CN106710162A - Boiler furnace scaling early warning method based on decision tree system - Google Patents

Abstract

The invention discloses a boiler furnace scaling early warning method based on a decision tree system. The method comprises the following steps that step (1): boiler room environment and boiler operation parameter data A are acquired, then a boiler furnace scaling critical value B is acquired, and a furnace scaling error rate table T is established; step (2): the decision tree system is established, a contrast system is established, and logic matching is performed on the decision tree system and the contrast system; step (3): an electronic sensor acquires real-time boiler furnace data to be transmitted to the decision tree system so as to obtain the furnace scaling height value probability P; step (4): if P is greater than 0.8, boiler furnace scaling is higher than the furnace scaling critical value, and a center console gives alarm prompting; and step (5): boiler work personnel obtain the alarm prompt given by the center console and then make confirmation, judgment of the decision tree system is wrong if boiler furnace scaling occurs after confirmation and the decision tree system is corrected. Automatic judgment of boiler furnace scaling can be realized and judgment is accurate.

Description

A kind of boiler furnace flue fouling method for early warning based on decision tree system

Technical field

The invention belongs to early warning technology field, more particularly to a kind of boiler furnace flue fouling based on decision tree system h systems is pre- Alarm method.

Background technology

Current domestic each flue fouling early warning system is provided with electronic sensor prompt system.Traditional electronic sensor Principle is, by the low value high of flue fouling, to be perceived by electronic sensor and for the numerical value of each section timely to feed back to middle control System.Work points out to learn the low value high of boiler furnace flue fouling by the picture and text of central control system.But high temperature due to generator tube, The corrosivity of stove water, a certain degree of influence is caused on electronic sensor so that cause mistake to estimate in flue fouling value of feedback Value, or there is falsity, cause major accident occur with the judgement for causing boiler staff generation mistake.And sensitivity is high Electronic sensor it is expensive, replacing is difficult, and is replaced as frequently as so that producing family's very headache.So current domestic stove Courage fouling early warning system cannot accurately react the flue fouling low value high of boiler.Most electronic sensor uses electricity before this The principles of chemistry produce electrification to the free metal ion in water, and the low value high of flue fouling is pointed out by the transmission of electric signal. But it is that underwater gold belongs to that ion motion is active to cause certain interference to result that furnace temperature is too high.

The content of the invention

The purpose of the present invention is that and overcomes the deficiencies in the prior art, there is provided a kind of Boiler Furnace based on decision tree system Courage fouling method for early warning, can immediate correction electronic sensor under circumstances data error, remind boiler staff's stove The situation of courage fouling so that staff obtains an accurate flue fouling condition to ensure the operation of boiler normal table, To extend the life-span for using of electronic sensor, the maintenance cost of boiler is reduced, realizes the automatization judgement of boiler furnace flue fouling, Accuracy of judgement, no longer needs artificial judgment, mitigates the labour intensity of staff.

To achieve these goals, the invention provides a kind of pre- police of boiler furnace flue fouling based on decision tree system Method, comprises the following steps：

Step (1), acquisition boiler room environment and boiler operating parameter data A, then boiler furnace flue fouling critical value B is obtained, Mutual pace of learning in data A and critical value BCorrespondence goes out error rate table t, is by the numerical quantization in error rate table t Flue fouling error rate table T is set up after decimal between 0-1；

Step (2)：Flue fouling error rate table T in step (1) sets up decision tree as decision tree system skeleton System, while the historical data for obtaining staff's artificial judgment boiler furnace flue fouling low value high sets up contradistinction system, by decision-making Tree system carries out logic and matches with contradistinction system；

Step (3)：Real-time boiler furnace flue data are obtained by electronic sensor, and is transmitted to decision tree system, decision tree Flue fouling low value probability P high is obtained after system repeatedly training；

Step (4)：Decision tree system judges the size of flue fouling low value probability P high, if P is more than 0.8, illustrates pot The fouling of stove flue is higher than flue fouling critical value, and result is transferred to console by decision tree system, and console provides alarm； If P is less than 0.8, boiler furnace flue fouling is illustrated less than flue fouling critical value, console will not provide alarm；

Step (5)：After boiler staff obtains the alarm that console sends, to the actual flue fouling condition of boiler Confirmed, if it is confirmed that rear boiler furnace flue fouling then illustrates decision tree system misjudgment less than flue fouling critical value, this When boiler staff correct result is inputed into contradistinction system, now contradistinction system is matched with decision tree system logic again After correct decision tree system；If it is confirmed that rear boiler furnace flue fouling then illustrates that decision tree system judges higher than flue fouling critical value Correctly；

Step (6)：Repeat step (3)-(5), so constantly circulation constantly corrects decision tree system until decision tree system Accuracy of judgement, no longer needs staff's artificial judgment boiler furnace flue fouling condition.

Further, the formula of decision tree system meets in step (2)：

Wherein：X_SIt is feedback score, X_BHIt is convolution constant, K_XIt is the converse feedback number of plies, S_OIt is vector convolution constant, K_OHIt is fixed Adopted vector constant collection, f_pIt is subset probability, b_HIt is counts, K_hFor error in judgement is counted.

Beneficial effects of the present invention：The present invention can immediate correction electronic sensor under circumstances data error, carry The situation of awake boiler staff's flue fouling so that staff family obtains an accurate flue fouling condition to ensure The operation of boiler normal table, to extend the life-span for using of electronic sensor, reduces the maintenance cost of boiler, realizes Boiler Furnace The automatization judgement of courage fouling, accuracy of judgement no longer needs artificial judgment, mitigates the labour intensity of staff.

Brief description of the drawings

In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing The accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are only this Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can be with Other accompanying drawings are obtained according to these accompanying drawings.

Fig. 1 is the flow chart of the embodiment of the present invention.

Specific embodiment

Invention is further illustrated below in conjunction with the accompanying drawings, but is not limited to the scope of the present invention.

Embodiment

As shown in figure 1, a kind of boiler furnace flue fouling method for early warning based on decision tree system that the present invention is provided, including such as Lower step：

Step (1), acquisition boiler room environment and boiler operating parameter data A, then boiler furnace flue fouling critical value B is obtained, Mutual pace of learning in data A and critical value BCorrespondence goes out error rate table t, is by the numerical quantization in error rate table t Flue fouling error rate table T is set up after decimal between 0-1；

Boiler room environmental data includes：Boiler room size, there is a several usable boilers, the species of boiler, uses Time, energy supply type etc..Boiler operating parameter data include：Furnace temperature, cigarette temperature, hydraulic pressure, vapour pressure, water inlet pump discharge, burning Machine temperature, air channel data, flue etc..

Step (2)：Flue fouling error rate table T in step (1) sets up decision tree as decision tree system skeleton System, while the historical data for obtaining staff's artificial judgment boiler furnace flue fouling low value high sets up contradistinction system, by decision-making Tree system carries out logic and matches with contradistinction system；

Step (3)：Real-time boiler furnace flue data are obtained by electronic sensor, and is transmitted to decision tree system, decision tree Flue fouling low value probability P high is obtained after system repeatedly training；

Step (4)：Decision tree system judges the size of flue fouling low value probability P high, if P is more than 0.8, illustrates pot The fouling of stove flue is higher than flue fouling critical value, and result is transferred to console by decision tree system, and console provides alarm； If P is less than 0.8, boiler furnace flue fouling is illustrated less than flue fouling critical value, console will not provide alarm；

Step (5)：After boiler staff obtains the alarm that console sends, to the actual flue fouling condition of boiler Confirmed, if it is confirmed that rear boiler furnace flue fouling then illustrates decision tree system misjudgment less than flue fouling critical value, this When boiler staff correct result is inputed into contradistinction system, now contradistinction system is matched with decision tree system logic again After correct decision tree system；If it is confirmed that rear boiler furnace flue fouling then illustrates that decision tree system judges higher than flue fouling critical value Correctly；

Step (6)：Repeat step (3)-(5), so constantly circulation constantly corrects decision tree system until decision tree system Accuracy of judgement, no longer needs staff's artificial judgment boiler furnace flue fouling condition.

The formula of decision tree system meets in step (2)：

Wherein：X_SIt is feedback score, X_BHIt is convolution constant, K_XIt is the converse feedback number of plies, S_OIt is vector convolution constant, K_OHIt is fixed Adopted vector constant collection, f_pIt is subset probability, b_HIt is counts, K_hFor error in judgement is counted.

The present invention can immediate correction electronic sensor under circumstances data error, remind boiler staff's flue The situation of fouling so that staff family obtains an accurate flue fouling condition to ensure the operation of boiler normal table, To extend the life-span for using of electronic sensor, the maintenance cost of boiler is reduced, realizes the automatization judgement of boiler furnace flue fouling, Accuracy of judgement, no longer needs artificial judgment, mitigates the labour intensity of staff.

General principle of the invention, principal character and advantages of the present invention has been shown and described above.The technology of the industry Personnel it should be appreciated that the present invention is not limited to the above embodiments, simply explanation described in above-described embodiment and specification this The principle of invention, various changes and modifications of the present invention are possible without departing from the spirit and scope of the present invention, these changes Change and improvement all fall within the protetion scope of the claimed invention.The claimed scope of the invention by appending claims and its Equivalent is defined.

Claims (2)

1. a kind of boiler furnace flue fouling method for early warning based on decision tree system, it is characterised in that comprise the following steps：

Step (1), acquisition boiler room environment and boiler operating parameter data A, then boiler furnace flue fouling critical value B is obtained, according to Mutual pace of learning in data A and critical value BCorrespondence go out error rate table t, by the numerical quantization in error rate table t be 0-1 it Between decimal after set up flue fouling error rate table T；

Step (2)：Flue fouling error rate table T in step (1) sets up decision tree system as decision tree system skeleton System, while the historical data for obtaining staff's artificial judgment boiler furnace flue fouling low value high sets up contradistinction system, by decision tree System carries out logic and matches with contradistinction system；

Step (3)：Real-time boiler furnace flue data are obtained by electronic sensor, and is transmitted to decision tree system, decision tree system Flue fouling low value probability P high is obtained after repetition training；

Step (4)：Decision tree system judges the size of flue fouling low value probability P high, if P is more than 0.8, illustrates Boiler Furnace Courage fouling is higher than flue fouling critical value, and result is transferred to console by decision tree system, and console provides alarm；If P is less than 0.8, then illustrate that boiler furnace flue fouling is less than flue fouling critical value, and console will not provide alarm；

Step (5)：After boiler staff obtains the alarm that console sends, the actual flue fouling condition of boiler is carried out Confirm, if it is confirmed that rear boiler furnace flue fouling then illustrates decision tree system misjudgment less than flue fouling critical value, now pot Correct result is inputed to contradistinction system by stove staff, is repaiied after now contradistinction system is matched with decision tree system logic again Positive decision tree system；If it is confirmed that rear boiler furnace flue fouling then illustrates that decision tree system judges just higher than flue fouling critical value Really；

Step (6)：Repeat step (3)-(5), so constantly circulation constantly corrects decision tree system until decision tree system judges Accurately, staff's artificial judgment boiler furnace flue fouling condition is no longer needed.

2. a kind of boiler furnace flue fouling method for early warning based on decision tree system according to claim 1, it is characterised in that The formula of decision tree system meets in step (2)：

$\frac{{\mathrm{dX}}_S}{dt}=(1-f_p)b_H{X}_{BH}-k_h\left(\frac{\frac{X_S}{X_{BH}}}{K_X+\frac{X_S}{X_{BH}}}\right)\left(\frac{S_O}{K_{OH}+S_O}\right)X_{BH};$

Wherein：X_SIt is feedback score, X_BHIt is convolution constant, K_XIt is the converse feedback number of plies, S_OIt is vector convolution constant, K_OHFor define to Amount constant collection, f_pIt is subset probability, b_HIt is counts, K_hFor error in judgement is counted.