A stabilization control algorithm for blast furnace top gas pressure in TRT power generation system
By using the algorithm of fuzzy controller and center of gravity method in the TRT power generation system, the problems of disturbance and hysteresis of the blast furnace top gas pressure are solved, and the automatic stable control of the blast furnace top gas pressure is realized, which improves the system response time and control Accuracy.
Patent Information
- Application Number
- CN202210762860.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-06-29
- Publication Date
- 2024-12-03
- Estimated Expiration
- 2042-06-29
AI Technical Summary
In the TRT power generation system, there is a large disturbance and hysteresis of the gas pressure on the top of the blast furnace, resulting in large pressure fluctuations, and it is difficult for conventional PID control algorithms to achieve stable control.
Using the algorithm based on the fuzzy controller, the pressure deviation and pressure change trend are used as input parameters for fuzzy, fuzzy reasoning and defuzzy, dynamically adjust the control interval time, calculate the final output value through the center of gravity method to control the opening of the static blade, and realize the blast furnace Automatic stability control of gas pressure on the furnace top.
Automatic control of gas pressure on the top of the TRT blast furnace is realized, which improves the system response time and control accuracy, reduces pressure fluctuations, and improves the stability of the system.
Smart Images

Figure CN114995159B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of industrial process monitoring, and in particular to an algorithm suitable for stabilizing gas pressure control of a blast furnace top in a TRT power generation system. Background Art
[0002] The steel industry currently uses blast furnace residual pressure recovery devices to recover some energy. The blast furnace gas top pressure recovery turbine power generation device (TRT) utilizes the pressure energy and thermal energy of the blast furnace top gas produced by blast furnace smelting to make the gas work through a turbine expander, converting it into mechanical energy, and then converting the mechanical energy into electrical energy.
[0003] The TRT unit adjusts the blast furnace top gas pressure by the static blade opening. When controlling, the pressure must be kept stable, which makes the blast furnace easier to control and has a positive effect on the blast furnace output. However, in actual applications, the pressure has a large disturbance, that is, the pressure fluctuates greatly with the frequent loading process of the blast furnace, and the response is fast, resulting in a rapid increase or decrease in pressure. At the same time, the system controls the furnace top pressure (greater than 50 meters) at a distance by adjusting the static blade opening of the unit. The system has a certain lag. The lag inertia after the system adjustment is completed causes the system to overshoot frequently, resulting in large fluctuations in the blast furnace top gas pressure.
[0004] At present, TRT power generation regulates the gas pressure on the top of the blast furnace by using the PID control algorithm. PID is a linear controller that uses the proportional (P), integral (I), and differential (D) of the deviation between the given value and the actual feedback (e(t) = set value - feedback value) to form a control quantity through a linear combination to control the control object. PID control algorithm formula
[0005] Therefore, as long as the deviation exists, the pressure needs to be adjusted. When the deviation always exists, the integral keeps increasing, resulting in excessive integral, which will cause integral saturation. When the deviation direction changes, it takes a period of time for the output u(n) to leave the saturation zone. This causes adjustment lag and causes obvious overshoot in the system.
[0006] At the same time, when the system is disturbed, the pressure suddenly increases or decreases rapidly, and the system deviation increases rapidly. The system must be adjusted quickly. Although rapid adjustment can reduce the deviation, the lag of the system will cause over-adjustment to be overcome later. The overshoot will increase, and rapid adjustment will cause the system stability to decrease. A small fluctuation will cause the system to oscillate.
[0007] In summary, TRT regulation of blast furnace top gas pressure has the typical characteristics of large disturbance and hysteresis, and fast disturbance response speed, and it is difficult to obtain an accurate mathematical model. It is necessary to design an algorithm that meets the requirements of stable control of blast furnace top pressure and realize smooth and automatic control of blast furnace top pressure. Summary of the invention
[0008] The technical task of the present invention is to provide a TRT power generation system blast furnace top gas pressure stabilization control algorithm to achieve automatic control of the TRT blast furnace top gas pressure in view of the above deficiencies in the prior art.
[0009] The technical solution of the present invention to solve the technical problem is: a TRT power generation system blast furnace top gas pressure stabilization control algorithm, characterized in that: the algorithm is based on a fuzzy controller, the fuzzy controller includes a fuzzification interface, fuzzy reasoning, defuzzification, and a knowledge base; the calculated pressure deviation e and pressure change trend de are used as input parameters of the fuzzy controller to enter the fuzzification interface, and the two parameters are fuzzified, and the number of fuzzified membership language value variables of the two parameters is 7; according to the membership function curve, the membership of the 7 language value variables of the two variables is calculated; then according to the membership values of each language value variable of the two variables, fuzzy reasoning is performed in combination with the knowledge base to obtain the respective output values under all rules; finally, according to the centroid method, the fuzzy reasoning value is defuzzified to obtain the final output value u(t) of the fuzzy controller; according to the final output value u(t) obtained by the fuzzy control algorithm, the static blade opening is controlled, so as to obtain n furnace top pressure values for curve fitting, and the pressure change trend de is obtained; the shortest interval time and the longest interval time are set, and the interval time is obtained by linear transformation according to the pressure change trend de.
[0010] Furthermore, the number of variables i in the above-mentioned pressure deviation e membership function is 7, namely NB, NM, NS, Z0, PS, PM, and PB; among the variables, except ZO which adopts a trapezoidal membership function, the others all adopt a triangular membership function, and the membership value range is (0~1).
[0011] Furthermore, the above-mentioned blast furnace top pressure change trend de adopts a membership function language value variable number j of 7, namely NB, NM, NS, Z0, PS, PM, PB; each variable adopts a triangular membership function, and the membership value range is (0-1).
[0012] Furthermore, the above-mentioned knowledge base content is a rule derivation table.
[0013] Furthermore, the above defuzzification method adopts the centroid method, and its calculation formula is as follows:
[0014]
[0015] Furthermore, the pressure deviation e is obtained by calculation according to the set pressure and the feedback pressure.
[0016] Furthermore, the above curve fitting method specifically collects the most recent pressure, obtains the curve equation through curve fitting, and then obtains the pressure change trend according to the most recent value tangent. According to the current situation analysis, the curve fitting adopts a linear equation, that is, the least squares method is adopted; its formula is y=a+bx; the calculation formulas of a and b are:
[0017]
[0018]
[0019] Finally, b is used as the tangent of the curve as the pressure change trend de, that is, de=b as the change trend.
[0020] Furthermore, the collection cycle of the above collection is 1s, and the number of collections n is 5-20.
[0021] Compared with the prior art, the present invention has the following outstanding beneficial effects:
[0022] 1. The fuzzy control algorithm is used to realize the automatic control of the gas pressure on the top of the TRT blast furnace;
[0023] 2. According to the change rate of the blast furnace top pressure, the algorithm execution interval is dynamically adjusted to improve the system response time and the control accuracy of the system. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] Figure 1 It is the structural principle diagram of the fuzzy controller of the present invention.
[0025] Figure 2 It is a membership function curve diagram of the furnace top pressure deviation e of the present invention.
[0026] Figure 3 It is a membership function diagram of the furnace top pressure change rate of the present invention. DETAILED DESCRIPTION
[0027] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods.
[0028] For the purpose of the following detailed description, it should be understood that the present invention may take various alternative variations and step sequences unless explicitly indicated to the contrary. In addition, except in any operating examples or where otherwise indicated, all numbers expressing, for example, the amounts of ingredients used in the specification and claims should be understood in all cases as modified by the term "about". At the very least, and without attempting to limit the application of the doctrine of equivalents to the scope of the claims, each numerical parameter should at least be interpreted in terms of the number of reported significant digits and by applying ordinary rounding techniques.
[0029] Notwithstanding that the numerical ranges and parameters setting forth the broad scope of the invention are approximations, the numerical values set forth in the specific examples are reported as precisely as possible. Any numerical value, however, inherently contains certain errors necessarily resulting from the standard deviation found in their respective testing measurements.
[0030] It should also be understood that any numerical range stated herein is intended to include all subranges subsumed therein. For example, a range of "1 to 10" is intended to include all subranges between (and including) the stated minimum value of 1 and the stated maximum value of 10, that is, having a minimum value equal to or greater than 1 and a maximum value equal to or less than 10.
[0031] In this application, unless otherwise specifically stated, the use of the singular includes the plural and the plural includes the singular. In addition, in this application, unless otherwise specifically stated, the use of "or" means "and / or", even if "and / or" may be explicitly used in some cases. Further, in this application, unless otherwise specifically stated, the use of "a" or "an" means "at least one / kind". For example, "a" first material, "a" coating composition, etc. refer to one or more items in any of these items.
[0032] Since the blast furnace top gas pressure has typical characteristics of disturbance, hysteresis and fast disturbance response speed, the conventional PID control algorithm control leads to large fluctuations in the blast furnace top pressure (deviation exceeds 8kpa), which cannot meet the blast furnace top pressure stability requirements. However, the operator's real-time control can achieve better results. The operator's control strategy is mainly to observe the trend of the top pressure curve to make dynamic judgments and make rapid adjustments, thereby achieving the purpose of early prediction.
[0033] The basis for manual operation is based on two points: pressure deviation and pressure change trend. Then, based on these two parameters, output control is performed according to long-term work experience. For example, if the current pressure positive deviation is large (the actual pressure is much greater than the set value), and the pressure continues to increase slowly, it is necessary to increase the static blade opening. This process involves three parameters: pressure deviation, pressure change trend, and output value. In manual judgment, these three parameters are generally planned with fuzzy concepts of large, slightly large, just right, slightly small, and small. Experienced operators make a series of rules based on these parameters and the fuzzy values of the parameter sizes to control, and these rules are all based on the experience values solidified in the minds of the operators.
[0034] The fuzzy control algorithm improves the controllability, adaptability and rationality of the control algorithm by simulating the approximate reasoning and comprehensive decision-making process of humans, and has a better effect on solving complex systems. Therefore, the improved solution of the present invention is to use the fuzzy control algorithm to achieve automatic and stable adjustment of the top pressure.
[0035] The structure of the present invention is as follows Figure 1 As shown, the fuzzy controller includes fuzzification interface, fuzzy reasoning, defuzzification, and knowledge base.
[0036] The fuzzy interface uses two variables: pressure deviation (e) and pressure change trend (de).
[0037] (1) Blast furnace top pressure deviation (e)
[0038] Pressure deviation (e) = set pressure - actual pressure
[0039] Membership function μ(e i ) The number of language value variables i is 7, namely NB (negative large), NM (negative medium), NS (negative small), ZO (exactly right), PS (positive small), PM (positive medium), and PB (positive large). Among the variables, except ZO (zero), which adopts trapezoidal membership function, the others adopt triangular membership function. The membership value range is (0~1). The membership function curve is shown in Figure 2 shown.
[0040] (2) Blast furnace top pressure change trend (de)
[0041] In order to obtain the pressure change trend of the blast furnace top, the pressure curve must be analyzed and predicted to determine the current pressure change trend. The curve fitting method can be used for pressure curve analysis, that is, collecting the most recent pressure, obtaining the curve equation through curve fitting, and then obtaining the pressure change trend based on the tangent of the most recent value. According to the current situation analysis, the curve fitting adopts a linear equation, that is, the least squares method is used.
[0042] Its formula is y=a+bx;
[0043] The calculation formula of a,b is:
[0044]
[0045]
[0046] Finally, b is used as the tangent of the curve as the pressure change trend de, that is, de=b as the change trend.
[0047] The algorithm needs to determine the number of data collection (n) and the interval period. The collection period is 1s. If the number of data collection (n) is too long, it cannot reflect the latest trend status. If it is too short, the change is too fast. The general value is 5 to 20.
[0048] Membership function μ(de j ) The number of language value variables j is 7, namely NB (negative large), NM (negative medium), NS (negative small), ZO (just right), PS (positive small), PM (positive medium), PB (positive large). Each variable uses a triangular membership function, and the membership value range is (0~1). The membership function curve is shown in Figure 3 shown.
[0049] The form of fuzzy reasoning is: if e is A and de is B, then the output is U ij .
[0050] Among them, A is the 7 fuzzy subsets (NB, NM, NS, Z0, PS, PM, PB) on the domain of pressure deviation (e), B is the 7 fuzzy subsets (NB, NM, NS, Z0, PS, PM, PB) on the domain of pressure change trend (de), U ij The value of is summarized based on the experience of experienced operators and is adjusted and optimized in practical debugging. The final rule derivation table is the knowledge base content, and its output relationship is shown in the following table:
[0051]
[0052]
[0053] When implementing fuzzy control, many control rules are subjected to the above-mentioned inference calculations, and then the control output is obtained by combining the inference results obtained by the calculations.
[0054] The defuzzification method adopts the centroid method, and its calculation formula is as follows:
[0055]
[0056] Where μ(e i) is the ambiguity calculated by the fuzzy function corresponding to the seven fuzzy subsets (NB, NM, NS, Z0, PS, PM, PB) on the domain of pressure deviation (e); μ(de j ) is the fuzziness calculated by the fuzzy function corresponding to the 7 fuzzy subsets (NB, NM, NS, Z0, PS, PM, PB) on the pressure change trend (de) domain. The output value U is finally obtained as the incremental value of the control stator opening. A positive value indicates an increase in the stator opening, and a negative value indicates a decrease in the stator opening.
[0057] The control of the present invention is performed as follows Figure 1 As shown, firstly, the pressure deviation e is calculated based on the set pressure and the feedback pressure (furnace top pressure); the collection cycle is 1s, and the latest n furnace top pressure values are collected for curve fitting to obtain the pressure change trend de.
[0058] The calculated pressure deviation e and pressure change trend de are used as the input parameters of the fuzzy controller, and the two parameters are fuzzified. The number of fuzzy membership language value variables of the two parameters is 7. Figure 2 , Figure 3 The membership function curve of the two variables is used to calculate the membership of the seven language value variables of each variable. Then, according to the membership values of each language value variable of the two variables, fuzzy reasoning is performed in combination with the knowledge base to obtain the respective output values under all rules. Finally, according to the centroid method, the fuzzy reasoning value is defuzzified to obtain the final output value u(t) of the fuzzy controller. According to the final output value u(t) obtained by the fuzzy control algorithm, the static blade opening (controlled object) is controlled to obtain n furnace top pressure values for curve fitting and obtain the pressure change trend de.
[0059] After one output, due to the lag of the system, manual work can dynamically adjust the next output time interval according to the pressure change trend, but automatic control cannot obtain a suitable control execution cycle. If the execution cycle is too long, it cannot respond to the system's rapid change requirements; if it is too short, the system has not responded to the last output, which aggravates the overshoot of the system. Therefore, the conventional fuzzy control algorithm is not very ideal for regulating pressure, and the control interval execution time must be dynamically adjusted. In order to meet the system requirements, the interval time of the control algorithm output adopts a dynamic change value, which is adjusted according to the speed of the furnace pressure change rate, thereby solving the problem of slow response and large lag caused by the fixed execution cycle of the algorithm, and can also get a timely response to the rapid pressure change.
[0060] The dynamic interval time is obtained by adjusting the pressure in the following way: the faster the furnace pressure changes, the faster the system enters the rapid change stage and needs to be adjusted in time. At this time, the interval time is shortened, and vice versa. The system sets the shortest interval time and the longest interval time, and obtains the interval time by linear transformation according to the pressure change trend (de).
[0061] Matters not described in detail in the embodiments of the present invention are well-known technologies in the art and are well known to those skilled in the art, and will not be described again.
[0062] It should be noted that the specific embodiments of the present invention have been described in detail. For those skilled in the art, various obvious changes to the present invention without departing from the spirit and scope of the present invention are within the protection scope of the present invention.
Claims
1. A blast furnace top gas pressure stabilization control algorithm for a TRT power generation system, characterized by: The algorithm is based on a fuzzy controller, which includes a fuzzification interface, fuzzy reasoning, defuzzification, and a knowledge base; the calculated pressure deviation e and pressure change trend de are used as input parameters of the fuzzy controller and enter the fuzzification interface, and the two parameters are fuzzified, and the number of fuzzy membership language value variables of the two parameters is 7; according to the membership function curve, the membership of the 7 language value variables of the two variables is calculated; then, according to the membership values of the language value variables of the two variables, fuzzy reasoning is performed in combination with the knowledge base to obtain the respective output values under all rules; finally, according to the centroid method, the fuzzy reasoning value is defuzzified to obtain the final output value u(t) of the fuzzy controller; the static blade opening is controlled according to the final output value u(t) obtained by the fuzzy control algorithm, so as to obtain n furnace top pressure values for curve fitting and obtain the pressure change trend de; the shortest interval time and the longest interval time are set, and the interval time is obtained by linear transformation according to the pressure change trend de; The defuzzification method adopts the centroid method, and its calculation formula is as follows: The curve fitting method is specifically to collect the most recent pressure, obtain the curve equation through curve fitting, and then obtain the pressure change trend according to the most recent value tangent. According to the current situation analysis, the curve fitting adopts a linear equation, that is, the least square method is adopted; its formula is y=a+bx; the calculation formulas of a and b are: Finally, b is used as the tangent of the curve as the pressure change trend de, that is, de=b as the change trend.
2. The TRT power generation system blast furnace top gas pressure stabilization control algorithm according to claim 1 is characterized by: The pressure deviation e membership function has 7 value variables i, namely NB, NM, NS, Z0, PS, PM, and PB; among the variables, except ZO which adopts a trapezoidal membership function, the others all adopt a triangular membership function, and the membership value range is (0~1).
3. The blast furnace top gas pressure stabilization control algorithm of the TRT power generation system according to claim 1 is characterized by: The pressure change trend de adopts a membership function language value variable number j of 7, namely NB, NM, NS, Z0, PS, PM, and PB; each variable adopts a triangular membership function, and the membership value range is (0-1).
4. The TRT power generation system blast furnace top gas pressure stabilization control algorithm according to claim 1, characterized in that: The content of the knowledge base is a rule derivation table.
5. The TRT power generation system blast furnace top gas pressure stabilization control algorithm according to claim 1, characterized in that: The pressure deviation e is obtained by calculation based on the set pressure and the feedback pressure.
6. The TRT power generation system blast furnace top gas pressure stabilization control algorithm according to claim 1, characterized in that: The collection period of the collection is 1 second, and the number of collections n is 5 to 20.
