A multivariable coupled intelligent decoupling control method for ethylene cracking furnace

Abstract

The invention discloses a multivariable coupled intelligent decoupling control method for an ethylene cracking furnace, relates to the technical field of industrial automatic control, and solves the problems of relatively chaotic correlation between output variables and operating variables during the operation of the current ethylene cracking furnace, and relatively complicated decoupling process of the output variables and the operating variables. The method comprises the following steps: calculating variable correlation coefficients between all variables based on historical variable data; analyzing variables in a correlation coefficient matrix to obtain an operating variable set and an output variable set of the ethylene cracking furnace; analyzing all operating variables associated with the output variables based on the output variable set, the operating variable set, and a sensitivity matrix; and decoupling the output variables from the corresponding operating variables based on an output variable-operating variable table. The method obtains the coupling relationship between the output variables and the operating variables in the ethylene cracking furnace through analysis, and realizes decoupling of the output variables and the operating variables.

Description

Technical Field

[0001] The invention belongs to the technical field of industrial automatic control, and in particular relates to a multivariable coupling intelligent decoupling control method for an ethylene cracking furnace. Background Art

[0002] Ethylene cracking furnaces are important reactors in the petrochemical industry, cracking hydrocarbon feedstocks (such as naphtha, ethane, and propane) at high temperatures to produce light olefins such as ethylene and propylene. The core process takes place within the furnace tubes, where the feedstock is cracked by the heat of the flame, accompanied by complex thermodynamic and kinetic processes such as steam dilution, fuel combustion, and heat exchange. The operation of an ethylene cracking furnace involves multiple key variables, which exhibit significant correlations and coupling relationships. Systematic analysis of these variables facilitates process optimization and intelligent control.

[0003] In the prior art, multiple output variables and multiple operating variables exist simultaneously during the operation of an ethylene cracking furnace. The correlation between the output variables and the operating variables is relatively chaotic, making it difficult to accurately determine the correlation between the output variables and the operating variables. As a result, when one operating variable is changed, multiple output variables are changed simultaneously. In addition, the decoupling process of the output variables and the operating variables is relatively cumbersome and consumes a lot of manpower and material resources.

[0004] To this end, the present invention proposes a multivariable coupling intelligent decoupling control method for an ethylene cracking furnace. Summary of the Invention

[0005] The purpose of the present invention is to propose a multivariable coupled intelligent decoupling control method for an ethylene cracking furnace to solve the problems raised in the above background technology.

[0006] The technical problems to be solved by the present invention are:

[0007] How to obtain the coupling relationship between output variables and operation variables while decoupling them.

[0008] In order to achieve the above object, the present invention adopts the following technical solutions:

[0009] A multivariable coupled intelligent decoupling control method for an ethylene cracking furnace, the method comprising:

[0010] Step S1, obtaining historical variable data of the ethylene cracking furnace, and calculating the variable correlation coefficients between all variables based on the historical variable data;

[0011] Step S2, analyzing the variables in the correlation coefficient matrix to obtain an operating variable set and an output variable set of the ethylene cracking furnace;

[0012] Step S3: Analyze and obtain all the manipulated variables associated with the output variables based on the output variable set, the manipulated variable set, and the sensitivity matrix;

[0013] Step S4: acquiring real-time variable data of the ethylene cracking furnace, and decoupling the output variables from the corresponding operation variables according to the output variable-operation variable table.

[0014] Furthermore, the historical variable data are historical flame temperature, historical raw material flow, historical steam flow, historical combustion ratio, historical outlet temperature, historical coking rate, historical thermal efficiency and historical outlet pressure of the ethylene cracking furnace.

[0015] Furthermore, the step S1 includes the following sub-steps:

[0016] Step S11, setting the time node i in the historical variable data, where i=1, 2, ..., n, n is a positive integer, and then setting the variable Jj in the daily production activity of the ethylene cracking furnace, where j=a, b, ..., z;

[0017] Step S12: Integrate the historical variable data of the ethylene cracking furnace into a variable data matrix X based on the time nodes and variables. The variable data matrix X is specifically as follows:

[0018] ; Where Xi,j is the variable value of the j-th variable Xj at the i-th time node;

[0019] Step S13: normalize the variable values ​​in the variable data matrix. The normalization process is as follows:

[0020] Obtain the variable value corresponding to each column in the variable data matrix, calculate the mean and standard deviation of each column using the mean formula and standard deviation formula, subtract the variable value from the mean, divide the result by the standard deviation, and calculate the variable value Xi,j of the corresponding variable;

[0021] Step S14: Obtain the variable Xa in the ath column of the variable data matrix at all time nodes, add the variable values ​​of the variable Xa and take the average value to obtain the average variable value PJXa of the variable Xa. Similarly, obtain the variable Xb in the bth column at all time nodes and the corresponding average variable value PJXb; where a≤j, b≤j, and a<b;

[0022] Step S15: Calculate the variable correlation coefficient rab between the variable Xa and the variable Xb using the Pearson correlation coefficient formula, which is as follows:

[0023] ;

[0024] Among them, the value range of the variable correlation coefficient is [-1, 1], Xi,a is the variable value of the variable Xa in the i-th row, and Xi,b is the variable value of the variable Xb in the i-th row;

[0025] Step S16, repeating steps S14 to S15, to obtain the variable correlation coefficient between any two variables among all the variables.

[0026] Furthermore, step S2 includes the following sub-steps:

[0027] Step S21, constructing a correlation coefficient matrix R of the ethylene cracking furnace based on the variable correlation coefficients. The correlation coefficient matrix is ​​as follows:

[0028] ;

[0029] Step S22: construct a strong coupling adjacency matrix based on the variable correlation coefficient between variable Xa and variable Xb.

[0030] Step S23: Based on the coupling value in the strong coupling adjacency matrix, the influence out-degree YXC of the variable Xa is calculated by the formula, which is as follows:

[0031] , where the formula is to sum the coupling values ​​of variable Xa in the jth row of the strong coupling adjacency matrix; the out-degree of influence is the number of other variables that are affected by variable Xa and are highly correlated with variable Xa when variable Xa actively changes;

[0032] Step S24: Repeat steps S2201 to S2202 to obtain the coupling value Zba between the variable Xb and the variable Xa. Based on the coupling value in the strong coupling adjacency matrix, the influence in-degree YXR of the variable Xa is calculated by the formula. The specific formula is as follows:

[0033] , where Zba is whether variable Xb has an impact on variable Xa. If Zba=1, variable Xb has an impact on variable Xa; if Zba=0, variable Xb has no impact on variable Xa. The formula is to add up the coupling values ​​of variable Xa in the j-th column in the strong coupling adjacency matrix, that is, the number of variables that actively affect variable Xa. The influence in-degree is the number of other variables that are highly correlated with variable Xa and will have an impact on variable Xa.

[0034] Furthermore, the step S2 further includes the following sub-steps:

[0035] Step S25: When the influence in-degree is zero, it is determined that the influence out-degree of variable Xa is greater than the influence in-degree, and the variable type of variable Xa is an operation variable;

[0036] When the influence in-degree is not zero, the control dominant fraction KZF of variable Xa is calculated by the formula KZF=YXC / YXR;

[0037] Step S26: If the control dominance score of variable Xa is greater than or equal to 1, it is determined that the influence out-degree of variable Xa is greater than the influence in-degree, and the variable type of variable Xa is a manipulated variable;

[0038] If the control dominance score of variable Xa is less than one, it is determined that the influence out-degree of variable Xa is less than the influence in-degree, and the variable type of variable Xa is an output variable;

[0039] Step S27 , repeating steps S23 to S26 , analyzing and obtaining the variable types of all variables, merging all operation variables into an operation variable set, and merging all output variables into an output variable set.

[0040] Furthermore, the construction process of the strong coupling adjacency matrix is ​​as follows:

[0041] Step S2201: When the correlation coefficient of any variable in the correlation coefficient matrix is ​​greater than or equal to the correlation threshold, it is determined that the correlation between the corresponding variable Xa and the variable Xb is highly correlated, and the variables Xa and Xb are coupled, and the process proceeds to the next step.

[0042] When the correlation coefficient of any variable is less than the correlation threshold, the correlation between the corresponding variable Xa and variable Xb is judged to be low correlation, and there is no coupling between the variable Xa and variable Xb;

[0043] Step S2202, setting the coupling value Zab between the variable Xa and the variable Xb;

[0044] When variables Xa and Xb are highly correlated, Zab=1;

[0045] When variables Xa and Xb are lowly correlated, Zab=0;

[0046] Step S2203: construct a strong coupling adjacency matrix Z based on the coupling values ​​of variables Xa and Xb. The strong coupling adjacency matrix is ​​specifically as follows:

[0047] .

[0048] Furthermore, step S3 includes the following sub-steps:

[0049] Step S31: Obtain the manipulated variables CZBe in the manipulated variable set, where e=1, 2, 3, 4, and the output variables SCBf in the output variable set, where f=1, 2, 3, 4. Construct a sensitivity matrix G based on the output variable set and the manipulated variable set. The sensitivity matrix is ​​as follows:

[0050] , where Gfe is specifically the influence degree value, which indicates the influence degree of the operating variable on the output variable;

[0051] Step S32, calculating all the influence degree values ​​in the sensitivity matrix by the least square method;

[0052] Step S33: when any influence degree value in the sensitivity matrix is ​​greater than or equal to the influence threshold, the corresponding operating variable is associated with the output variable;

[0053] When any impact value in the sensitivity matrix is ​​less than the impact threshold, no operation is performed;

[0054] Step S34: Count all the operation variables associated with any output variable in the output variable set and construct an output variable-operation variable table.

[0055] Furthermore, the real-time variable data are the real-time flame temperature, real-time raw material flow, real-time steam flow, real-time combustion ratio, real-time outlet temperature, real-time coking rate, real-time thermal efficiency and real-time outlet pressure of the ethylene cracking furnace.

[0056] Furthermore, step S4 includes the following sub-steps:

[0057] Step S41, taking the first manipulated variable corresponding to the output variable in the output variable-manipulated variable table as a candidate master control variable;

[0058] Step S42: Automatically control candidate master control variables corresponding to historical output variables according to the order of output variables from top to bottom in the output variable-operation variable table; wherein the real-time output variables in the real-time variable data are real-time outlet temperature, real-time coking rate, real-time thermal efficiency, and real-time outlet pressure; and the real-time operation variables in the real-time variable data are specifically real-time flame temperature, real-time raw material flow rate, real-time steam flow rate, and real-time combustion ratio;

[0059] Step S43: acquiring all real-time output variables after the automatic control candidate master control variables, and analyzing the real-time output variables to obtain the master control variables corresponding to the real-time output variables.

[0060] Furthermore, the analysis process of the main control variables corresponding to the real-time output variables is as follows:

[0061] Step S4301: Subtract the variable values ​​of all real-time output variables from the variable values ​​of corresponding historical output variables to obtain the change of the output variables;

[0062] Step S4302: When the ethylene cracking furnace automatically controls the candidate master control variable, if there is only one real-time output variable whose change is greater than or equal to the change threshold, and the candidate master control variable matches the corresponding real-time output variable, then the corresponding candidate master control variable is recorded as the master control variable;

[0063] Step S4303: When, among the candidate master control variables for automatic control of the ethylene cracking furnace, there is only one real-time output variable whose change is greater than or equal to the change threshold, and the candidate master control variable does not match the corresponding real-time output variable, the candidate master control variable is removed from the real-time operation variables corresponding to the real-time output variables, and the second-ranked real-time operation variable is selected as the candidate master control variable. Steps S41 to S43 are repeated until the master control variable corresponding to the output variable is obtained.

[0064] Step S4304: When, among the candidate master control variables for automatic control of the ethylene cracking furnace, there are two or more real-time output variables whose change amounts are greater than or equal to the change amount threshold, or there is no real-time output variable whose change amount is greater than or equal to the change amount threshold, the corresponding candidate master control variable is eliminated from the manipulated variables corresponding to the output variables, and the manipulated variable ranked second is selected as the candidate master control variable. Steps S41 to S43 are repeated until the master control variable corresponding to the output variable is obtained.

[0065] Step S4305: repeat the above steps. When all the output variables corresponding to the main controlled variables are obtained, it is determined that the decoupling of the ethylene cracking furnace is completed.

[0066] Compared with the prior art, the present invention has the following beneficial effects:

[0067] The present invention calculates variable correlation coefficients between all variables based on historical variable data, then constructs a correlation coefficient matrix of an ethylene cracking furnace based on the variable correlation coefficients, analyzes the variables in the correlation coefficient matrix, and obtains an operating variable set and an output variable set of the ethylene cracking furnace through analysis. Based on the output variable set, the operating variable set, and the sensitivity matrix, all operating variables associated with the output variables are analyzed and obtained. Finally, based on an output variable-operating variable table and in combination with real-time variable data of the ethylene cracking furnace, the output variables are decoupled from corresponding operating variables. The present invention can analyze and obtain the coupling relationship between the output variables and the operating variables in the ethylene cracking furnace, and realize the decoupling of the output variables and the operating variables. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] To facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0069] Figure 1 is a flow chart of the method of the present invention;

[0070] Figure 2 This is a determination flow chart of steps S1 to S4 in the present invention;

[0071] Figure 3 It is a structural schematic diagram of the electronic device in the present invention. DETAILED DESCRIPTION

[0072] The technical solutions of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0073] Example 1: Please refer to Figure 1 and Figure 2 As shown, the technical solution provided by the present invention is: a multivariable coupled intelligent decoupling control method for an ethylene cracking furnace, the method is specifically as follows:

[0074] Step S1, obtaining historical variable data of the ethylene cracking furnace, and calculating the variable correlation coefficients between all variables based on the historical variable data;

[0075] The variables in the ethylene cracking furnace are flame temperature, feed flow, steam flow, combustion ratio, outlet temperature, coking rate, thermal efficiency, outlet pressure, gas pressure, product yield and tail gas content.

[0076] In practice, in this method, the historical variable data are the historical flame temperature, historical raw material flow rate, historical steam flow rate, historical combustion ratio, historical outlet temperature, historical coking rate, historical thermal efficiency and historical outlet pressure of the ethylene cracking furnace;

[0077] In specific implementation, the historical variable data consists of the corresponding variable values ​​of all variables of the ethylene cracking furnace in the past week;

[0078] In this embodiment, step S1 includes the following sub-steps:

[0079] Step S11, setting the time node i in the historical variable data, where i=1, 2, ..., n, n is a positive integer, and then setting the variable Jj in the daily production activity of the ethylene cracking furnace, where j=a, b, ..., z;

[0080] Specifically, the time node i is: the time node at which the ethylene cracking furnace stops production during the daily production activities of the ethylene cracking furnace;

[0081] Step S12: Integrate the historical variable data of the ethylene cracking furnace into a variable data matrix X based on the time nodes and variables. The variable data matrix X is specifically as follows:

[0082] ;

[0083] Among them, Xi,j is the variable value of the j-th variable Xj at the i-th time node;

[0084] Step S13: normalize the variable values ​​in the variable data matrix. The normalization process is as follows:

[0085] Obtain the variable value corresponding to each column in the variable data matrix, calculate the mean and standard deviation of each column using the mean formula and standard deviation formula, subtract the variable value from the mean, divide the result by the standard deviation, and calculate the variable value Xi,j of the corresponding variable;

[0086] Step S14: Obtain the variable Xa in the ath column of the variable data matrix at all time nodes, add the variable values ​​of the variable Xa and take the average value to obtain the average variable value PJXa of the variable Xa. Similarly, obtain the variable Xb in the bth column at all time nodes and the corresponding average variable value PJXb.

[0087] Where a≤j, b≤j, and a<b;

[0088] Step S15: Calculate the variable correlation coefficient rab between the variable Xa and the variable Xb using the Pearson correlation coefficient formula. The formula is as follows:

[0089] ;

[0090] Among them, the value range of the variable correlation coefficient is [-1, 1];

[0091] Specifically, Xi,a is the variable value of the variable Xa in the i-th row, and Xi,b is the variable value of the variable Xb in the i-th row;

[0092] Step S16, repeating steps S14 to S15, to obtain the variable correlation coefficient between any two variables among all the variables.

[0093] Step S2, analyzing the variables in the correlation coefficient matrix to obtain an operating variable set and an output variable set of the ethylene cracking furnace;

[0094] In this embodiment, step S2 includes the following sub-steps:

[0095] Step S21, constructing a correlation coefficient matrix R of the ethylene cracking furnace based on the variable correlation coefficients. The correlation coefficient matrix is ​​specifically as follows:

[0096] ;

[0097] Step S22: construct a strong coupling adjacency matrix based on the variable correlation coefficient between the variable Xa and the variable Xb. The construction process is as follows:

[0098] Step S2201: When the correlation coefficient of any variable in the correlation coefficient matrix is ​​greater than or equal to the correlation threshold, it is determined that the correlation between the corresponding variable Xa and the variable Xb is highly correlated, and the variables Xa and Xb are coupled, and the process proceeds to the next step.

[0099] When the correlation coefficient of any variable is less than the correlation threshold, the correlation between the corresponding variable Xa and variable Xb is judged to be low correlation, and there is no coupling between the variable Xa and variable Xb;

[0100] For example, in the correlation coefficient matrix, if the variable correlation coefficient r12 is greater than or equal to the correlation threshold, it is determined that the variable X1 and the variable X2 are highly correlated, and there is coupling between the variable X1 and the variable X2;

[0101] The variable correlation coefficient r14 is less than the correlation threshold, and it is determined that variables X1 and X4 are lowly correlated, and there is no coupling between variables X1 and X4;

[0102] Step S2202, setting the coupling value Zab between the variable Xa and the variable Xb;

[0103] When variables Xa and Xb are highly correlated, Zab=1;

[0104] When variables Xa and Xb are lowly correlated, Zab=0;

[0105] It should be noted that Zab specifically refers to whether variable Xa has an impact on variable Xb. If Zab=1, variable Xa has an impact on variable Xb; if Zab=0, variable Xa has no impact on variable Xb.

[0106] Step S2203: construct a strong coupling adjacency matrix Z based on the coupling values ​​of variables Xa and Xb. The strong coupling adjacency matrix is ​​specifically as follows:

[0107] , where the coupling values ​​in the strong coupling adjacency matrix are only examples; the number of coupling values ​​in the strong coupling adjacency matrix is ​​j×j;

[0108] Step S23: Based on the coupling value in the strong coupling adjacency matrix, the influence out-degree YXC of the variable Xa is calculated by the formula, which is as follows:

[0109] , where the formula is specifically to sum the coupling values ​​of variable Xa in the j-th row in the strong coupling adjacency matrix;

[0110] Specifically, the out-degree of influence is the number of variables corresponding to all other variables that are affected by variable Xa and are highly correlated with variable Xa when variable Xa actively changes;

[0111] Step S24: Repeat steps S2201 to S2202 to obtain the coupling value Zba between the variable Xb and the variable Xa. Based on the coupling value in the strong coupling adjacency matrix, the influence in-degree YXR of the variable Xa is calculated by the formula. The specific formula is as follows:

[0112] , where Zba specifically refers to whether variable Xb has an impact on variable Xa. If Zba=1, variable Xb has an impact on variable Xa; if Zba=0, variable Xb has no impact on variable Xa;

[0113] The formula is specifically to add up the coupling values ​​of variable Xa in the jth column of the strong coupling adjacency matrix, that is, the number of variables corresponding to the variables that actively affect variable Xa;

[0114] Specifically, the influence in-degree is the number of other variables that are highly correlated with variable Xa and will have an impact on variable Xa;

[0115] Step S25: When the influence in-degree is zero, it is determined that the influence out-degree of variable Xa is greater than the influence in-degree, and the variable type of variable Xa is an operation variable;

[0116] When the influence in-degree is not zero, the control dominant fraction KZF of the variable Xa is calculated by the formula, which is as follows:

[0117] KZF=YXC / YXR;

[0118] Step S26: If the control dominance score of variable Xa is greater than or equal to 1, it is determined that the influence out-degree of variable Xa is greater than the influence in-degree, and the variable type of variable Xa is a manipulated variable;

[0119] If the control dominance score of variable Xa is less than one, it is determined that the influence out-degree of variable Xa is less than the influence in-degree, and the variable type of variable Xa is an output variable;

[0120] Among them, the manipulated variables are specifically the variables that need to be manually regulated; the output variables are specifically the variables that are affected when the manipulated variables change;

[0121] Step S27, repeating steps S23 to S26, analyzing and obtaining the variable types of all variables, merging all operation variables into an operation variable set, and merging all output variables into an output variable set;

[0122] In practice, the variables in the set of operational variables are specifically the historical flame temperature, historical feedstock flow rate, historical steam flow rate, and historical combustion ratio of the ethylene cracking furnace;

[0123] The variables in the output variable set are specifically the historical outlet temperature, historical coking rate, historical thermal efficiency and historical outlet pressure of the ethylene cracking furnace.

[0124] Step S3: Analyze and obtain all the manipulated variables associated with the output variables based on the output variable set, the manipulated variable set, and the sensitivity matrix;

[0125] In this embodiment, step S3 includes the following sub-steps:

[0126] Step S31: Obtain the manipulated variables CZBe in the manipulated variable set, where e=1, 2, 3, 4, and the output variables SCBf in the output variable set, where f=1, 2, 3, 4. Construct a sensitivity matrix G based on the output variable set and the manipulated variable set. The sensitivity matrix is ​​as follows:

[0127] , where Gfe is specifically the influence degree value, which indicates the influence degree of the operating variable on the output variable;

[0128] Step S32, calculating all the influence degree values ​​in the sensitivity matrix by the least square method;

[0129] Among them, calculating the influence degree value in the sensitivity matrix by the least square method is an existing technology;

[0130] For example, when the impact value G 32 =4, which means that every time the manipulated variable CZB2 changes by one unit, the corresponding output variable SCB3 changes by four units;

[0131] Step S33: when any influence degree value in the sensitivity matrix is ​​greater than or equal to the influence threshold, the corresponding operating variable is associated with the output variable;

[0132] When any impact value in the sensitivity matrix is ​​less than the impact threshold, no operation is performed;

[0133] In step S34, all operation variables associated with any output variable in the output variable set are counted and an output variable-operation variable table is constructed. The output variable-operation variable table is as follows:

[0134] Output variable-operation variable table:

[0135]

[0136] Among them, in the output variable-operation variable table, the operation variables are sorted in descending order according to the influence degree value.

[0137] Step S4, obtaining real-time variable data of the ethylene cracking furnace, and decoupling the output variable from the corresponding operation variable according to the output variable-operation variable table;

[0138] The real-time variable data specifically include the real-time flame temperature, real-time raw material flow, real-time steam flow, real-time combustion ratio, real-time outlet temperature, real-time coking rate, real-time thermal efficiency and real-time outlet pressure of the ethylene cracking furnace;

[0139] In this embodiment, step S4 includes the following sub-steps:

[0140] Step S41, taking the first manipulated variable corresponding to the output variable in the output variable-manipulated variable table as a candidate master control variable;

[0141] Step S42, automatically controlling the candidate master control variables corresponding to the historical output variables according to the order of the output variables from top to bottom in the output variable-operation variable table;

[0142] Among them, the real-time output variables in the real-time variable data are real-time outlet temperature, real-time coking rate, real-time thermal efficiency and real-time outlet pressure;

[0143] The real-time operation variables in the real-time variable data are specifically real-time flame temperature, real-time raw material flow, real-time steam flow and real-time combustion ratio;

[0144] In specific implementation, automatically controlling the candidate master control variable specifically involves increasing or decreasing the variable value of the candidate master control variable by a fixed ratio;

[0145] Step S43: Obtain all real-time output variables after the automatic control candidate master variables, and analyze the real-time output variables to obtain the master variables corresponding to the real-time output variables. The specific analysis process is as follows:

[0146] Step S4301: Subtract the variable values ​​of all real-time output variables from the variable values ​​of corresponding historical output variables to obtain the change of the output variables;

[0147] Step S4302: When the ethylene cracking furnace automatically controls the candidate master control variable, if there is only one real-time output variable whose change is greater than or equal to the change threshold, and the candidate master control variable matches the corresponding real-time output variable, then the corresponding candidate master control variable is recorded as the master control variable;

[0148] Step S4303: When, among the candidate master control variables for automatic control of the ethylene cracking furnace, there is only one real-time output variable whose change is greater than or equal to the change threshold, and the candidate master control variable does not match the corresponding real-time output variable, the candidate master control variable is removed from the real-time operation variables corresponding to the real-time output variables, and the second-ranked real-time operation variable is selected as the candidate master control variable. Steps S41 to S43 are repeated until the master control variable corresponding to the output variable is obtained.

[0149] Step S4304: When, among the candidate master control variables for automatic control of the ethylene cracking furnace, there are two or more real-time output variables whose change amounts are greater than or equal to the change amount threshold, or there is no real-time output variable whose change amount is greater than or equal to the change amount threshold, the corresponding candidate master control variable is eliminated from the manipulated variables corresponding to the output variables, and the manipulated variable ranked second is selected as the candidate master control variable. Steps S41 to S43 are repeated until the master control variable corresponding to the output variable is obtained.

[0150] Step S4305: repeat the above steps. When all the output variables corresponding to the main controlled variables are obtained, it is determined that the decoupling of the ethylene cracking furnace is completed.

[0151] In this application, if a corresponding calculation formula appears, the above calculation formula is dimensionless and its numerical calculation is performed. The weight coefficient, proportional coefficient and other coefficients in the formula are set to a result value obtained by quantifying each parameter. Regarding the size of the weight coefficient and the proportional coefficient, as long as it does not affect the proportional relationship between the parameter and the result value, it is acceptable.

[0152] Example 2: The present invention also provides a computer device for running the multivariable coupled intelligent decoupling control method for an ethylene cracking furnace; see Figure 3 The structure diagram of a computer device provided by an embodiment of the present invention is shown, wherein the computer device includes a memory and a processor, wherein the memory is used to store one or more computer instructions, and the one or more computer instructions are executed by the processor to implement the above-mentioned multi-variable coupled intelligent decoupling control method for an ethylene cracking furnace;

[0153] Further, Figure 3 The computer device shown further includes a system bus and a communication interface, and the processor, the communication interface and the memory are connected via the communication bus;

[0154] Among them, the memory may include high-speed random access memory (RAM), and may also include non-volatile memory (non-volatile memory), such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface (which can be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. can be used. The system bus can be an ISA bus, PCI bus or EISA bus, etc. The system bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, Figure 3 Only one bidirectional arrow is used, but it does not mean that there is only one communication bus or one type of system bus;

[0155] The processor may be an integrated circuit chip with signal processing capabilities. During implementation, each step of the above method may be completed by hardware integrated logic circuits in the processor or by software instructions. The above processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The methods, steps, and logic block diagrams disclosed in the embodiments of the present invention may be implemented or executed. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in conjunction with the embodiments of the present invention may be directly implemented and executed by a hardware decoding processor, or by a combination of hardware and software modules in the decoding processor. The software module may be located in a storage medium well-known in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or the like. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the method of the above embodiment in combination with its hardware.

[0156] Embodiment 3: The embodiment of the present invention further provides a computer storage medium, which stores computer-executable instructions. When the computer-executable instructions are called and executed by a processor, the computer-executable instructions prompt the processor to implement the above-mentioned multivariable coupled intelligent decoupling control method for an ethylene cracking furnace. The specific implementation can be found in the method embodiment and will not be repeated here.

[0157] A computer program product for a multivariable coupled intelligent decoupling control method for an ethylene cracking furnace provided in an embodiment of the present invention includes a computer storage medium storing program code. The instructions included in the program code can be used to execute the method in the previous method embodiment. The specific implementation can be found in the method embodiment and will not be repeated here.

[0158] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working process of the system and / or device described above can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0159] In addition, in the description of the embodiments of the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; mechanical connections or electrical connections; direct connections or indirect connections through an intermediate medium; and internal connections between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.

[0160] If the functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as USB flash drives, mobile hard drives, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0161] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multivariable coupled intelligent decoupling control method for an ethylene cracking furnace, characterized in that: Methods include: Step S1, obtaining historical variable data of the ethylene cracking furnace, and calculating the variable correlation coefficients between all variables based on the historical variable data; The step S1 includes the following sub-steps: Step S11, setting the time node i in the historical variable data, where i=1, 2, ..., n, n is a positive integer, and then setting the variable Jj in the daily production activity of the ethylene cracking furnace, where j=a, b, ..., z; Step S12: Integrate the historical variable data of the ethylene cracking furnace into a variable data matrix X based on the time nodes and variables. The variable data matrix X is specifically as follows: ; Where Xi,j is the variable value of the j-th variable Xj at the i-th time node; Step S13: normalize the variable values ​​in the variable data matrix. The normalization process is as follows: Obtain the variable value corresponding to each column in the variable data matrix, calculate the mean and standard deviation of each column using the mean formula and standard deviation formula, subtract the variable value from the mean, divide the result by the standard deviation, and calculate the variable value Xi,j of the corresponding variable; Step S14: Obtain the variable Xa in the ath column of the variable data matrix at all time nodes, add the variable values ​​of the variable Xa and take the average value to obtain the average variable value PJXa of the variable Xa. Similarly, obtain the variable Xb in the bth column at all time nodes and the corresponding average variable value PJXb; where a≤j, b≤j, and a<b; Step S15: Calculate the variable correlation coefficient rab between the variable Xa and the variable Xb using the Pearson correlation coefficient formula, which is as follows: ； Among them, the value range of the variable correlation coefficient is [-1, 1], Xi,a is the variable value of the variable Xa in the i-th row, and Xi,b is the variable value of the variable Xb in the i-th row; Step S16, repeating steps S14 to S15 to obtain the variable correlation coefficient between any two variables among all variables; Step S2, analyzing the variables in the correlation coefficient matrix to obtain an operating variable set and an output variable set of the ethylene cracking furnace; The step S2 includes the following sub-steps: Step S21, constructing a correlation coefficient matrix R of the ethylene cracking furnace based on the variable correlation coefficients. The correlation coefficient matrix is ​​as follows: ； Step S22: construct a strong coupling adjacency matrix based on the variable correlation coefficient between the variable Xa and the variable Xb. Step S23: Based on the coupling value in the strong coupling adjacency matrix, the influence out-degree YXC of the variable Xa is calculated by the formula, which is as follows: , where the formula is to sum the coupling values ​​of variable Xa in the jth row of the strong coupling adjacency matrix; the out-degree of influence is the number of other variables that are affected by variable Xa and are highly correlated with variable Xa when variable Xa actively changes; Step S24: Repeat steps S2201 to S2202 to obtain the coupling value Zba between the variable Xb and the variable Xa. Based on the coupling value in the strong coupling adjacency matrix, the influence in-degree YXR of the variable Xa is calculated by the formula. The specific formula is as follows: , where Zba is whether variable Xb has an impact on variable Xa. If Zba=1, variable Xb has an impact on variable Xa; if Zba=0, variable Xb has no impact on variable Xa. The formula is to add up the coupling values ​​of variable Xa in the jth column of the strong coupling adjacency matrix, that is, the number of variables that actively affect variable Xa. The influence in-degree is the number of other variables that are highly correlated with variable Xa and will have an impact on variable Xa. Step S25: When the influence in-degree is zero, it is determined that the influence out-degree of variable Xa is greater than the influence in-degree, and the variable type of variable Xa is an operation variable; When the influence in-degree is not zero, the control dominant fraction KZF of variable Xa is calculated by the formula KZF=YXC / YXR; Step S26: If the control dominance score of variable Xa is greater than or equal to 1, it is determined that the influence out-degree of variable Xa is greater than the influence in-degree, and the variable type of variable Xa is a manipulated variable; If the control dominance score of variable Xa is less than one, it is determined that the influence out-degree of variable Xa is less than the influence in-degree, and the variable type of variable Xa is an output variable; Step S27, repeating steps S23 to S26, analyzing and obtaining the variable types of all variables, merging all operation variables into an operation variable set, and merging all output variables into an output variable set; Step S3: Analyze and obtain all the manipulated variables associated with the output variables based on the output variable set, the manipulated variable set, and the sensitivity matrix; Step S4: acquiring real-time variable data of the ethylene cracking furnace, and decoupling the output variables from the corresponding operation variables according to the output variable-operation variable table.

2. The multivariable coupled intelligent decoupling control method for an ethylene cracking furnace according to claim 1, characterized in that: The historical variable data include the historical flame temperature, historical raw material flow, historical steam flow, historical combustion ratio, historical outlet temperature, historical coking rate, historical thermal efficiency and historical outlet pressure of the ethylene cracking furnace.

3. The multivariable coupled intelligent decoupling control method for an ethylene cracking furnace according to claim 1, characterized in that: The construction process of the strong coupling adjacency matrix is ​​as follows: Step S2201: When the correlation coefficient of any variable in the correlation coefficient matrix is ​​greater than or equal to the correlation threshold, it is determined that the correlation between the corresponding variable Xa and the variable Xb is highly correlated, and the variables Xa and Xb are coupled, and the process proceeds to the next step. When the correlation coefficient of any variable is less than the correlation threshold, the correlation between the corresponding variable Xa and variable Xb is judged to be low correlation, and there is no coupling between the variable Xa and variable Xb; Step S2202, setting the coupling value Zab between the variable Xa and the variable Xb; When variables Xa and Xb are highly correlated, Zab=1; When variables Xa and Xb are lowly correlated, Zab=0; Step S2203: construct a strong coupling adjacency matrix Z based on the coupling values ​​of variables Xa and Xb. The strong coupling adjacency matrix is ​​specifically as follows: 。 4. The multivariable coupled intelligent decoupling control method for an ethylene cracking furnace according to claim 1, characterized in that: The step S3 includes the following sub-steps: Step S31: Obtain the manipulated variables CZBe in the manipulated variable set, where e=1, 2, 3, 4, and the output variables SCBf in the output variable set, where f=1, 2, 3, 4. Construct a sensitivity matrix G based on the output variable set and the manipulated variable set. The sensitivity matrix is ​​as follows: , where Gfe is specifically the influence degree value, which indicates the influence degree of the operating variable on the output variable; Step S32, calculating all the influence degree values ​​in the sensitivity matrix by the least square method; Step S33: when any influence degree value in the sensitivity matrix is ​​greater than or equal to the influence threshold, the corresponding operating variable is associated with the output variable; When any impact value in the sensitivity matrix is ​​less than the impact threshold, no operation is performed; Step S34: Count all the operation variables associated with any output variable in the output variable set and construct an output variable-operation variable table.

5. The multivariable coupled intelligent decoupling control method for an ethylene cracking furnace according to claim 4, characterized in that: The real-time variable data include the real-time flame temperature, real-time raw material flow, real-time steam flow, real-time combustion ratio, real-time outlet temperature, real-time coking rate, real-time thermal efficiency and real-time outlet pressure of the ethylene cracking furnace.

6. The multivariable coupled intelligent decoupling control method for an ethylene cracking furnace according to claim 5, characterized in that: The step S4 includes the following sub-steps: Step S41, taking the first manipulated variable corresponding to the output variable in the output variable-manipulated variable table as a candidate master control variable; Step S42: Automatically control candidate master control variables corresponding to historical output variables according to the order of output variables from top to bottom in the output variable-operation variable table; wherein the real-time output variables in the real-time variable data are real-time outlet temperature, real-time coking rate, real-time thermal efficiency, and real-time outlet pressure; and the real-time operation variables in the real-time variable data are specifically real-time flame temperature, real-time raw material flow rate, real-time steam flow rate, and real-time combustion ratio; Step S43: acquiring all real-time output variables after the automatic control candidate master control variables, and analyzing the real-time output variables to obtain the master control variables corresponding to the real-time output variables.

7. The multivariable coupled intelligent decoupling control method for an ethylene cracking furnace according to claim 6, characterized in that: The analysis process of the main control variables corresponding to the real-time output variables is as follows: Step S4301: Subtract the variable values ​​of all real-time output variables from the variable values ​​of corresponding historical output variables to obtain the change of the output variables; Step S4302: When the ethylene cracking furnace automatically controls the candidate master control variable, if there is only one real-time output variable whose change is greater than or equal to the change threshold, and the candidate master control variable matches the corresponding real-time output variable, then the corresponding candidate master control variable is recorded as the master control variable; Step S4303: When, among the candidate master control variables for automatic control of the ethylene cracking furnace, there is only one real-time output variable whose change is greater than or equal to the change threshold, and the candidate master control variable does not match the corresponding real-time output variable, the candidate master control variable is removed from the real-time operation variables corresponding to the real-time output variables, and the second-ranked real-time operation variable is selected as the candidate master control variable. Steps S41 to S43 are repeated until the master control variable corresponding to the output variable is obtained. Step S4304: When, among the candidate master control variables for automatic control of the ethylene cracking furnace, there are two or more real-time output variables whose change amounts are greater than or equal to the change amount threshold, or there is no real-time output variable whose change amount is greater than or equal to the change amount threshold, the corresponding candidate master control variable is eliminated from the manipulated variables corresponding to the output variables, and the manipulated variable ranked second is selected as the candidate master control variable. Steps S41 to S43 are repeated until the master control variable corresponding to the output variable is obtained. Step S4305: repeat the above steps. When all the output variables corresponding to the main controlled variables are obtained, it is determined that the decoupling of the ethylene cracking furnace is completed.

Images