CN103672949A - Heating furnace combustion control technology for overcoming fuel gas heat value and production rhythm fluctuation - Google Patents

Abstract

Aiming at the problems of unstable gas calorific value and frequent changes of production rhythm in the combustion control of heating furnaces, the present invention proposes a new method of combustion control. This method retains the traditional temperature-gas flow cascade control, that is, the temperature control is the main loop, and the gas flow is the secondary loop. The cascade system structure can better overcome the fluctuation of the gas pressure; the introduction of a productivity fuzzy feedforward module is superimposed on the In terms of the set value of the secondary loop, it can effectively improve the rapidity of the temperature control system following the temperature set value when the production rhythm changes. At the same time, the air flow is optimized and set according to the productivity model, and the reasonable optimal air demand can be accurately obtained according to the calculation of the heat supply, and the best air-fuel ratio can be objectively obtained, which overcomes the gas calorific value that cannot be solved in the previous combustion control technology The defect that the air-fuel ratio is difficult to automatically correct when it fluctuates. The control technology in this invention not only better solves the problem of combustion efficiency when the heat value and production rhythm fluctuate, but also can greatly improve the control quality of the furnace temperature. The control scheme is simple and easy to implement and maintain on site.

Description

A Combustion Control Technology for Heating Furnace Overcomes Calorific Value and Production Rhythm Fluctuation of Gas

technical field

The invention relates to a combustion control technology of a hot rolling heating furnace.

Background technique

The purpose of the traditional combustion control of the heating furnace is to ensure the accuracy of the furnace temperature control on the one hand, and on the other hand to ensure a reasonable ratio of gas to combustion air flow, so as to ensure that it can be fully burned and achieve the highest combustion efficiency. Excellent heating furnace control technology can not only improve the heating quality of products, but also has specific significance for energy saving and emission reduction.

In fact, most of the heating furnaces operating in China currently use a mixture of blast furnace gas and coke oven gas, and its calorific value is often affected by the working conditions of the blast furnace and coke oven and often fluctuates. The cascade ratio and double-crossing limiting combustion control strategies are still the most common control schemes in the combustion control of heating furnaces. This scheme is based on the constant calorific value of gas. When the calorific value of the mixed gas fluctuates, it will inevitably lead to a mismatch between the gas flow and the air flow, thus wasting energy. At present, the common solutions are as follows: ①The calorific value of gas is an ideal method to determine the air-fuel ratio, but the calorific value signal detection often lacks stability, and the instrument is expensive; The actual effect is not good due to the durability stability of the instrument and the large hysteresis characteristics of the object; ③The air-fuel ratio self-optimization, these methods are theoretically feasible, but in the actual process due to the existence of random fluctuations and disturbances in the process, it is difficult Guarantee a complete self-optimization process, so it is difficult to run stably for a long time. Therefore, in practical applications, most heating furnaces manually correct the air-fuel ratio through the actual experience of the operator.

Generally speaking, the production rhythm of the heating furnace follows the rolling rhythm. With the change of product specifications, roll changes, failures, etc., the rolling rhythm will inevitably change, which often leads to large fluctuations in the temperature of the heating furnace. fluctuation. The main reason for this situation is the large inertia of the temperature object. If the gas cannot be increased or decreased in time when the rolling rhythm changes, it will inevitably cause a period of gas excess or shortage. As a result, the temperature will rise or fall rapidly, and It must be slow to adjust. At present, most heating furnace temperature control inevitably requires manual intervention when the rolling rhythm fluctuates greatly.

Contents of the invention

When the calorific value of gas fluctuates, the ratio of air and fuel flow (air-fuel ratio) is no longer a fixed value, and should change with the fluctuation of calorific value. Since the self-optimization method is difficult to apply in practical applications, it is necessary to make the ratio loop Untie and take control separately. The specific control strategy is shown in Figure 1.

According to the productivity model, the air flow rate is optimized, and the reasonable optimal air demand can be obtained according to the heat supply calculation; at the same time, the temperature control is based on the cascade control system structure (using the gas flow control as a secondary loop can be better overcome Disturbance of gas pressure), in order to improve the response speed of temperature control when the production rhythm fluctuates, a fuzzy feed-forward controller is set up, which can give a reasonable gas pressure according to the value of the production rhythm (production rate) and its change rate. Flow feed-forward value, in the case that the temperature has not yet responded due to lag, the gas is adjusted in advance.

(1) Air demand calculation strategy based on productivity optimization model

From the perspective of heat supply balance, assuming that the productivity of the heating furnace is P tons/hour, the temperature of the steel when it enters the furnace is Tin, the set temperature of the furnace is Tsp, and the specific heat of the steel is C, then the steel should be heated to the desired value The heat to be absorbed per unit time is Q1=C*P*(Tsp-Tin). Assuming that the heating furnace gas is a mixture of blast furnace gas and coke oven gas, its calorific value is about 5900~9200 KJ/Nm ³ . The air ratio k (air-fuel ratio) required for fuel gas should be linearly related to the calorific value R, that is, k=n*R, where n is a constant coefficient.

Let the calorific value of the gas be R, that is, the heat that can be emitted by the complete combustion of each cubic meter of gas is R, and the heat that can be emitted by the complete combustion of the gas per unit time is Fgas*R, where Fgas is the gas flow. In addition, the exhaust gas temperature will take away part of the heat Q2, the cooling water will take away part of the heat Q3, plus the inevitable part of the heat loss Q4 of the furnace itself, including the escaped gas from the furnace door and other loose places. The heat loss caused by incomplete combustion and the heat loss caused by incomplete combustion, etc., all need to be provided by the calorific value of gas combustion. From the perspective of heating demand balance, there is Fgas*R= Q1+Q2+Q3+Q4.

Let k be the air-fuel ratio that should be set in theory, that is, k cubic meters of air are required for complete combustion of every cubic meter of gas. ) changes with changes. And there is, Fair=k*Fgas, where Fair is the ideal air flow to ensure complete combustion. So there is, Fair=k*(Q1+Q2+Q3+Q4)/R.

As has been analyzed above, the air-fuel ratio k is approximately proportional to the calorific value R of the gas, then the above formula can be written as: Fair= n*(Q1+Q2+Q3+Q4)= n*【C*P*(Tsp-Tin )+Q2+Q3+Q4]. It can be seen from the formula that under normal production conditions, Q2, Q3 and Q4 are basically stable, and Tsp and Tin are almost unchanged when the varieties are similar, so the above formula can be simply described as F=c*P+d In the formula, the values of c and d are difficult to calculate accurately. In practice, they can be calculated according to data interpolation under several typical production rates of waiting temperature, normal production and high-speed production. Therefore, the air demand is a function of the production rate P and the set value of the tapping temperature Tsp, and a reasonable air demand value can be obtained according to this model. In other words, when the production rate and tapping temperature are set constant, the air supply volume can remain constant. Even if the furnace temperature fluctuates at this time, it means that the calorific value of the gas has changed. In this case, you only need to adjust the amount of gas properly, and the furnace temperature will naturally stabilize.

When the burner works under different loads, if you want to achieve complete combustion, you must also consider the problem of air excess coefficient. As shown in Figure 2, a larger excess coefficient is required to work under a small load, and a smaller excess coefficient is required to work at full load. Therefore, the effective air-fuel ratio k1 should contain two parts: the theoretical air-fuel ratio k and the excess air coefficient b, with k1=k*b. The present invention fits the optimization curve in Fig. 2 into a nonlinear function b=f(Fgas), and the final air requirement Fairsp is: Fairsp=b*(c*P+d).

Finally, the calculation method of the "air demand model" module in Figure 1 is realized.

(2) Temperature control strategy based on productivity fuzzy feedforward

The furnace temperature control of the heating furnace adopts the cascade-fuzzy feedforward control strategy, and the structure is shown in the upper part of Figure 1. The system has the following two characteristics: first, a fuzzy feed-forward controller based on productivity is added to the original temperature control loop to improve the system's rapid response to production rhythm fluctuations; second, maintaining a cascade structure can effectively control pressure fluctuations. Better anti-interference ability.

The fuzzy feedforward controller takes the production rate P and the production rate change rate P' as the input, and the gas flow feedforward value FFSP as the output. The fuzzy controller design needs to follow the following three steps.

Step one, blurring. The parameters of the controller are selected according to the production process of a heating furnace. The fuzzy domain of productivity P is the range of all productivity (unit t/h), and the fuzzy domain of productivity change rate P' is the maximum fluctuation range that often occurs in production. The fuzzy domain of the output value FFSP is the feed-forward correction range of the gas flow. The input quantity and change quantity of the fuzzy controller are divided into n levels, and a unified fuzzy subset is defined to select the word set, and the triangular membership function is selected for fuzzification.

Step 2, making fuzzy rules. The principle of making fuzzy rules comes from the expert-level operation experience of excellent fire adjusters when the production rhythm of the heating furnace process fluctuates. Its core idea is to emphasize the importance of the rate of change, and to adjust the fuel more timely when the production rhythm fluctuates.

For example: if the productivity P and the change rate P' are both "positive", FFSP takes "negative";

When the IF productivity P is "positive" and the rate of change P' is "negative", FFSP takes "positive", and so on.

Step three, defuzzification. The maximum degree of membership method is used for defuzzification to solve the clear value of FFSP.

The fuzzy feed-forward controller mainly works when the production rhythm fluctuates. the

Implementation

Step 1: Design the structure of the control system according to Figure 1.

Step 2, determination of parameters c and d in the air demand model. Measure the data of several typical production rates under waiting temperature, normal production, and high-speed production respectively, and define them as Fair0, Fair1, Fair2, P0, P1, and P2 respectively (it will be better if more sets of data are obtained). The c and d parameters in the model are calculated by least squares interpolation.

Step three, the determination of the b parameter in the air demand model. The nonlinear functional relationship b=f(Fgas) of the curve in Figure 2 can be fitted to a high-order polynomial form, which is convenient for calculation.

Let x=Fgas, and the value of x is between 0-100, then there is b=a0+a1*x+a2*x^2, where: a0=1.3675; a1=-0.0093; a2=0.0001.

The fourth step is to determine the domain of discourse of variables during fuzzification. The parameters of the controller are selected according to the production process of a heating furnace. The fuzzy domain of productivity P is the range of all productivity (unit t/h), which is [0,200], and the fuzzy domain of productivity change rate P’ is [-50,50]. The fuzzy domain of the output value FFSP is the feed-forward correction range of the gas flow, which takes 20-30% of the upper limit of the gas range, and the value in this example is [-2000,2000]. The input quantity and change quantity of the fuzzy controller are divided into 7 levels, the fuzzy subset selection word set is: {NB, NM, NS, ZO, PS, PM, PB}, and the triangular membership function is selected for fuzzification.

Step five, the determination of fuzzy rules. According to the experience of experts, the fuzzy rule table is obtained as shown in Figure 3.

Description of drawings

Figure 1 is a schematic diagram of the combustion control strategy.

Figure 2 is a diagram of the relationship between load and excess air coefficient.

Figure 3 is a table of fuzzy rules.

Claims (2)

1. the air requirement calculative strategy based on production rate optimization model

From heat supply balance angle analysis, steel is heated to desired value is the function of productivity ratio P and target temperature Tsp in requisition for the heat Q1 absorbing in the unit interval, simultaneously, consideration exhaust gas temperature will be taken away a part of heat Q2, cooling water will be taken away a part of heat Q3, add the inevitable a part of heat loss Q4 of stove itself, comprise heat loss that heat loss that fire door and other imprecise local emergent gas are taken away and imperfect combustion cause etc., so the Q1+Q2+Q3+Q4 that always recepts the caloric; Because caloric value provides by gas-fired, gross calorific power is that gas flow Fgas and calorific value R are long-pending in the situation that of completing combustion, according to thermal balance principle--gross calorific power and total caloric receptivity equate, and the chemically correct fuel k that considers air and gas flow in completing combustion situation, can obtain theory needs air capacity Fair=k* (Q1+Q2+Q3+Q4)/R; Under the normal condition of production, Q2, Q3 and Q4 are basicly stable, therefore above formula can simply be described as the form of Fair=c*P+d, in formula, the value of c and d is difficult to accurate Calculation, can be respectively according to treating that the data interpolating in temperature, normal production, the several typical production rate of high-speed production situation calculates in practice;

While considering that burner is worked under different load, if want, realize completing combustion and also must consider the problem of coefficient of excess air, as shown in Figure 2, introduce coefficient of excess air b, have the revised air-fuel ratio k1=k*b of optimization; The present invention fits to nonlinear function b=f (Fgas) by Fig. 2 Optimal Curve, and final air requirement Fairsp is b* (c*P+d).

2. concrete implementation step

Step 1, designs control system structure according to Fig. 1;

Step 2, in air requirement model, c, d parameter determines

Measure respectively the data for the treatment of in temperature, normal production, the several typical production rate of high-speed production situation, be defined as respectively Fair0, Fair1, Fair2, P0, P1, if P2(obtains more multi-group data better effects if), utilize least square interpolation calculation to draw c, the d parameter in model;

Step 3, in air requirement model, b parameter determines

The nonlinear function b=f (Fgas) of Fig. 2 curve can fit to higher order polynomial form, is convenient to calculate; Make x=Fgas, x value, between 0-100, has b=a0+a1*x+a2*x^2, wherein: a0=1.3675; A1=-0.0093; A2=0.0001;

Step 4, the determining of obfuscation variations per hour domain

According to the selected controller parameter of heating furnace production process, comprise: the fuzzy domain of the fuzzy domain of productivity ratio P, the fuzzy domain of productivity ratio rate of change P ', output valve FFSP, the input quantity of fuzzy controller and variable quantity are all divided into 7 grades, and fuzzy subset selects word set to be: { NB, NM, NS, ZO, PS, PM, PB}, selects Triangleshape grade of membership function to carry out obfuscation;

Step 5, according to the fuzzy reasoning table obtaining according to expertise (as shown in Figure 3) system fuzzy rule.

Images