device, characterized in that it comprises the following steps:

[0007] Step S1, obtaining an operation data set during the operation of the coal gasification device, and preprocessing the operation data in the operation data set to obtain a preprocessed operation data set;

[0008] Step S2, using the operation data in the preprocessed operation data set to pre-build a coal gasification device initialization model library for simulating the operation of the coal gasification device; wherein the coal gasification device initialization model library includes subunit initialization models corresponding to W different subunits, the different subunit initialization models include M feed categories, each feed category includes N operating condition categories, each subunit initialization model corresponds to a subunit operation condition in the simulated coal gasification device, and W is the total number of sub-operation units actually included in the coal gasification device;

[0009] Step S3, setting the assessment technical parameters and the constraint conditions corresponding to each assessment technical parameter based on the process diagnosis requirements, and developing a process diagnosis model library based on the initialization model, assessment technical parameters and constraint conditions in the coal gasification unit initialization model library;

[0010] Step S4, obtaining the actual operation data of the coal gasification device again, and inputting the obtained actual operation data into the process diagnosis model library for diagnosis, so as to obtain the process diagnosis result of the coal gasification device.

[0011] Optionally, in the coal gasification device process diagnosis method, the actual operation data in step S4 is retrieved real-time operation data of the coal gasification device or manually input operation data of the coal gasification device.

[0012] Furthermore, in the process diagnosis method for the coal gasification unit, the models in the process diagnosis model library are diagnostic models capable of performing simulation analysis on key technical parameters of the coal gasification unit, the key technical parameters are independent variable technical parameters relative to other technical parameters in the coal gasification unit, and the other technical parameters in the coal gasification unit are dependent variable technical parameters relative to the key technical parameters.

[0013] As a further improvement, the coal gasification device process diagnosis method also includes a process of displaying the retrieved real-time operation data in a visual manner; or / and a process of displaying the coal gasification device process diagnosis result in a visual manner.

[0014] Improved, in the coal gasification device process diagnosis method, in step S2, the process of constructing the coal gasification device initialization model includes the following steps a1 to a3:

[0015] Step a1, collecting the coal gasification device operation data, and performing validity processing on the coal gasification device operation data to form a basic database; wherein the coal gasification device operation data includes the coal gasification device real-time database and the historical data collected by LIMS;

[0016] Step a2, based on the basic database data, using clustering algorithms to perform feed clustering and operating condition clustering in sequence to form a sample data set library;

[0017] Step a3: constructing a coal gasification unit initialization model based on the formed sample set library data.

[0018] Furthermore, in the process diagnosis method for the coal gasification device, in step a2, firstly, the parameter data in the feed characteristic parameter set are clustered into M feed categories, and then each of the M feed categories is clustered into N operating condition categories according to the parameter data in the operating condition characteristic parameter set; wherein the clustering distance is calculated as follows:

[0019]

[0020] Among them, n and m are the numbers of the feature parameter set Y, n =(x n1 ,x n2 ,…,x nJ ), Y m =(x m1 ,x m2 ,…,x mJ ), cosθ nm is the feature parameter set Y n With the characteristic parameter set Y m The cosine distance between nj is the feature parameter set Y n The data of the jth characteristic parameter in the corresponding x mj is the feature parameter set Y m The data of the jth characteristic parameter in, J is the characteristic parameter set Y n The total number of characteristic parameters in the characteristic parameter set Y n With the characteristic parameter set Y m Have the same total number of characteristic parameters.

[0021] Optionally, in the coal gasification unit process diagnosis method, the clustering algorithm is a mean shift clustering algorithm or a self-organizing neural network clustering algorithm.

[0022] Furthermore, in the process diagnosis method of the coal gasification unit, the effectiveness processing in step a1 includes preprocessing, steady-state detection and data correction, and the preprocessing includes invalid data elimination, interpolation fitting filling, filtering denoising and data normalization processing.

[0023] Furthermore, in the process diagnosis method of the coal gasification device, in step a1, the invalid data elimination is to eliminate the operating data that meets the large error condition in the collected coal gasification device operating data, and the operating data that meets the large error condition is the operating data whose fluctuation coefficient value in the collected coal gasification device operating data is greater than the preset fluctuation coefficient threshold value; wherein, the collected coal gasification device operating data is the operating data collected within a preset time interval, and the fluctuation coefficient value of the kth operating data collected within the preset time interval is marked as τ k , the preset fluctuation coefficient threshold corresponding to the preset time interval is marked as τ th :

[0024]

[0025] a k is the kth operating data value of the coal gasification device operating data collected within the preset time interval, a k-1 is the k-1th operating data value of the coal gasification device operating data collected within the preset time interval, μ a is the average of all operating data values ​​of the coal gasification device operating data collected within the preset time interval; K is the total amount of coal gasification device operating data collected within the preset time interval.

[0026] Further improvement, in the coal gasification device process diagnosis method, the steady-state detection is a heuristic steady-state detection; the steady-state detection process includes the following steps b1 to b4:

[0027] Step b1, calculating the light filtering value and the heavy filtering value corresponding to each valid data in the coal gasification device operation data collected in step a1; wherein the valid data in the coal gasification device operation data collected is the remaining data after the invalid data is eliminated;

[0028] (Y L ) t =f L ·y t +(1-f L )·(Y L ) t-1 ;

[0029] (Y H ) t =f H ·y t +(1-f H )· (Y H ) t-1 ;

[0030] (Y L )0=y t1 , (Y H)0=y t1 ;

[0031] t1≤t≤t Q ;

[0032] Among them, the time period (t1,t Q ) is the collection time period of the coal gasification device operation data collected in step a1, (Y L ) t is the period (t1,t Q ) is the lightly filtered value of the valid data corresponding to the t-th time in the coal gasification device operation data collected in (Y H ) t is the time period (t1,t Q ) in the coal gasification device operation data collected in the re-filtered value of the valid data corresponding to the t-th time; (Y L ) t-1 is the period (t1,t Q ) is the lightly filtered value of the valid data corresponding to the t-1th moment in the coal gasification device operation data collected in (Y H ) t-1 is the time period (t1,t Q ) in the coal gasification device operation data collected in the re-filtered value of the valid data corresponding to the t-1th time, f L is the light filter coefficient, f H is the re-filtering coefficient; y t is the actual value of the valid data corresponding to the tth moment; Q is the value in the time period (t1, t Q ) The total number of all valid data corresponding to the time in the coal gasification unit operation data collected in );

[0033] Step b2, respectively calculating the absolute value of the light and heavy filter value difference corresponding to each valid data in the collected coal gasification device operation data, and obtaining the absolute value maximum of all the absolute values ​​obtained; wherein the absolute value of the light and heavy filter value difference of the valid data corresponding to the t-th time in the collected coal gasification device operation data is marked as The maximum absolute value among all the absolute values ​​obtained is marked as

[0034]

[0035] Step b3, making a parameter steady-state judgment based on the obtained absolute value maximum:

[0036] When the absolute maximum value is obtained When it is less than the preset parameter steady-state tolerance value χ, the factory value of the valid data corresponding to the time t is determined to be a steady-state value; otherwise, the factory value of the valid data corresponding to the time t is determined to be a non-steady-state value;

[0037] Step b4, making a steady-state judgment of the device according to the refiltered value of the valid data corresponding to the last moment and the refiltered value of the valid data corresponding to the first moment:

[0038] When the absolute value of the difference between the two is less than the preset trend tolerance value, Determine the time period [t1,t Q ]The state of the coal gasification unit corresponding to the valid data within t1,t2 is steady state; otherwise, it is determined that the time period [t1,t Q ]The state of the coal gasification unit corresponding to the valid data in is non-steady state.

[0039] Improved, in the coal gasification device process diagnosis method, the data correction includes the following steps: based on the original measurement data, using the material balance relationship or energy balance relationship of the coal gasification device production process, determining the correction condition based on the minimum sum of squares of deviations between the correction value and its corresponding measurement value, and establishing a correction mathematical expression based on least squares solution; wherein the correction mathematical expression is as follows:

[0040]

[0041]

[0042] Among them, X is the vector of measured values ​​of measured variables, and U is the vector of unmeasured or estimated parameters. is a vector composed of the correction values ​​of the measured variables, P is the variance-covariance matrix of the measurement error, and F is the constraint equation of the coal gasification unit model; wherein the constraint equation includes the material balance equation, the energy balance equation, the chemical reaction rate equation and the chemical balance equation, the heat mass and momentum transfer equation; P determines the weight of each variable for adjustment, and assigns a higher weight to instruments with higher precision.

[0043] The technical solution adopted by the present invention to solve the second technical problem is: a coal gasification device process diagnosis system, which implements the coal gasification device process diagnosis method, and is characterized in that the coal gasification device process diagnosis system includes:

[0044] A data acquisition module is used to obtain a set of operating data during the operation of the coal gasification device;

[0045] The data preprocessing module preprocesses the operation data in the operation data set collected by the data collection module to obtain the preprocessed operation data set, and performs category processing on the preprocessed operation data set data;

[0046] The device model management module manages the pre-built coal gasification device initial model and process diagnosis model for simulating the operation of the coal gasification device;

[0047] The process diagnosis module is responsible for the setting of the assessment technical parameters and calling the diagnosis model corresponding to the category to which the preprocessed running data set determined by the data preprocessing module belongs, and inputting the preprocessed running data set into the process diagnosis model for diagnosis and outputting the process diagnosis result.

[0048] Compared with the prior art, the advantages of the present invention are as follows: the gasification device process diagnosis method in the present invention forms a sample database by preprocessing the operation data of the gasification device, and pre-builds a gasification device initialization model library that simulates the operation of the gasification device based on the data in the sample database, and then develops a process diagnosis model library based on the gasification device initialization model, assessment technical parameters and constraints in the initialization model library, and automatically matches and calls the current operation data of the gasification device with the diagnosis model in the process diagnosis model library through the developed gasification device process diagnosis system, so as to realize online diagnosis of the gasification device. The process diagnosis model in the gasification device process diagnosis model library developed by the method covers various working conditions of the gasification device, and the model has high robustness. The user interface of the developed gasification device process diagnosis system is easy to operate, which greatly reduces the difficulty of using the model, and can perform continuous and normalized process diagnosis on the gasification device, and guide the gasification device to perform refined operation optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 It is a schematic flow chart of a process diagnosis method for a coal gasification device in an embodiment of the present invention;

[0050] Figure 2 Schematic diagram of a process diagnosis system for a coal gasification unit in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The present invention is further described in detail below with reference to the accompanying drawings.

[0052] This embodiment provides a method for diagnosing the process of a coal gasification device. Figure 1 As shown, the process diagnosis method of the coal gasification device in this embodiment includes the following steps:

[0053] Step S1, obtaining an operation data set during the operation of the coal gasification device, and preprocessing the operation data in the operation data set to obtain a preprocessed operation data set;

[0054] Step S2, using the operation data in the preprocessed operation data set to pre-build a coal gasification device initialization model library for simulating the operation of the coal gasification device; wherein the coal gasification device initialization model library includes subunit initialization models corresponding to W different subunits, the different subunit initialization models include M feed categories, each feed category includes N operating condition categories, each subunit initialization model corresponds to a subunit operation condition in the simulated coal gasification device, and W is the total number of sub-operation units actually included in the coal gasification device;

[0055] That is to say, if the coal gasification device is formed by 10 sub-units, then a sub-unit initialization model simulating its operation status must be constructed for each sub-unit, that is, 10 sub-unit initialization models need to be constructed in advance; of course, each sub-unit initialization model includes input data and output data, for example, the input data is the operation data of the sub-unit, and the output data is the performance parameter corresponding to the sub-unit;

[0056] Specifically in this embodiment, the construction process of the coal gasification device initialization model includes the following steps a1 to a3:

[0057] Step a1, collecting the operation data of the coal gasification device, and performing validity processing on the operation data of the coal gasification device to form a basic database; wherein the operation data of the coal gasification device includes the real-time database of the coal gasification device and the historical data collected by LIMS; the validity processing here includes preprocessing, steady-state detection and data correction, and the preprocessing includes invalid data elimination, interpolation fitting filling, filtering denoising and data normalization processing;

[0058] Step a2, based on the basic database data, using clustering algorithms to perform feed clustering and operating condition clustering in sequence to form a sample data set library;

[0059] Step a3, constructing a coal gasification device initialization model based on the formed sample number set library data;

[0060] Step S3, based on the process diagnosis requirements, set the assessment technical parameters and the constraints corresponding to each assessment technical parameter, and then develop a process diagnosis model library based on the initialization model, assessment technical parameters and constraints in the coal gasification device initialization model library; wherein the model in the process diagnosis model library here is a diagnosis model capable of simulating and analyzing the key technical parameters of the coal gasification device, the key technical parameters are independent variable technical parameters relative to other technical parameters in the coal gasification device, and the other technical parameters in the coal gasification device are dependent variable technical parameters relative to the key technical parameters;

[0061] Step S4, the actual operation data of the coal gasification device is obtained again, and the obtained actual operation data is input into the process diagnosis model library for diagnosis to obtain the process diagnosis result of the coal gasification device. For example, the actual operation data here can be the real-time operation data of the coal gasification device retrieved or the operation data of the coal gasification device input manually.

[0062] It should be noted that in step a2 of this embodiment, firstly, the parameter data in the feed characteristic parameter set are clustered into M feed categories, and then each of the M feed categories is clustered into N operating condition categories according to the parameter data in the operating condition characteristic parameter set; wherein, the cluster distance is calculated as follows:

[0063]

[0064] Among them, n and m are the numbers of the feature parameter set Y, n =(x n1 ,x n2 ,…,x nJ ), Y m =(x m1 ,x m2 ,…,x mJ ), cosθ nm is the feature parameter set Y n With the characteristic parameter set Y m The cosine distance between nj is the feature parameter set Y n The data of the jth characteristic parameter in the corresponding x mj is the feature parameter set Y m The data of the jth characteristic parameter in, J is the characteristic parameter set Y n The total number of characteristic parameters in the characteristic parameter set Y n With the characteristic parameter set Y m The total number of characteristic parameters is the same. Specifically in this embodiment, the feed characteristic parameter set is a set of characteristic parameters related to feeds such as feed composition and feed properties; the process characteristic parameter set is a set of parameters related to process parameters such as processing capacity and operating temperature; wherein, when clustering parameter data in the feed characteristic parameter set and clustering parameter data in the operating condition characteristic parameter set, the clustering distance calculation method used in the above formula is used.

[0065] In addition, with respect to step a1 of this embodiment, the invalid data elimination is to eliminate the operating data that meets the large error condition in the collected coal gasification device operating data, and the operating data that meets the large error condition is the operating data whose fluctuation coefficient value in the collected coal gasification device operating data is greater than the preset fluctuation coefficient threshold value; wherein, the collected coal gasification device operating data is the operating data collected within a preset time interval, and the fluctuation coefficient value of the kth operating data collected within the preset time interval is marked as τ k , the preset fluctuation coefficient threshold corresponding to the preset time interval is marked as τ th :

[0066]

[0067] a k is the kth operating data value of the coal gasification device operating data collected within the preset time interval, a k-1 is the k-1th operating data value of the coal gasification device operating data collected within the preset time interval, μ a is the average of all operating data values ​​of the coal gasification device operating data collected within the preset time interval; K is the total amount of coal gasification device operating data collected within the preset time interval.

[0068] Specifically, in this embodiment, the steady-state detection in step a1 is a heuristic steady-state detection; the steady-state detection process includes the following steps b1 to b4:

[0069] Step b1, calculating the light filtering value and the heavy filtering value corresponding to each valid data in the coal gasification device operation data collected in step a1; wherein the valid data in the coal gasification device operation data collected is the remaining data after the invalid data is eliminated;

[0070] (Y L ) t =f L ·y t +(1-f L )·(Y L ) t-1 ;

[0071] (Y H ) t =f H ·y t +(1-f H )·(Y H ) t-1 ;

[0072] (Y L )0=y t1 , (Y H )0=y t1 ;

[0073] t1≤t≤t Q ;

[0074] Among them, the time period (t1,t Q ) is the collection time period of the coal gasification device operation data collected in step a1, (Y L ) t is the time period (t1,t Q ) is the lightly filtered value of the valid data corresponding to the t-th time in the coal gasification device operation data collected in (Y H ) t is the time period (t1,t Q ) in the coal gasification device operation data collected in the re-filtered value of the valid data corresponding to the t-th time; (Y L ) t-1 is the time period (t1,t Q ) is the lightly filtered value of the valid data corresponding to the t-1th moment in the coal gasification device operation data collected in (Y H ) t-1 is the time period (t1,t Q ) in the coal gasification device operation data collected in the re-filtered value of the valid data corresponding to the t-1th time, f L is the light filter coefficient, f H is the re-filtering coefficient; y t is the actual value of the valid data corresponding to the tth moment; Q is the value in the time period (t1, t Q ) The total number of all valid data corresponding to the time in the coal gasification unit operation data collected in );

[0075] Step b2, respectively calculating the absolute value of the light and heavy filter value difference corresponding to each valid data in the collected coal gasification device operation data, and obtaining the absolute value maximum of all the absolute values ​​obtained; wherein the absolute value of the light and heavy filter value difference of the valid data corresponding to the t-th time in the collected coal gasification device operation data is marked as

[0076] The maximum absolute value among all the absolute values ​​obtained is marked as

[0077]

[0078] Step b3, making a parameter steady-state judgment based on the obtained absolute value maximum:

[0079] When the absolute maximum value is obtained When it is less than the preset parameter steady-state tolerance value χ, the factory value of the valid data corresponding to the time t is determined to be a steady-state value; otherwise, the factory value of the valid data corresponding to the time t is determined to be a non-steady-state value;

[0080] It should be noted that, in this embodiment, the calculation method of the preset parameter steady-state tolerance value χ is as follows:

[0081] χ=(2±b)·σ,b>0,

[0082] Among them, a k is the preset time interval (t1,t Q ), where K is the kth operating data value of the coal gasification device operating data collected within the preset time interval (t1, t Q ) The total amount of coal gasification unit operation data collected in

[0083] Step b4, making a steady-state judgment of the device according to the refiltered value of the valid data corresponding to the last moment and the refiltered value of the valid data corresponding to the first moment:

[0084] When the absolute value of the difference between the two is less than the preset trend tolerance value ε, that is, Determine the time period [t1,t Q ] the state of the coal gasification unit corresponding to the valid data within the time period [t1,t Q ] The state of the coal gasification device corresponding to the valid data in ] is non-steady state. In this embodiment, the calculation method of the preset trend tolerance value ε is as follows:

[0085] 0<ε<5%·μ a ,

[0086] Among them, a k is the preset time interval (t1,t Q ) in the coal gasification device operation data collected within the preset time interval (t1, t Q ) The total amount of coal gasification unit operation data collected.

[0087] In addition, in step a1, the data correction process includes the following steps: based on the original measurement data, using the material balance relationship or energy balance relationship of the coal gasification device production process, determining the correction condition based on the minimum sum of squares of deviations between the correction value and its corresponding measurement value, and establishing a correction mathematical expression based on the least squares method; wherein the correction mathematical expression is as follows:

[0088]

[0089]

[0090] Among them, X is the vector of measured values ​​of measured variables, and U is the vector of unmeasured or estimated parameters. is a vector composed of the correction values ​​of the measured variables, P is the variance-covariance matrix of the measurement error, and F is the constraint equation of the coal gasification unit model; wherein the constraint equation includes the material balance equation, the energy balance equation, the chemical reaction rate equation and the chemical balance equation, the heat mass and momentum transfer equation; P determines the weight of each variable for adjustment, and assigns a higher weight to instruments with higher precision.

[0091] This embodiment also provides a coal gasification device process diagnosis system for implementing the above-mentioned coal gasification device process diagnosis method. Specifically, the coal gasification device process diagnosis system of this embodiment includes:

[0092] Data acquisition module 1, to obtain a set of operating data during the operation of the coal gasification device;

[0093] The data preprocessing module 2 preprocesses the operation data in the operation data set collected by the data collection module 1 to obtain a preprocessed operation data set, and performs category processing on the preprocessed operation data set data;

[0094] The device model management module 3 manages the pre-built coal gasification device initial model and process diagnosis model for simulating the operation of the coal gasification device;

[0095] The process diagnosis module 4 is responsible for setting the assessment technical parameters and calling the diagnosis model corresponding to the category to which the preprocessed operation data set determined by the data preprocessing module belongs, and inputting the preprocessed operation data set into the process diagnosis model for diagnosis, and outputting the process diagnosis result.

[0096] Although the preferred embodiments of the present invention are described in detail above, it should be clearly understood that various modifications and variations are possible for those skilled in the art. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. Process diagnosis method of coal gasification unit, It is characterized in that The steps include: Step S1, obtaining an operation data set during the operation of the coal gasification device, and preprocessing the operation data in the operation data set to obtain a preprocessed operation data set; Step S2, using the operation data in the preprocessed operation data set to pre-build a coal gasification device initialization model library for simulating the operation of the coal gasification device; wherein the coal gasification device initialization model library includes subunit initialization models corresponding to W different subunits, and the different subunit initialization models include M feed categories, each feed category includes N operating condition categories, and each subunit initialization model corresponds to a subunit operation condition in the simulated coal gasification device, and W is the total number of sub-operation units actually included in the coal gasification device; Step S3, based on the process diagnosis requirements, set the assessment technical parameters and the constraints corresponding to each assessment technical parameter, and then develop a process diagnosis model library based on the initialization model, assessment technical parameters and constraints in the coal gasification device initialization model library; wherein the model in the process diagnosis model library is a diagnosis model capable of simulating and analyzing the key technical parameters of the coal gasification device, the key technical parameters are independent variable technical parameters relative to other technical parameters in the coal gasification device, and the other technical parameters in the coal gasification device are dependent variable technical parameters relative to the key technical parameters; Step S4, obtaining the actual operation data of the coal gasification device again, and inputting the obtained actual operation data into the process diagnosis model library for diagnosis, to obtain the process diagnosis result of the coal gasification device; wherein: In step S2, the construction process of the coal gasification unit initialization model includes the following steps a1 to a3: Step a1, collecting the coal gasification device operation data, and performing validity processing on the coal gasification device operation data to form a basic database; wherein the coal gasification device operation data includes the coal gasification device real-time database and the historical data collected by LIMS; Step a2, based on the basic database data, using clustering algorithms to perform feed clustering and operating condition clustering in sequence to form a sample data set library; Step a3, constructing a coal gasification device initialization model based on the formed sample data set library data; The validity processing in step a1 includes preprocessing, steady-state detection and data correction. The preprocessing includes invalid data elimination, interpolation fitting filling, filtering denoising and data normalization processing; and the steady-state detection is a heuristic steady-state detection; the steady-state detection process includes the following steps b1 to b4: Step b1, calculating the light filtering value and the heavy filtering value corresponding to each valid data in the coal gasification device operation data collected in step a1; wherein the valid data in the coal gasification device operation data collected is the remaining data after the invalid data is eliminated; (Y L ) t =f L ·y t +(1-f L )·(Y L ) t-1 ; (Y H ) t =f H ·y t +(1-f H )·(Y H ) t-1 ; t 1 ≤t≤t Q ; Among them, the time period (t 1 ,t Q ) is the collection time period of the coal gasification device operation data collected in step a1, (Y L ) t For the period (t 1 ,t Q ) is the lightly filtered value of the valid data corresponding to the t-th time in the coal gasification device operation data collected in (Y H ) t For the period (t 1 ,t Q ) in the coal gasification device operation data collected in the re-filtered value of the valid data corresponding to the t-th time; (Y L ) t-1 For the period (t 1 ,t Q ) is the lightly filtered value of the valid data corresponding to the t-1th moment in the coal gasification device operation data collected in (Y H ) t-1 For the period (t 1 ,t Q ) in the coal gasification device operation data collected in the re-filtered value of the valid data corresponding to the t-1th time, f L is the light filter coefficient, f H is the re-filtering coefficient; y t is the actual value of the valid data corresponding to the tth moment; Q is the value in the time period (t 1 ,t Q ) The total number of all valid data corresponding to the time in the coal gasification unit operation data collected in ); Step b2, respectively calculating the absolute value of the light and heavy filter value difference corresponding to each valid data in the collected coal gasification device operation data, and obtaining the absolute value maximum of all the absolute values ​​obtained; wherein the absolute value of the light and heavy filter value difference of the valid data corresponding to the t-th time in the collected coal gasification device operation data is marked as The maximum absolute value among all the absolute values ​​obtained is marked as Step b3, making a parameter steady-state judgment based on the obtained absolute value maximum: When the absolute maximum value is obtained When it is less than the preset parameter steady-state tolerance value χ, the factory value of the valid data corresponding to the time t is determined to be a steady-state value; otherwise, the factory value of the valid data corresponding to the time t is determined to be a non-steady-state value; Step b4, making a steady-state judgment of the device according to the refiltered value of the valid data corresponding to the last moment and the refiltered value of the valid data corresponding to the first moment: When the absolute value of the difference between the two is less than the preset trend tolerance value, Determine the time period [t 1 ,t Q ] the coal gasification unit state corresponding to the valid data within the time period [t 1 ,t Q ]The state of the coal gasification unit corresponding to the valid data in ] is non-steady state.

2. The process diagnosis method for a coal gasification device according to claim 1, It is characterized in that It also includes a process of displaying the retrieved real-time operation data in a visual manner; or / and, a process of displaying the process diagnosis result of the coal gasification device in a visual manner.

3. The process diagnosis method for a coal gasification device according to claim 1, It is characterized in that In step a2, firstly, the parameter data in the feed characteristic parameter set are clustered into M feed categories, and then each of the M feed categories is clustered into N operating condition categories according to the parameter data in the operating condition characteristic parameter set; wherein the cluster distance is calculated as follows: Among them, n and m are the numbers of the feature parameter set Y, n =(x n1 ,x n2 ,…,x nJ ), Y m =(x m1 ,x m2 ,…,x mJ ), cosθ nm is the feature parameter set Y n With the characteristic parameter set Y m The cosine distance between nj is the feature parameter set Y n The data of the jth characteristic parameter in the corresponding x mj is the feature parameter set Y m The data of the jth characteristic parameter in, J is the characteristic parameter set Y n The total number of characteristic parameters in the characteristic parameter set Y n With the characteristic parameter set Y m Have the same total number of characteristic parameters.

4. The process diagnosis method for a coal gasification device according to claim 1, It is characterized in that In step a1, the invalid data is eliminated by eliminating the operating data that meets the large error condition in the collected coal gasification device operating data, and the operating data that meets the large error condition is the operating data whose fluctuation coefficient value is greater than the preset fluctuation coefficient threshold value in the collected coal gasification device operating data; wherein, the collected coal gasification device operating data is the operating data collected within a preset time interval, and the fluctuation coefficient value of the kth operating data collected within the preset time interval is marked as τ k , the preset fluctuation coefficient threshold corresponding to the preset time interval is marked as τ th : a k is the kth operating data value of the coal gasification device operating data collected within the preset time interval, a k-1 is the k-1th operating data value of the coal gasification device operating data collected within the preset time interval, μ a is the average of all operating data values ​​of the coal gasification device operating data collected within the preset time interval; K is the total amount of coal gasification device operating data collected within the preset time interval.

5. The process diagnosis method for a coal gasification device according to claim 1, It is characterized in that The data correction includes the following steps: based on the original measurement data, using the material balance relationship or energy balance relationship of the coal gasification device production process, determining the correction condition based on the minimum sum of squares of deviations between the correction value and its corresponding measurement value, and establishing a correction mathematical expression based on the least squares method; wherein the correction mathematical expression is as follows: Among them, X is the vector of measured values ​​of measured variables, and U is the vector of unmeasured or estimated parameters. is a vector composed of the correction values ​​of the measured variables, P is the variance-covariance matrix of the measurement error, and F is the constraint equation of the coal gasification unit model; wherein the constraint equation includes the material balance equation, the energy balance equation, the chemical reaction rate equation and the chemical balance equation, the heat mass and momentum transfer equation; P determines the weight of each variable for adjustment, and assigns a higher weight to instruments with higher precision.

6. A coal gasification device process diagnosis system, which implements the coal gasification device process diagnosis method according to claim 1, It is characterized in that The coal gasification unit process diagnostic system includes: A data acquisition module is used to obtain a set of operating data during the operation of the coal gasification device; The data preprocessing module preprocesses the operation data in the operation data set collected by the data collection module to obtain the preprocessed operation data set, and performs category processing on the preprocessed operation data set data; The device model management module manages the pre-built coal gasification device initial model and process diagnosis model for simulating the operation of the coal gasification device; The process diagnosis module is responsible for the setting of the assessment technical parameters and calling the diagnosis model corresponding to the category to which the preprocessed running data set determined by the data preprocessing module belongs, and inputting the preprocessed running data set into the process diagnosis model for diagnosis and outputting the process diagnosis result.