Pneumatic ash conveying fault early warning method and device, electronic equipment and storage medium

Abstract

The application provides a pneumatic ash conveying fault early warning method and device, electronic equipment and storage medium. First, the historical operation data of the pneumatic ash conveying system is processed by using a Monte Carlo algorithm with a specified number of random vectors, and a first ash conveying pressure curve is generated. The first ash conveying pressure curve includes an abnormal ash conveying curve and a normal ash conveying curve. Then, after generating a second ash conveying pressure curve based on real-time operation data of the pneumatic ash conveying system, the second ash conveying pressure curve can be classified based on the abnormal ash conveying curve and the normal ash conveying curve to predict the fault point of the pneumatic ash conveying system through the classification result. Based on the application, the ash conveying condition of the pneumatic ash conveying system can be analyzed and predicted in real time without manual intervention, with high automation and strong real-time performance, thereby improving the overall performance of the pneumatic ash conveying system operation.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of flue gas treatment, more particularly to a pneumatic ash conveying fault early warning method and device, electronic equipment and storage medium. BACKGROUND

[0002] The positive pressure dense phase pneumatic ash conveying is based on the principle of gas-solid two-phase flow, and utilizes the static pressure and / or dynamic pressure of compressed air to convey materials at high concentration and high efficiency. The ash must be fully fluidized in the bin pump, and is conveyed while being fluidized. The entire pneumatic ash conveying system is composed of five parts: air source part, conveying part, pipeline part, ash storage part and control part. According to the size of the system, a single bin pump can be operated, or multiple bin pumps can be combined to form a system for operation. Each group of bin pumps conveying dry ash is one working cycle, and each cycle is divided into four stages: feeding stage, pressurized fluidization stage, conveying stage and waiting stage.

[0003] The pneumatic ash conveying system generally uses a PLC (Programmable Logic Controller) for automatic control of ash conveying. Under normal circumstances, the system operates automatically without human intervention, but in abnormal circumstances, the system will issue a failure alarm, affecting the ash conveying effect. At present, the method based on the process principle is basically adopted, and the artificial method is used to analyze and judge the ash conveying pressure history curve, and the fault point of the ash conveying system is predicted according to the shape / curvature of the curve, which provides a reference basis for the subsequent fault handling method.

[0004] However, the artificial analysis method has the problems of large workload, complete dependence on the experience accumulation of workers, easy misjudgment and omission, and large time lag and low efficiency, which cannot realize real-time early warning and judgment of the fault point. SUMMARY

[0005] Therefore, in order to solve the above problems, the present application provides a pneumatic ash conveying fault early warning method, device, electronic equipment and storage medium, and the technical scheme is as follows:

[0006] A pneumatic ash conveying fault early warning method, the method comprising:

[0007] The historical operation data of the pneumatic ash conveying system is processed by using a Monte Carlo algorithm to generate a first ash conveying pressure curve, the number of random vectors of the Monte Carlo algorithm is a specified order of magnitude, and the first ash conveying pressure curve includes an ash conveying abnormal curve and an ash conveying normal curve;

[0008] A second ash conveying pressure curve is generated according to the real-time operation data of the pneumatic ash conveying system, and the second ash conveying pressure curve is classified based on the ash conveying abnormal curve and the ash conveying normal curve, so as to predict the fault point of the pneumatic ash conveying system through the classification result.

[0009] Preferably, the processing of the historical operation data of the pneumatic ash conveying system by the Monte Carlo algorithm to generate the first ash conveying pressure curve comprises:

[0010] According to the ash conveying unit of the pneumatic ash conveying system, the historical ash conveying pressure time sequence of each ash conveying stage is classified and intercepted from the historical operation data thereof;

[0011] The historical ash conveying pressure time sequence is stored in a database, and the historical ash conveying pressure time sequence is aligned as N dimensions in the database;

[0012] The first ash conveying pressure curve is generated according to the historical ash conveying pressure time sequence of N dimensions;

[0013] The running time corresponding to the maximum ash conveying pressure is determined from the historical ash conveying pressure time sequence of N dimensions by using the bubble method, and the rectangle enveloped by the first ash conveying pressure curve is divided into a first sub-rectangle and a second sub-rectangle according to the running time from early to late with the determined running time as the dividing line;

[0014] The curve picture of the first ash conveying pressure curve is generated by using the least square method, and the curve picture is divided into a first sub-picture corresponding to the first sub-rectangle and a second sub-picture corresponding to the second sub-rectangle;

[0015] The first area of the first ash conveying pressure curve in the first sub-picture and the second area of the first ash conveying pressure curve in the second sub-picture are calculated by using the Monte Carlo algorithm with the specified number of random vectors as the number of dimensions;

[0016] The type of the first ash conveying pressure curve is determined by comparing the size of the first area and the second area.

[0017] Preferably, the specified number of dimensions includes multiple dimensions; the processing of the historical operation data of the pneumatic ash conveying system by the Monte Carlo algorithm to generate the first ash conveying pressure curve further comprises:

[0018] For each dimension in the multiple dimensions, the classification accuracy of the Monte Carlo algorithm with the number of random vectors as the dimension is calculated;

[0019] The first ash conveying pressure curve corresponding to the Monte Carlo algorithm with the highest classification accuracy is selected.

[0020] Preferably, the multiple dimensions include 1000, 10,000, and 100,000;

[0021] The first ash conveying pressure curve corresponding to the Monte Carlo algorithm with the highest classification accuracy is selected.

[0022] The first dust conveying pressure curve corresponding to the Monte Carlo algorithm with 100,000 random vectors is selected.

[0023] The device comprises:

[0024] The curve generation module is configured to process historical operation data of the pneumatic dust conveying system by using a Monte Carlo algorithm to generate a first dust conveying pressure curve, wherein the number of random vectors of the Monte Carlo algorithm is a specified order of magnitude, and the first dust conveying pressure curve comprises a dust conveying abnormal curve and a dust conveying normal curve.

[0025] The curve classification module is configured to generate a second dust conveying pressure curve according to real-time operation data of the pneumatic dust conveying system, classify the second dust conveying pressure curve based on the dust conveying abnormal curve and the dust conveying normal curve, and predict a fault point of the pneumatic dust conveying system through a classification result.

[0026] Preferably, the curve generation module is specifically configured to:

[0027] According to a dust conveying unit of the pneumatic dust conveying system, historical dust conveying pressure time series of each dust conveying stage are classified and intercepted from historical operation data; the historical dust conveying pressure time series are stored in a database and aligned as N dimensions in the database; a first dust conveying pressure curve is generated according to the N-dimensional historical dust conveying pressure time series; a running time corresponding to a maximum dust conveying pressure is determined from the N-dimensional historical dust conveying pressure time series by using a bubble method, and a rectangle enveloped by the first dust conveying pressure curve is divided into a first sub-rectangle and a second sub-rectangle according to the determined running time as a dividing line and in the order of running time from early to late; a curve picture of the first dust conveying pressure curve is generated by using a least square method, and the curve picture is divided into a first sub-picture corresponding to the first sub-rectangle and a second sub-picture corresponding to the second sub-rectangle; a first area of the first dust conveying pressure curve in the first sub-picture and a second area of the first dust conveying pressure curve in the second sub-picture are calculated by using the Monte Carlo algorithm with the number of random vectors being the specified order of magnitude; and the type of the first dust conveying pressure curve is determined by comparing the sizes of the first area and the second area.

[0028] Preferably, the specified order of magnitude comprises a plurality of orders of magnitude; and the curve generation module is further configured to:

[0029] For each order of magnitude in the plurality of orders of magnitude, a classification accuracy of the Monte Carlo algorithm with the number of random vectors being the order of magnitude is calculated; and the first dust conveying pressure curve corresponding to the Monte Carlo algorithm with the highest classification accuracy is selected.

[0030] Preferably, the multiple orders of magnitude comprise: 1000, 10000, and 100000; the curve generation module is configured to select the first dust conveying pressure curve corresponding to the Monte Carlo algorithm with the highest classification accuracy, and specifically configured to:

[0031] The first dust conveying pressure curve corresponding to the Monte Carlo algorithm with the random vector quantity of 100000 is selected.

[0032] An electronic device, comprising: at least one memory and at least one processor; the memory stores an application program, and the processor invokes the application program stored in the memory, and the application program is used to implement the pneumatic dust conveying fault early warning method.

[0033] A storage medium, which stores computer program code, and the computer program code implements the pneumatic dust conveying fault early warning method when executed.

[0034] Compared with the prior art, the present application has the following beneficial effects:

[0035] The present application provides a kind of pneumatic dust conveying fault early warning method, device, electronic equipment and storage medium, first adopt the Monte Carlo algorithm with the random vector quantity of specified order of magnitude to the historical operation data of pneumatic dust conveying system Process, generate first dust conveying pressure curve, the first dust conveying pressure curve includes dust conveying abnormal curve and dust conveying normal curve;Further, after generating second dust conveying pressure curve according to the real-time operation data of pneumatic dust conveying system, the second dust conveying pressure curve can be classified based on dust conveying abnormal curve and dust conveying normal curve, to predict the fault point of pneumatic dust conveying system by classification result.The pneumatic dust conveying system of the present application can be analyzed and predicted in real time, without manual intervention, high degree of automation, strong real-time, to improve the overall performance of pneumatic dust conveying system operation. BRIEF DESCRIPTION OF DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present application or the prior art, the drawings needed in the embodiments or prior art description will be briefly introduced below. Obviously, the drawings in the following description are only embodiments of the present application, and those skilled in the art can obtain other drawings according to the provided drawings without creative labor.

[0037] Figure 1 is an example diagram of dust conveying abnormal curve 1;

[0038] Figure 2 is an example diagram of dust conveying abnormal curve 2;

[0039] Figure 3 is an example diagram of dust conveying abnormal curve 3;

[0040] Figure 4 is an example graph of abnormal ash conveying curve 4;

[0041] Figure 5 is an example graph of normal ash conveying curve;

[0042] Figure 6 is a method flow chart of the pneumatic ash conveying fault early warning method provided by the embodiment of the present application;

[0043] Figure 7 is a part of the method flow chart of the pneumatic ash conveying fault early warning method provided by the embodiment of the present application;

[0044] Figure 8 is an example graph of the first ash conveying pressure curve provided by the embodiment of the present application;

[0045] Figure 9 is an example graph of the curve picture provided by the embodiment of the present application;

[0046] Figure 10 is a structural schematic diagram of the pneumatic ash conveying fault early warning device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0047] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all the other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0048] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0049] For the convenience of understanding the present application, taking the pneumatic ash conveying system of a certain power plant as an example, the conveying units are configured as follows: two warehouse pumps of SCR (Selective Catalytic Reduction, selective catalytic reduction method) form one conveying unit, a total of four conveying units, sharing one ash conveying pipeline (ash pipe one); two warehouse pumps of an electric field form one conveying unit, a total of four conveying units, two conveying units in the same column sharing one ash conveying pipeline (ash pipe two and ash pipe three); four warehouse pumps of a second electric field form one conveying unit, a total of two conveying units, sharing one ash conveying pipeline (ash pipe four); four warehouse pumps of a third and fourth electric field form one conveying unit, a total of four conveying units, sharing one ash conveying pipeline (ash pipe five).

[0050] The typical ash conveying pressure curves of the pneumatic ash conveying system mainly include the following five kinds:

[0051] 1) See Figure 1 , Figure 1 This is an example diagram of the ash conveying anomaly curve 1. Figure 1 As shown, the abnormal ash conveying curve 1 is mainly characterized by a significant increase in the ash conveying pressure curve, which remains at a relatively high pressure value. The main reason is that the ash conveying pipeline is blocked by foreign objects or ash accumulation.

[0052] 2) See Figure 2 , Figure 2 This is an example diagram of the ash conveying anomaly curve 2. Figure 2 As shown, the abnormal ash conveying curve 2 is mainly characterized by a rapid increase and a slow decrease in ash conveying pressure, which is mainly due to the poor fluidization effect of the silo pump in the conveying unit.

[0053] 3) See Figure 3 , Figure 3 This is an example diagram of ash conveying anomaly curve 3. (See example diagram.) Figure 3 As shown, the abnormal ash conveying curve 3 is mainly characterized by a slow rise and a rapid fall in the ash conveying pressure curve, which is mainly caused by a malfunction of the feed valve / primary air valve / tertiary air valve.

[0054] 4) See Figure 4 , Figure 4 This is an example diagram of ash conveying anomaly curve 4. Figure 4 As shown, the abnormal ash conveying curve 4 is mainly characterized by a rapid rise and fall in the ash conveying pressure curve. The main reason is that the balancing valve failure results in a small amount of ash being conveyed and a shorter ash conveying time.

[0055] 5) See Figure 5 , Figure 5 This is an example diagram of the normal ash conveying curve. (Example:) Figure 5 As shown, the normal ash conveying curve is mainly characterized by a uniform rate of increase and decrease in ash conveying pressure.

[0056] At present, the method of manually analyzing the historical ash conveying pressure curve of the pneumatic ash conveying system based on process knowledge to predict the system failure point requires rich process experience and knowledge, and is labor-intensive, real-time, and prone to misjudgment and omission.

[0057] To address the aforementioned issues, this invention provides a pneumatic ash conveying fault early warning method, which can analyze and predict the ash conveying conditions of the pneumatic ash conveying system in real time without manual intervention. It features a high degree of automation and strong real-time performance, thereby improving the overall performance of the pneumatic ash conveying system.

[0058] See Figure 6 , Figure 6A method flowchart of the pneumatic ash conveying fault early warning method provided by the embodiment of the present application, the pneumatic ash conveying fault early warning method comprises the following steps:

[0059] S10, the historical operation data of the pneumatic ash conveying system is processed by using the Monte Carlo algorithm, and a first ash conveying pressure curve is generated, the number of random vectors of the Monte Carlo algorithm is a specified order of magnitude, and the first ash conveying pressure curve comprises an ash conveying abnormal curve and an ash conveying normal curve.

[0060] In the embodiment of the present application, the number of random vectors of the Monte Carlo algorithm can be specified by the user, that is, the number of random vectors is a specified order of magnitude, which meets the accuracy in the actual application scenario. Specifically, the Monte Carlo algorithm with random vector data of a specified order of magnitude is used to calculate the area of the typical historical operation data of the pneumatic ash conveying system, and comparison and analysis are performed according to the line characteristics to generate the above-mentioned five types of ash conveying pressure curves, that is, the first ash conveying pressure curve, which comprises Figure 1 to Figure 4 the ash conveying abnormal curve shown in the figure, and Figure 5 the ash conveying normal curve shown in the figure.

[0061] In the specific implementation process, the step S10 of processing the historical operation data of the pneumatic ash conveying system by using the Monte Carlo algorithm to generate the first ash conveying pressure curve can adopt the following steps, and the method flowchart is as shown in Figure 7 .

[0062] S101, according to the ash conveying unit of the pneumatic ash conveying system, the historical ash conveying pressure time sequence of each ash conveying stage is classified and cut from the historical operation data.

[0063] In the embodiment of the present application, a batch of representative historical operation data can be selected from the production site of the pneumatic ash conveying system, and the historical operation data contains the typical operation conditions of the pneumatic ash conveying system in a period of time, such as the time period containing various ash conveying faults mentioned above.

[0064] In addition, the pneumatic ash conveying system can be divided into multiple ash conveying units, each ash conveying unit can independently run according to a certain time sequence rule, and specifically contains waiting, feeding, aeration, conveying and purging of each ash conveying stage. In this regard, the operation parameters of each ash conveying stage can be classified and cut from the historical operation data of the pneumatic ash conveying system according to different ash conveying units, and the data from the opening state of the discharge valve to the closing state of the discharge valve can be cut, because only the ash conveying pressure curve in this period of time can analyze the ash conveying fault condition. Based on this, the historical ash conveying pressure time sequence of each ash conveying stage can be obtained, that is, the sampled ash conveying pressure values of the corresponding ash conveying stage are sampled at a certain sampling frequency according to the time sequence.

[0065] S102, store the historical dust conveying pressure time sequence into a database, and align the historical dust conveying pressure time sequence into N dimensions in the database.

[0066] In the embodiment of the present application, the dust conveying pressure sampled according to time sequence for each dust conveying stage can be stored into the database according to time sequence, and then aligned into N dimensions in the database, N=Tdust conveying / t sampling, wherein Tdust conveying is the timing time of the dust conveying stage, and t sampling is the sampling time.

[0067] Specifically, there are two conditions for the end of the dust conveying stage, i.e. the pipeline pressure drops to the set conveying end pressure or the timing time of the dust conveying stage, for which the present application is uniformly aligned according to the timing time of the dust conveying stage, and the insufficient is zero-filled, and the excess is deleted, so as to ensure that the dimension of the historical dust conveying pressure time sequence is N.

[0068] S103, generate a first dust conveying pressure curve according to the N-dimensional historical dust conveying pressure time sequence.

[0069] In the embodiment of the present application, for the N-dimensional historical dust conveying pressure time sequence, the dust conveying pressure can be plotted according to time sequence, so as to generate the corresponding dust conveying pressure curve, i.e. the first dust conveying pressure curve. See Figure 8 , Figure 8 The first dust conveying pressure curve provided by the embodiment of the present application has a horizontal coordinate of running time and a vertical coordinate of dust conveying pressure.

[0070] Continuing to refer to Figure 8 For the rectangle enveloped by the first dust conveying pressure curve, the area S thereof can be calculated, S=P UP *N, wherein P UP is the upper limit value of the dust conveying pressure, which can be defined in advance according to the dust conveying process, so as to ensure that the numerical value of all the dust conveying pressures in the dust conveying stage is lower than it Figure 8 P UP is 0.25).

[0071] S104, determine the running time corresponding to the maximum dust conveying pressure from the N-dimensional historical dust conveying pressure time sequence by using the bubble method, and divide the rectangle enveloped by the first dust conveying pressure curve into a first sub-rectangle and a second sub-rectangle according to the running time as the dividing line in the order from early to late running time.

[0072] In the embodiment of the present application, for the first dust conveying pressure curve, the bubble method can be used to determine the time sequence value (i.e. the running time) corresponding to the maximum numerical value of the dust conveying pressure (i.e. the maximum dust conveying pressure) in the historical dust conveying pressure time sequence, and the rectangle enveloped by the first dust conveying pressure curve is divided into two rectangular regions, i.e. the first sub-rectangle and the second sub-rectangle, according to the determined running time as the dividing line, continuing to refer to Figure 8, the first sub-rectangle, i.e. the rectangular area on the left side of M (the maximum conveying pressure corresponding to the operation time), and the second sub-rectangle, i.e. the rectangular area on the right side of M.

[0073] Continuing to refer to Figure 8 , for the first sub-rectangle, its area S1 can be calculated, S1 = P UP *M; for the second sub-rectangle, its area S2 can be calculated, S2 = P UP *(N-M).

[0074] S105, a curve picture of the first conveying pressure curve is generated by using the least square method, and the curve picture is divided into a first sub-picture corresponding to the first sub-rectangle and a second sub-picture corresponding to the second sub-rectangle.

[0075] In the embodiment of the present application, for the first conveying pressure curve, a curve picture of a fixed size corresponding thereto can be generated by using the least square method to ensure that the total pixels of the curve picture are consistent, the curve picture is black and white, black filling mode is used inside the curve envelope, and white is used outside the curve envelope. Referring to Figure 9 , Figure 9 is an example of the curve picture provided in the embodiment of the present application. As shown in Figure 9 , the left picture (i.e. the first sub-picture) can be divided according to the area corresponding to the first sub-rectangle in the curve picture, and the right picture (i.e. the second sub-picture) can be divided according to the area corresponding to the second sub-rectangle in the curve picture.

[0076] S106, the first area of the first conveying pressure curve in the first sub-picture and the second area of the first conveying pressure curve in the second sub-picture are calculated respectively by using the Monte Carlo algorithm with a specified number of random vectors.

[0077] In the embodiment of the present application, the Monte Carlo algorithm can be used to calculate the area (i.e. the first area) of the first conveying pressure curve in the first sub-picture, i.e. the area enveloped by the first conveying pressure curve in the first sub-rectangle; similarly, the Monte Carlo algorithm can be used to calculate the area (i.e. the second area) of the first conveying pressure curve in the second sub-picture, i.e. the area enveloped by the first conveying pressure curve in the second sub-rectangle.

[0078] Specifically, for the first area, n two-dimensional random vectors R UP with uniform distribution in the interval from 0 to M, the maximum value of which is less than P i , R i = (x i , y i ), x i ∈ [0, M], y i ∈ [0, P UP, i is in the range of 1-n, and n is the number of random vectors. Further, n two-dimensional random vectors R i points are randomly punched into the first sub-rectangle, and then all pixel points are traversed to obtain the color corresponding to each pixel point, as shown in Figure 9 , to calculate the first area C1 = the number of black points / total point number * S1.

[0079] Similarly, for the second area, the computer's high-level language (such as the random number generation function of PYTHON) can be used to generate n two-dimensional random vectors R UP with uniform distribution from interval M to N, the maximum value of which is less than P i , R i = (x i , y i ), x i ∈ [M, N], y i ∈ [0, PUP], i is in the range of 1-n, and n is the specified order of magnitude. Further, n two-dimensional random vectors R i points are randomly punched into the second sub-rectangle, and then all pixel points are traversed to obtain the color corresponding to each pixel point, as shown in Figure 9 , to calculate the second area C2 = the number of black points / total point number * S2.

[0080] In S107, the type of the first conveying pressure curve is determined by comparing the size of the first area and the second area.

[0081] In the embodiment of the present application, two threshold values δ1 and δ2 can be set. Taking the conveying pressure curve shown in Figure 1 to Figure 5 as an example for illustration:

[0082] When C1-C2>δ2, it is conveying abnormal curve 1; when C2-C1>δ1, it is conveying abnormal curve 2; when δ1<C1-C2<δ2, it is conveying abnormal curve 3; when C1-C2<δ1, C1<Smin, and C2<Smin, it is conveying abnormal curve 4; when C1-C2<δ1, C1>Smin and C2>Smin, it is conveying normal curve. Wherein, Smin = 0.2 * S. In this way, the type of the first conveying pressure curve can be determined, and the classification is completed.

[0083] On this basis, by iterating the database, the classification result of the Monte Carlo algorithm with the specified order of magnitude of the random vector number is compared with the real classification label, and the classification accuracy of the Monte Carlo algorithm with the specified order of magnitude is obtained.

[0084] In actual application, by modifying the random vector number n, different orders of magnitude can be specified, and each order of magnitude can complete the classification through the above steps, so as to determine the classification accuracy of the Monte Carlo algorithm of each order of magnitude.

[0085] To this end, the specified order of magnitude includes multiple orders of magnitude, and the step S20 of "processing the historical operation data of the pneumatic ash conveying system by using the Monte Carlo algorithm to generate the first ash conveying pressure curve" can further include the following steps:

[0086] For each order of magnitude in the multiple orders of magnitude, the classification accuracy of the Monte Carlo algorithm with the random vector quantity of the order of magnitude is calculated; and the first ash conveying pressure curve corresponding to the Monte Carlo algorithm with the highest classification accuracy is selected.

[0087] In the embodiment of the application, the Monte Carlo algorithm with the highest classification accuracy can be selected, and the classification result thereof is used as the basis for classifying the real-time ash conveying pressure curve in actual production.

[0088] Through the inventor's test verification, it is found that the random vector quantity n can be modified to 1000, 10000, and 100000, and the above steps are repeated, and it is verified that the classification accuracy of the Monte Carlo algorithm with the random vector quantity of 100000 is the highest. To this end, the random vector quantity of the Monte Carlo algorithm can be specified as 100000 to classify the ash conveying pressure curve after the end of the ash conveying period in real time.

[0089] S20, generating a second ash conveying pressure curve according to the real-time operation data of the pneumatic ash conveying system, classifying the second ash conveying pressure curve based on the ash conveying abnormal curve and the ash conveying normal curve, and predicting the fault point of the pneumatic ash conveying system through the classification result.

[0090] In the embodiment of the application, for the operation data (i.e. real-time operation data) generated in real time for each ash conveying stage of the pneumatic ash conveying system, the ash conveying pressure of the corresponding ash conveying stage can be sampled at a certain sampling frequency according to the time sequence to obtain the corresponding ash conveying pressure curve, i.e. the second ash conveying pressure curve.

[0091] To this end, based on the different types of ash conveying pressure curves obtained in step S10, the second ash conveying pressure curve can be classified and identified to determine the type to which the second ash conveying pressure curve belongs. Of course, in actual production, the Monte Carlo algorithm with the random vector quantity of high accuracy can also be used to classify the ash conveying pressure curve after the end of the ash conveying period in real time. Taking the random vector quantity of 100000 as an example, the random vector quantity of the Monte Carlo algorithm can be specified as 100000, and then steps S104-S107 are repeated, i.e. the following steps are sequentially executed, which are not limited in the embodiment of the application:

[0092] The bubble method is used to determine the operation time corresponding to the maximum dust conveying pressure from the second dust conveying pressure curve, and the determined operation time is used as a demarcation line, and the rectangle enveloped by the second dust conveying pressure curve is divided into a first sub-rectangle and a second sub-rectangle in the order of operation time from early to late; the least square method is used to generate a curve picture of the second dust conveying pressure curve, and the curve picture is divided into a first sub-picture corresponding to the first sub-rectangle and a second sub-picture corresponding to the second sub-rectangle; the Monte Carlo algorithm with 100,000 random vectors is used to calculate a first area of the second dust conveying pressure curve in the first sub-picture and a second area of the second dust conveying pressure curve in the second sub-picture, and the type of the second dust conveying pressure curve is determined by comparing the first area and the second area.

[0093] The pneumatic dust conveying fault early warning method provided by the embodiment can analyze and predict the dust conveying conditions of the pneumatic dust conveying system in real time without manual intervention, and has high automation and strong real-time performance, thereby improving the overall performance of the pneumatic dust conveying system.

[0094] Based on the pneumatic dust conveying fault early warning method provided in the above embodiment, the embodiment further provides a device for executing the above pneumatic dust conveying fault early warning method, and a structure diagram of the device is as shown in Figure 10 The device comprises:

[0095] The curve generation module 10 is configured to process historical operation data of the pneumatic dust conveying system by using the Monte Carlo algorithm to generate a first dust conveying pressure curve, the number of random vectors of the Monte Carlo algorithm is a specified order of magnitude, and the first dust conveying pressure curve comprises a dust conveying abnormal curve and a dust conveying normal curve.

[0096] The curve classification module 20 is configured to generate a second dust conveying pressure curve according to real-time operation data of the pneumatic dust conveying system, and classify the second dust conveying pressure curve based on the dust conveying abnormal curve and the dust conveying normal curve, so as to predict a fault point of the pneumatic dust conveying system through a classification result.

[0097] Optionally, the curve generation module 10 is specifically configured to:

[0098] According to the pneumatic ash conveying system, the historical ash conveying pressure time sequence of each ash conveying stage is classified and intercepted from the historical operation data of the ash conveying unit; the historical ash conveying pressure time sequence is stored in the database, and the historical ash conveying pressure time sequence is aligned as N-dimensional in the database; the first ash conveying pressure curve is generated according to the N-dimensional historical ash conveying pressure time sequence; the bubble method is used to determine the running time corresponding to the maximum ash conveying pressure from the N-dimensional historical ash conveying pressure time sequence, and the first ash conveying pressure curve is divided into a first sub-rectangle and a second sub-rectangle according to the running time from early to late as the dividing line; the least square method is used to generate the curve picture of the first ash conveying pressure curve, and the curve picture is cut into the first sub-picture corresponding to the first sub-rectangle and the second sub-picture corresponding to the second sub-rectangle; the Monte Carlo algorithm with a specified number of random vectors is used to calculate the first area of the first ash conveying pressure curve in the first sub-picture and the second area in the second sub-picture; the type of the first ash conveying pressure curve is determined by comparing the size of the first area and the second area.

[0099] Optionally, the specified number of orders includes a plurality of orders; the curve generation module 10 is further configured to:

[0100] For each order in the plurality of orders, calculate the classification accuracy of the Monte Carlo algorithm with a random vector number of the order; select the first ash conveying pressure curve corresponding to the Monte Carlo algorithm with the highest classification accuracy.

[0101] Optionally, the plurality of orders includes 1000, 10000, and 100000; the curve generation module 10 for selecting the first ash conveying pressure curve corresponding to the Monte Carlo algorithm with the highest classification accuracy is specifically configured to:

[0102] Select the first ash conveying pressure curve corresponding to the Monte Carlo algorithm with a random vector number of 100000.

[0103] It should be noted that the detailed functions of each module in the embodiments of the present application can be referred to the above-mentioned disclosure part of the pneumatic ash conveying fault early warning method embodiments, which will not be repeated here.

[0104] Based on the pneumatic ash conveying fault early warning method provided in the above embodiments, the present embodiment further provides an electronic device, which comprises at least one memory and at least one processor; the memory stores an application program, and the processor calls the application program stored in the memory, and the application program is used to realize the pneumatic ash conveying fault early warning method.

[0105] Based on the pneumatic ash conveying fault early warning method provided in the above embodiments, the present embodiment further provides a storage medium, which stores computer program code, and the computer program code is executed to realize the pneumatic ash conveying fault early warning method.

[0106] The above describes in detail the pneumatic ash conveying fault early warning method, device, electronic equipment and storage medium provided by the present application. The principles and implementation manners of the present application are described by using specific examples. The above description of the embodiments is only used to help understand the method of the present application and its core idea. Meanwhile, for those skilled in the art, according to the idea of the present application, the specific implementation manners and application ranges will be changed. Therefore, the content of the specification should not be understood as a limitation of the present application.

[0107] It should be noted that each embodiment in the specification adopts a progressive manner for description, and each embodiment focuses on the difference from other embodiments. The same and similar parts between each embodiment can be referred to each other. For the device disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the related parts can be referred to the method part.

[0108] It should be further noted that, in this document, the relationship terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements, or including the elements inherent in the process, method, article or equipment, or further including the elements inherent in the process, method, article or equipment. Without more limitations, the element defined by the statement "including a" does not exclude the presence of other identical elements in the process, method, article or equipment including the element.

[0109] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to the embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for early warning of pneumatic ash conveying failures, characterized in that, The method includes: The Monte Carlo algorithm is used to process the historical operating data of the pneumatic ash conveying system to generate a first ash conveying pressure curve. The number of random vectors in the Monte Carlo algorithm is on a specified order of magnitude. The first ash conveying pressure curve includes an abnormal ash conveying curve and a normal ash conveying curve. A second ash conveying pressure curve is generated based on the real-time operating data of the pneumatic ash conveying system. The second ash conveying pressure curve is classified based on the ash conveying abnormal curve and the ash conveying normal curve, so as to predict the fault point of the pneumatic ash conveying system through the classification results. The step of processing historical operating data of the pneumatic ash conveying system using the Monte Carlo algorithm to generate the first ash conveying pressure curve includes: Based on the ash conveying units of the pneumatic ash conveying system, the historical ash conveying pressure time series of each ash conveying stage is extracted from its historical operating data. The historical ash conveying pressure time series is stored in a database, and the historical ash conveying pressure time series is aligned to N dimensions in the database. The first ash conveying pressure curve is generated based on the N-dimensional historical ash conveying pressure time series. The bubble method is used to determine the running time corresponding to the maximum ash conveying pressure from the N-dimensional historical ash conveying pressure time series. The rectangle enclosed by the first ash conveying pressure curve is divided into a first sub-rectangle and a second sub-rectangle according to the determined running time and the order of running time from early to late, using the determined running time as the dividing line. The least squares method is used to generate a curve image of the first ash conveying pressure curve, and the curve image is divided into a first sub-image corresponding to the first sub-rectangle and a second sub-image corresponding to the second sub-rectangle; The Monte Carlo algorithm, with a random vector number of the specified order of magnitude, is used to calculate the first area of ​​the first ash conveying pressure curve within the first sub-image and the second area within the second sub-image. The type of the first ash conveying pressure curve is determined by comparing the size of the first area and the second area.

2. The method according to claim 1, characterized in that, The specified order of magnitude includes multiple orders of magnitude; the step of processing historical operating data of the pneumatic ash conveying system using the Monte Carlo algorithm to generate the first ash conveying pressure curve also includes: For each of the multiple orders of magnitude, calculate the classification accuracy of the Monte Carlo algorithm with the number of random vectors on that order of magnitude; Select the first ash conveying pressure curve corresponding to the Monte Carlo algorithm with the highest classification accuracy.

3. The method according to claim 2, characterized in that, The multiple orders of magnitude include: 1000, 10000, and 100000; The selection of the first ash conveying pressure curve corresponding to the Monte Carlo algorithm with the highest classification accuracy includes: Select the first ash conveying pressure curve corresponding to the Monte Carlo algorithm with a random vector quantity of 100,000.

4. A pneumatic ash conveying fault early warning device, characterized in that, The device includes: The curve generation module is used to process the historical operating data of the pneumatic ash conveying system using the Monte Carlo algorithm to generate a first ash conveying pressure curve. The number of random vectors in the Monte Carlo algorithm is on a specified order of magnitude. The first ash conveying pressure curve includes an abnormal ash conveying curve and a normal ash conveying curve. The curve classification module is used to generate a second ash conveying pressure curve based on the real-time operating data of the pneumatic ash conveying system, and classify the second ash conveying pressure curve based on the ash conveying abnormal curve and the ash conveying normal curve, so as to predict the fault point of the pneumatic ash conveying system through the classification results. Specifically, the curve generation module is used for: According to the ash conveying unit of the pneumatic ash conveying system, historical ash conveying pressure time series for each ash conveying stage are extracted from its historical operating data. These historical ash conveying pressure time series are stored in a database, and aligned to N dimensions within the database. A first ash conveying pressure curve is generated based on the N-dimensional historical ash conveying pressure time series. The bubble sort method is used to determine the operating time corresponding to the maximum ash conveying pressure from the N-dimensional historical ash conveying pressure time series. Using the determined operating time as a boundary, the rectangle enclosed by the first ash conveying pressure curve is divided into a first sub-rectangle and a second sub-rectangle according to the order of operating time from earliest to latest. A curve image of the first ash conveying pressure curve is generated using the least squares method, and the curve image is divided into a first sub-image corresponding to the first sub-rectangle and a second sub-image corresponding to the second sub-rectangle. The Monte Carlo algorithm with a random vector quantity of the specified order of magnitude is used to calculate the first area of ​​the first ash conveying pressure curve within the first sub-image and the second area within the second sub-image. By comparing the sizes of the first and second areas, the type of the first ash conveying pressure curve is determined.

5. The apparatus according to claim 4, characterized in that, The specified order of magnitude includes multiple orders of magnitude; the curve generation module is further used for: For each of the multiple orders of magnitude, calculate the classification accuracy of the Monte Carlo algorithm with that number of random vectors; select the first ash conveying pressure curve corresponding to the Monte Carlo algorithm with the highest classification accuracy.

6. The apparatus according to claim 5, characterized in that, The multiple orders of magnitude include: 1000, 10000, and 100000; the curve generation module, used to select the first ash conveying pressure curve corresponding to the Monte Carlo algorithm with the highest classification accuracy, is specifically used for: Select the first ash conveying pressure curve corresponding to the Monte Carlo algorithm with a random vector quantity of 100,000.

7. An electronic device, characterized in that, The electronic device includes: at least one memory and at least one processor; the memory stores an application program, and the processor calls the application program stored in the memory, the application program being used to implement the pneumatic ash conveying fault early warning method according to any one of claims 1-3.

8. A storage medium, characterized in that, The storage medium stores computer program code, which, when executed, implements the pneumatic ash conveying fault early warning method according to any one of claims 1-3.

Images