Boiler tube fault diagnosis and early warning method and system based on boiler wall temperature monitoring

Abstract

This invention discloses a boiler tube fault diagnosis and early warning method and system based on boiler wall temperature monitoring. The method obtains the wall temperature of all boiler tubes in the heat exchange equipment at the current moment and selects the n boiler tubes with the highest wall temperatures. It then obtains the ranking of the wall temperatures of the n boiler tubes from highest to lowest at multiple historical time points. Based on the ranking positions of the n boiler tubes at the historical and current time points, it determines the ranking distance feature value of the n boiler tubes and determines the state of the heat exchange equipment based on the ranking distance feature value. Finally, based on the ranking positions of the n boiler tubes at multiple historical time points, it determines the wall temperature early warning coefficient for each boiler tube and determines the state of each boiler tube. This method provides an effective solution to the problems of untimely detection of boiler tube hazards, low efficiency of manual judgment, and complex mechanism model construction.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of big data mining application technology in thermal power industry, and particularly relates to a boiler tube fault diagnosis and early warning method and system based on boiler wall temperature monitoring. BACKGROUND

[0002] The water wall, superheater, reheater and economizer in the boiler are important heat exchange equipment in the boiler furnace and flue, and their operating state directly affects the efficiency of the conversion of fuel chemical energy to working medium energy in the boiler and the safety and stability of the boiler operation, and the four heat exchange equipment are all composed of multiple boiler tubes. However, due to the factors such as variable boiler coal quality, variable load, improper combustion adjustment, etc., the heat transfer efficiency of the boiler tube is often reduced or even the tube bursts for maintenance. With the informationization construction of the power plant, a large number of wall temperature measuring points are arranged on the boiler tube to monitor the working state of the tube. However, the wall temperature information can only provide surface or lag information, and if the wall temperature is obviously overheated, the tube may have already appeared a fault condition, missing the best maintenance time window. In addition, there are many wall temperature measuring points, and it is difficult for the production and management personnel of the power plant to quickly and acutely capture the potential hidden danger and abnormality of the boiler tube, and it is also impossible to locate the specific tube that may appear a fault. SUMMARY

[0003] In view of the problems in the prior art, the present application provides a boiler tube fault diagnosis and early warning method based on boiler wall temperature monitoring, which utilizes the monitored boiler tube row wall temperature big data, and based on historical data analysis and sorting feature criterion method, provides a method for early warning of tubes with hidden faults or trends, and provides a health monitoring, fault early warning and operation maintenance auxiliary decision guidance function of the boiler tube operating state for the production and management personnel of the power plant.

[0004] The present application is realized by the following technical solutions:

[0005] A boiler tube fault diagnosis and early warning method based on boiler wall temperature monitoring, comprising the following steps:

[0006] Step 1, acquiring the wall temperature of all boiler tubes in the heat exchange equipment at the current time t0, and selecting the top n boiler tubes {k1, k2, k3, …, kn} with the highest wall temperature; n};

[0007] Step 2, acquiring the sorting positions of the n boiler tubes {k1, k2, k3, …, kn} from high to low in the wall temperature at the corresponding historical time; n};

[0008] Step 3, determining the sorting distance feature value of the n boiler tubes according to the sorting positions of the n boiler tubes at the historical time and the current time, and determining the state of the heat exchange equipment according to the sorting distance feature value;

[0009] Step 4, according to the n boiler tubes {k1, k2, k3, …, kn} in the sorting position in multiple historical moments, determine the wall temperature early warning coefficient of each boiler tube, determine the state of each boiler tube. n Step 4, according to the n boiler tubes {k1, k2, k3, …, kn} in the sorting position in multiple historical moments, determine the wall temperature early warning coefficient of each boiler tube, determine the state of each boiler tube.

[0010] Preferably, the heat exchange equipment is a water wall, a superheater, a reheater or an economizer.

[0011] Preferably, the sorting distance characteristic value is determined according to the maximum boiler tube wall temperature sorting distance in step 3;

[0012] Preferably, the expression of the sorting distance characteristic value is as follows:

[0013]

[0014] Wherein, d i is the maximum boiler tube wall temperature sorting distance.

[0015] Preferably, the method for determining the state of the heat exchange equipment according to the sorting distance characteristic value is as follows:

[0016] Compare the sorting distance characteristic value d with the set distance alarm threshold d cr , if d < d cr , it means that the heat exchange equipment is normal at the current moment; if d > d cr , step 4 is executed.

[0017] Preferably, in step 2 and step 4, multiple historical moments are selected respectively to determine the sorting position of the n boiler tubes, and the number of selected historical moments in step 2 is less than the number of selected historical moments in step 4.

[0018] Preferably, the calculation method of the wall temperature early warning coefficient P i in step 4 is as follows:

[0019]

[0020] Wherein, m is the number of selected historical moments, X i is the maximum value of the wall temperature of the boiler tube k i (i = 1 ~ n) in the sorting order at m historical moments, s i,Tj is the sorting position sequence number of the wall temperature of the boiler tube k i at T j historical moments.

[0021] Preferably, the method for determining the state of each boiler tube in step 4 is as follows:

[0022] Compare the wall temperature early warning coefficient P i of each boiler tube with the early warning coefficient threshold P crComparison, when P i <P cr Then boiler tube k i Normal status; when P i >P cr Then boiler tube k i The status is abnormal.

[0023] A system for boiler tube fault diagnosis and early warning based on boiler wall temperature monitoring includes,

[0024] The boiler tube selection module is used to obtain the wall temperature of all boiler tubes in the heat exchanger at the current time t0, and select the n boiler tubes {k1,k2,k3,…,kn} with the highest wall temperatures. n};

[0025] The sorting module is used to retrieve n boiler tubes {k1,k2,k3,…,k} at multiple historical time points. n The wall temperature is sorted from highest to lowest at the corresponding historical moment;

[0026] The equipment status judgment module is used to determine the sorting distance feature value of n boiler tubes based on their sorting positions at historical and current times, and to determine the status of the heat exchange equipment based on the sorting distance feature value.

[0027] The boiler tube status determination module is used to determine the status of n boiler tubes {k1,k2,k3,…,k n By determining the ranking position of each boiler tube in multiple historical moments, the wall temperature warning coefficient of each boiler tube is determined, and the status of each boiler tube is determined.

[0028] Compared with the prior art, the present invention has the following beneficial technical effects:

[0029] This invention provides a boiler tube fault diagnosis and early warning method based on boiler wall temperature monitoring. Based on a large amount of wall temperature monitoring data from power plants, and combined with the influence analysis of the spatial relationship between each tube bank, the wall temperature ranking characteristics of each tube bank are obtained. Furthermore, through historical data mining and feature value extraction based on time-varying trends, a criterion for tube bank fault diagnosis is derived. This method utilizes large amounts of monitored boiler tube bank wall temperature data, based on historical data analysis and ranking feature criteria, to provide early warning for tubes with potential faults or trends. It offers an effective solution to the problems of untimely detection of boiler tube faults, low efficiency of manual judgment, and complex mechanism model construction. Attached Figure Description

[0030] Figure 1 This is a flowchart of the boiler tube fault diagnosis and early warning method based on boiler wall temperature monitoring according to the present invention. Detailed Implementation

[0031] The present invention will now be described in further detail with reference to the accompanying drawings. These descriptions are intended to explain the invention and not to limit it.

[0032] See Figure 1 A boiler tube fault diagnosis and early warning method based on boiler wall temperature monitoring includes the following steps:

[0033] Step 1: Based on the location of all boiler tubes in the heat exchange equipment, obtain the wall temperature of each boiler tube and form a set M of wall temperature measuring points, M = {a1, a2, ..., a...} x}

[0034] The heat exchange equipment is a water-cooled wall, a superheater, a reheater, or an economizer.

[0035] Where x is the total number of boiler tubes in the heat exchange equipment, and a1~a x These are the tube serial numbers corresponding to each wall temperature measuring point.

[0036] Step 2: Obtain the wall temperature of each boiler tube at the current time t0, sort the boiler tubes from highest to lowest wall temperature, and select the top n boiler tubes {k1,k2,k3,…,kn} with the highest wall temperatures. n}

[0037] Where k1 represents the tube number with the highest wall temperature, k n The tube number representing the tube with the highest wall temperature (n).

[0038] Step 3: Obtain the wall temperature of each boiler tube at three historical time points (t1, t2, t3) with the same load state as the current time, and sort them from high to low wall temperature to determine the n boiler tubes {k1, k2, k3, ..., k} obtained in Step 2. n The sorting position at the corresponding historical moment.

[0039] The three historical moments should be selected in chronological order with a certain time span, and the interval between the earliest and latest moments should be avoided to be less than one month.

[0040] The current wall temperature is sorted by the index: {1, 2, 3, ..., n};

[0041] The wall temperature sorting position of each boiler tube at historical time t1 is {s} 1,t1 ,s 2,t1 ,s 3,t1 ,…s n,t1};

[0042] The wall temperature ranking position of each boiler tube at historical time t2 is {s} 1,t2 ,s 2,t2 ,s 3,t2 ,…s n,t2};

[0043] The wall temperature sequence position of each boiler tube at historical moment t3 is {s 1,t3 , 2,t3 , 3,t3 , …s n,t3}.

[0044] For example, the wall temperature sequence of boiler tube k1 at the current moment is 1, the wall temperature sequence at t1 is s 1,t1 , the wall temperature sequence at t2 is s 1,t2 , and the wall temperature sequence at t3 is s 1,t3 .

[0045] Step 4, the wall temperature sequence change distance d i of each boiler tube k i (i=1~n) at the current moment and each historical moment is calculated, i=1~n, and the sequence distance characteristic value d of the whole n boiler tubes {k1, k2, k3, …, kn} is calculated according to the wall temperature sequence change distance of each boiler tube.

[0046] Wherein, the distance calculation method is as follows:

[0047] d i,1 = s i,t1 -i

[0048] d i,2 = s i,t2 -i

[0049] d i,3 = s i,t3 -i

[0050] d i = max{d i,1 , d i,2 , d i,3}

[0051]

[0052] Step 5, the sequence distance characteristic value d is compared with the given distance alarm threshold d cr , if d cr , it indicates that the boiler tube sequence at the current moment is normal, and the current round of diagnosis is completed; if d>d cr , further judgment is made in step 6 to determine which tubes need to be warned.

[0053] The given distance alarm threshold d cr can be flexibly adjusted according to the design characteristics and operating conditions of different boilers.

[0054] Step 6, select m historical moments {T1, T2, …, T m}, determine the wall temperature sequence change distance d i of each boiler tube ki (i = 1 ~ n) in the corresponding historical time from high to low in the order of the position of the wall temperature {s i,T1 , i,T2 s i,Tm ,…s i,T2}, i = 1 ~ n.

[0055] For example, s i is the wall temperature of boiler pipe k n at T2 time, in all boiler pipe k1~k n from high to low in the order of the wall temperature.

[0056] The selection of m historical moments requires that the earliest historical moment is selected at the time when the boiler is started and enters stable operation, and m should be greater than 5.

[0057] Step 7, according to the order position of each boiler pipe k i (i = 1 ~ n) in the corresponding historical moment, determine the characteristic value of the wall temperature of each boiler pipe k i (i = 1 ~ n) with time, and take it as the wall temperature early warning coefficient P i , the early warning coefficient calculation method is as follows:

[0058] X i = max{s i,T1 ,s i,T2 ,…s i,Tm}

[0059]

[0060] Step 8, compare the wall temperature early warning coefficient P i of each boiler pipe with the early warning coefficient threshold P cr , if P i <P cr , the boiler pipe k i is normal; if P i >P cr , trigger the early warning signal of the boiler pipe k i .

[0061] Step 9, check the wall temperature early warning coefficient calculation and judgment of pipe serial number k i (i = 1 ~ n) whether to complete, if complete, end this round of diagnosis process; if not, repeat step 6 to step 8 until the completion.

[0062] A boiler pipe fault diagnosis and early warning method based on boiler wall temperature monitoring system, comprising,

[0063] Boiler pipe selection module, for obtaining the wall temperature of all boiler pipes in heat exchange equipment at current time t0, selecting the first n boiler pipes with the highest wall temperature {k1, k2, k3, …, k n};

[0064] The sorting module is used for obtaining the sorting positions of the n boiler pipes {k1, k2, k3,..., kn} in the plurality of historical moments from high to low according to the wall temperatures of the n boiler pipes in the plurality of historical moments. n} in the corresponding historical moments from high to low.

[0065] The device state judging module is used for determining the sorting distance characteristic values of the n boiler pipes according to the sorting positions of the n boiler pipes in the historical moments and the current moment, and determining the state of the heat exchange device according to the sorting distance characteristic values.

[0066] The boiler pipe state judging module is used for determining the wall temperature early warning coefficients of the n boiler pipes {k1, k2, k3,..., kn} according to the sorting positions of the n boiler pipes in the plurality of historical moments, and determining the states of the n boiler pipes. n

[0067] The method can provide a boiler four-pipe fault diagnosis and early warning method based on boiler wall temperature monitoring, can be applied to the automatic online calculation and judgment mode of the computer and the manual trigger type judgment mode, and the diagnosis and early warning period of each round can be freely set.

[0068] The method can effectively combine the large amount of wall temperature monitoring data of the power plant, analyze the wall temperature sorting characteristics of each tube bank in combination with the influence of the spatial position relationship of each tube bank, and obtain the criteria for tube bank fault diagnosis through the characteristic value extraction of the running historical data mining and the change trend with time, so that the problems of the boiler pipe hidden danger discovery not being timely, the low artificial judgment efficiency and the complex mechanism model construction are solved, and an effective scheme is provided.

[0069] The above content only illustrates the technical idea of the present application, and cannot limit the protection scope of the present application, and any modification made according to the technical idea of the present application on the basis of the technical scheme falls within the protection scope of the present application.​

Claims

1. A method for fault diagnosis and early warning of boiler tubes based on boiler wall temperature monitoring, characterized in that, Includes the following steps: Step 1: Obtain the current time t The wall temperature of all boiler tubes in the heat exchanger is selected from the highest wall temperature. n One boiler tube { k 1, k 2, k 3, …, k n }; Step 2: Obtain data from multiple historical moments. n One boiler tube { k 1, k 2, k 3, …, k n The wall temperature is sorted from highest to lowest at the corresponding historical moment; Step 3, according to n The order of each boiler tube at historical and current times, according to n The largest boiler tube wall temperature sorting change distance among all boiler tubes determines the sorting distance feature value; The expression for the sorting distance feature value is as follows: in, The largest boiler tube wall temperature variation distance; The status of the heat exchanger is determined based on the sorting distance feature value; Step 4, according to n One boiler tube { k 1, k 2, k 3, …, k n The wall temperature warning coefficient of each boiler tube is determined by the sorting position in multiple historical moments, and the status of each boiler tube is determined. The wall temperature warning coefficient P i The calculation method is as follows: in, m The number of historical moments selected, For boiler tubes k i The maximum value of the sequence number of the wall temperature (i=1~n) at m historical time points. For boiler tubes k i The wall temperature at T j The sorting position number at a historical moment.

2. The boiler tube fault diagnosis and early warning method based on boiler wall temperature monitoring according to claim 1, characterized in that, The heat exchange equipment is a water-cooled wall, a superheater, a reheater, or an economizer.

3. The boiler tube fault diagnosis and early warning method based on boiler wall temperature monitoring according to claim 1, characterized in that, The method for determining the state of the heat exchanger based on the sorting distance feature value is as follows: sort distance feature value d and the set distance alarm threshold d cr Compare, if d < d cr This indicates that the heat exchange equipment is functioning normally at the current moment; if d > d cr Then proceed to step 4.

4. The boiler tube fault diagnosis and early warning method based on boiler wall temperature monitoring according to claim 1, characterized in that, In steps 2 and 4, multiple historical moments are selected to determine... n The sorting position of each boiler tube, and the number of historical moments selected in step 2 is less than the number of historical moments selected in step 4.

5. A boiler tube fault diagnosis and early warning method based on boiler wall temperature monitoring according to claim 1, characterized in that, The method for determining the status of each boiler tube in step 4 is as follows: The wall temperature warning coefficient for each boiler tube P i With warning coefficient threshold P cr In comparison, when P i < P cr Then the boiler tube k i Normal status; when P i > P cr Then the boiler tube k i The status is abnormal.

6. A system for implementing the boiler tube fault diagnosis and early warning method based on boiler wall temperature monitoring as described in any one of claims 1-5, characterized in that, include: The boiler tube selection module is used to obtain the current time. t The wall temperature of all boiler tubes in the heat exchanger is selected from the highest wall temperature. n One boiler tube { k 1, k 2, k 3, …, k n }; The sorting module is used to retrieve data from multiple historical moments. n One boiler tube { k 1, k 2, k 3, …, k n The wall temperature is sorted from highest to lowest at the corresponding historical moment; The device status determination module is used to determine the status of the device based on the following information: n Determine the order position of each boiler tube at historical and current times. n The sorting distance characteristic value of each boiler tube is used to determine the status of the heat exchange equipment. The boiler tube status determination module is used to determine the status of the boiler tubes based on the following information: n One boiler tube { k 1, k 2, k 3, …, k n By determining the ranking position of each boiler tube in multiple historical moments, the wall temperature warning coefficient of each boiler tube is determined, and the status of each boiler tube is determined.

Images