Annealing furnace fault prediction control system and method based on data mining

Abstract

The present invention discloses an annealing furnace fault prediction control system and method based on data mining, relating to the field of industrial automation control technology. The present invention analyzes the historical data of the annealing furnace to obtain a target time; further analyzes the historical data based on the obtained target time, calculates the scale thickness threshold and wear characteristic value, and then calculates the fault severity threshold; and predicts the time of fault occurrence by monitoring the current annealing furnace fault severity in real time and comparing and analyzing the current fault severity with the fault severity threshold. The present invention achieves annealing furnace fault prediction through data mining, solves the problem of delayed fault processing existing in traditional technologies, improves production efficiency, and reduces losses.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial automation control, and in particular to an annealing furnace fault prediction control system and method based on data mining. Background Art

[0002] With the increasing automation of industrial production, annealing furnaces play a vital role in the metal processing process. Within the annealing furnace, closed-circuit cooling water circulates, absorbing heat from the furnace and dissipating it through the cooling system. This maintains a stable temperature for equipment in direct contact with the furnace, preventing damage from prolonged high temperatures. However, if a damage point develops in the cooling water jacket, cooling water can leak directly into the furnace, raising the dew point and ultimately causing annealing furnace failure. Causes of equipment cooling water jacket failure include: First, high impurity content in the closed-circuit cooling water, which easily leads to scaling and burn-through in high-temperature environments; and second, damage to the roller bearings or inaccurate assembly, which causes additional wear between the bearings and the bearing housing water jackets during operation, leading to wear-through of the jackets. Existing technologies often address faults only after they occur. This delay not only reduces production efficiency but can also lead to product quality issues and significant losses. Therefore, a data mining-based annealing furnace fault prediction control system and method are urgently needed to address these issues. Summary of the Invention

[0003] The purpose of the present invention is to provide an annealing furnace fault prediction control system and method based on data mining to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a data mining-based annealing furnace fault prediction control method, the control method comprising:

[0005] Step S100: obtaining a three-dimensional structural diagram of the annealing furnace, establishing a coordinate system in the three-dimensional structural diagram, capturing the position coordinates of the cooling water jacket of the equipment in the annealing furnace; collecting historical data of the cooling water jacket of the equipment in the annealing furnace, and classifying and storing the historical data;

[0006] Step S200: extracting historical data corresponding to the cooling water jacket of each device, analyzing the change pattern of various parameters in the historical data of the cooling water jacket of each device, and obtaining the target time;

[0007] Step S300: Analyzing the historical data based on the target time, calculating the scale thickness threshold and wear characteristic value in the cooling water jacket of each device, and analyzing the fault degree threshold of the cooling water jacket of each device based on the calculated scale thickness threshold and wear characteristic value;

[0008] Step S400: monitor the failure degree of the cooling water jacket of each device in real time and predict the failure occurrence time of the cooling water jacket of each device.

[0009] Furthermore, step S100 includes:

[0010] Step S101: In the three-dimensional structure diagram of the annealing furnace, a coordinate system is established with one corner of the bottom base of the annealing furnace as the coordinate origin, and the coordinates of the equipment cooling water jackets are marked in the three-dimensional structure diagram to extract the position coordinates W of each equipment cooling water jacket. i (X i ,Y i ,Z i ) ; Measure the actual distance D between each two cooling water jackets ij , in the three-dimensional structure diagram, the coordinate distance between each two cooling water jackets is calculated using the following formula:

[0011] ;

[0012] Among them, d ij Represents the position coordinate W i (X i ,Y i ,Z i ) corresponding to the cooling water jacket and position coordinate W j (X j ,Y j ,Z j ) The coordinate distance of the corresponding cooling water jacket in the three-dimensional structure diagram; D ij Represents the position coordinate W i (X i ,Y i ,Z i ) corresponding to the cooling water jacket and position coordinate W j (X j ,Y j ,Z j ) the actual distance between the corresponding cooling water jackets;

[0013] Step S102: Calculate distance deviation ; The distance deviation Distance deviation threshold from the preset Compare, if the distance deviation , then the origin position, coordinate axis direction and coordinate marking process of the coordinate system are corrected until the distance deviation ;

[0014] Step S103: extracting historical data of the equipment cooling water jacket corresponding to each position coordinate respectively, classifying the historical data corresponding to the same position coordinate into one category, and storing the historical data into a database according to the divided categories.

[0015] Furthermore, step S200 includes:

[0016] Step S201: extract the historical data of the equipment cooling water jacket corresponding to each position coordinate, the historical data including the wall temperature of the equipment cooling water jacket, the equipment cooling water jacket inlet temperature, the equipment cooling water jacket outlet temperature, the scale thickness, the scale thickness in the equipment cooling water jacket, and the jacket thickness of the equipment cooling water jacket; sort the wall temperature of the cooling water jacket corresponding to each collected position coordinate according to the time series, and calculate the wall temperature change rate at two adjacent time points Among them, C i 、C i+1 Respectively represent the wall temperature of the cooling water jacket collected at the i-th and i+1-th sampling points, t i , t i+1 Respectively represent the i-th and i+1-th sampling times, V i Indicates the rate of change of wall temperature between the i-th sampling point and the i+1-th sampling point;

[0017] Step S202: Compare the wall temperature change rate with a preset wall temperature change rate threshold, extract the sampling time corresponding to the wall temperature change rate greater than the wall temperature change rate threshold, sort the sampling times in ascending order, and calculate the time interval between two adjacent sampling times. ; where t j , t j+1 Respectively represent the jth and j+1th sampling times extracted; if the time interval Greater than the time interval threshold , then mark the sampling time corresponding to the time interval, the first extracted sampling time, and the last extracted sampling time; count the number of sampling points in two adjacent marked time intervals; if the number of sampling points is less than the set error point number, remove the sampling points included in the time interval; and select the minimum sampling time among the remaining sampling points as the first start time;

[0018] Step S203: Extract the inlet temperature C of the cooling water jacket of the equipment collected from the first start time to the last sampling time. 入 , the outlet temperature of the cooling water jacket of the equipment C 出 , calculate the heat exchange efficiency of cooling water corresponding to each sampling point ; Among them, T 参Indicates the reference temperature; calculates the mean heat exchange efficiency `η of all sampling points from the first start time to the current time; sets the heat exchange efficiency threshold η 阈 =`η·w; where w is the proportionality coefficient;

[0019] Step S204: All sampling times of the sampling points are collected into a time axis in chronological order, and the time axis is divided into two time axis according to the time interval threshold. , divided into several time windows, and the number of times below the heat exchange efficiency threshold η in each time window is counted 阈 The number of sampling points is calculated to calculate the heat exchange efficiency threshold η in each window. 阈 Sampling point density ; Among them, r j Indicates that the heat exchange efficiency in the jth window is lower than the threshold η 阈 The sampling point density, K j Indicates that the heat exchange efficiency in the jth window is lower than the threshold η 阈 The number of sampling points, N j Represents the total number of all sampling points in the j-th window;

[0020] Step S205: Set the sampling point density threshold r 阈 , if the sampling point density r in the window j <r 阈 , then delete the sampling points in the corresponding window; if the sampling point density in the window is r j >r 阈 , then retain the sampling points in the corresponding window; filter out the minimum sampling time among all sampling points as the target time;

[0021] In the above steps, the "wall temperature change rate" reflects the abnormal fluctuation of the equipment's cooling status. A sudden change in the wall temperature of the equipment's cooling water jacket may indicate scale deposition or increased wear, and is an important parameter indicator for early signs of failure; the "time interval threshold" is used to filter out occasional fluctuations in the equipment's cooling water jacket temperature, focusing on continuous abnormal temperature change cycles to avoid misjudgment; the "first start time" is used to determine the starting point of the heat exchange efficiency analysis to ensure that the data interval contains significantly abnormal data; the "heat exchange efficiency" directly reflects the heat dissipation capacity of the cooling water jacket. Thickening of scale or wear of the cooling water jacket will lead to a decrease in heat exchange efficiency, which is an important parameter for assessing the degree of equipment failure. The "heat exchange efficiency threshold" is set according to the "heat exchange efficiency mean", and the proportional coefficient is dynamically set through historical data to avoid misjudgment of the fixed threshold due to equipment differences or changes in operating conditions; the "time window" segments continuous data for easy analysis of the concentration of sampling points; the "sampling point density" is used to reflect the degree of abnormal fluctuation within the time period; the "sampling point density threshold" is used to filter low-probability anomalies and focus on the trend of continuously worsening faults; the "target time" is used to determine the benchmark point for fault evolution analysis, which is used for the subsequent retrospective calculation of scale thickness and wear characteristics to ensure the timeliness of fault prediction.

[0022] Furthermore, step S300 includes:

[0023] Step S301: Using the target time as a reference, the time interval threshold is set as the backtracking time. The scale thickness covered in the cooling water jacket of each device in the time period is collected, and the collected scale thickness is sorted according to the sampling time to construct a scale thickness set D i ={d1,d2,…,d n}; where D i represents the scale thickness set collected from the cooling water jacket of the i-th device during the time period, and n represents the number of samples collected; calculate the scale thickness threshold corresponding to the cooling water jacket of each device ;in, represents the scale thickness threshold corresponding to the cooling water jacket of the i-th equipment, d j Denotes the scale thickness set D i The thickness of the jth scale in ;

[0024] Step S302: obtain a thermal image of the cooling water jacket of each device using a thermal imager during the time period, identify the high-temperature area on the cooling water jacket of each device using the thermal image, mark the high-temperature area on the three-dimensional structural diagram, and obtain the area of ​​the high-temperature area on the cooling water jacket of the device by establishing a coordinate system in the three-dimensional structural diagram; use sensing technology to collect the water jacket thickness of the high-temperature area on the cooling water jacket of each device during the time period, and calculate the wear characteristic value ; Where Pi represents the wear characteristic value of the cooling water jacket of the i-th equipment, T j represents the temperature of the jth high temperature area on the cooling water jacket of the i-th equipment, S j represents the high temperature area of ​​the jth high temperature area on the cooling water jacket of the i-th equipment, h j represents the thickness of the jth high-temperature area on the cooling water jacket of the i-th equipment, a represents the thickness influence coefficient, and n represents the total number of high-temperature areas on the cooling water jacket of the i-th equipment;

[0025] Step S303: Calculate the fault level threshold of each device cooling water jacket according to the scale thickness threshold and wear characteristic value corresponding to each device cooling water jacket. ;in, represents the fault degree threshold of the cooling water jacket of the i-th equipment, w1 represents the influence weight of scale on the fault degree of the equipment cooling water jacket, and w2 represents the influence weight of wear on the fault degree of the equipment cooling water jacket;

[0026] In the above steps, the "target time" is the key node when the cooling performance begins to deteriorate. Using the "target time" as a benchmark, the time period backtracking can focus on the scale deposition data in the incipient stage of the fault, ensuring that the scale threshold reflects the scale growth trend before the actual fault; the "scale thickness threshold" is used as a benchmark to determine whether the scale deposition has reached the upper limit. When the "scale thickness" exceeds the threshold, the heat exchange efficiency will decrease; the "wear characteristic value" is used to provide a quantitative evaluation basis for wear-related faults, making up for the lag of visual monitoring.

[0027] Furthermore, step S400 includes:

[0028] Step S401: Monitor the cooling water jacket of the equipment in real time, obtain the impurity content C of the cooling water in the equipment cooling water jacket, the wall temperature T of the cooling water jacket, and the equipment vibration frequency f, and calculate the failure rate based on the real-time monitoring data of the cooling water jacket of the equipment ; Among them, R i represents the failure rate of the cooling water jacket of the i-th device, C t represents the impurity content threshold of the cooling water in the i-th cooling water jacket, f t Represents the equipment vibration frequency threshold of the cooling water jacket of the i-th equipment;

[0029] Step S402: Obtain the fault level G of each cooling water jacket in the current annealing furnace. i , based on the failure rate obtained by real-time monitoring, calculate the time required for the current failure level of each device cooling water jacket to reach the failure level threshold. The calculation formula is as follows:

[0030] ;

[0031] Among them, T i represents the failure time of the cooling water jacket of the i-th equipment, G i Indicates the current fault level of the cooling water jacket of the i-th device;

[0032] Step S403: The fault prediction data of each device cooling water jacket are aggregated into rows to construct an annealing furnace fault prediction table; the fault prediction information includes the location coordinates of the device cooling water jacket, the current fault degree, the fault degree threshold, and the fault occurrence time, and each row of fault prediction data in the annealing furnace fault prediction table is sorted in ascending order according to the fault occurrence time; when the fault occurrence time in the annealing furnace fault prediction table is less than the preset fault time threshold, the fault prediction data of the device cooling water jacket is sent to the user display interface, and an early warning prompt is issued.

[0033] Furthermore, in order to better implement the above method, an annealing furnace fault prediction control system based on data mining is also provided, which includes: a data acquisition module, a data analysis module, a threshold calculation module, and a fault prediction module;

[0034] A data acquisition module is used to obtain historical data of the cooling water jacket of the equipment in the annealing furnace and classify and store the obtained historical data;

[0035] A data analysis module, used to analyze the patterns of the historical data and determine the target time;

[0036] A threshold calculation module is used to calculate the fault degree threshold of the cooling water jacket of each device;

[0037] The fault prediction module is used to monitor the cooling water jacket of the equipment in real time and predict the time when the fault occurs.

[0038] Furthermore, the data acquisition module includes a coordinate processing unit and a data storage unit;

[0039] a coordinate processing unit, configured to establish a coordinate system in the three-dimensional structural diagram of the annealing furnace, correct the established coordinate system, and obtain the position coordinates of the cooling water jacket of each device in the annealing furnace according to the corrected coordinate system;

[0040] The data storage unit is used to obtain historical data of the cooling water jacket of each device, and classify and store the obtained historical data of the cooling water jacket of each device according to the position coordinates of the cooling water jacket of each device.

[0041] Furthermore, the data analysis module includes a temperature analysis unit, a time processing unit, an efficiency calculation unit, and a window screening unit;

[0042] The temperature analysis unit is used to sort the obtained wall temperature of the cooling water jacket of each device according to the time series and calculate the wall temperature change rate at two adjacent time points;

[0043] a time processing unit, processing the sampling time to obtain a time interval according to the wall temperature change rate at two adjacent time points, and comparing the obtained time interval with a preset time interval threshold to obtain a first start time;

[0044] an efficiency calculation unit, which obtains all sampling points from the first start time to the last sampling time, calculates the heat exchange efficiency of the cooling water corresponding to each sampling point, and sets a heat exchange efficiency threshold;

[0045] The window screening unit collects the acquired sampling points into a time axis in chronological order, divides the time axis into several time windows according to the time interval threshold, calculates the density of sampling points below the heat exchange efficiency threshold in each time window, and screens the time windows according to the sampling point density to determine the target time.

[0046] Furthermore, the threshold calculation module includes a scale threshold calculation unit, a wear calculation unit, and a fault threshold calculation unit;

[0047] The scale threshold calculation unit uses the target time as a reference and traces back a time period as the time interval threshold, collects the thickness of scale covered in the cooling water jacket of each device during the time period, and analyzes the collected scale thickness data to obtain the scale thickness threshold of the cooling water jacket of each device;

[0048] A wear calculation unit, which uses a thermal imager to obtain a thermal image of the cooling water jacket of each device, identifies and marks the high-temperature area on the cooling water jacket of each device based on the thermal image, and calculates the wear characteristic value of the cooling water jacket of each device;

[0049] The fault threshold calculation unit calculates the fault degree threshold of each device cooling water jacket according to the scale thickness threshold and wear characteristic value corresponding to each device cooling water jacket.

[0050] Furthermore, the fault prediction module includes: a data monitoring unit, a time prediction unit, and an early warning prompt unit;

[0051] a data monitoring unit, which monitors the cooling water jacket of the equipment in real time, obtains real-time monitoring data of the cooling water jacket of the equipment, and calculates the failure rate of the cooling water jacket of the equipment based on the real-time monitoring data;

[0052] The time prediction unit obtains the fault level of the cooling water jacket of each device in the current annealing furnace and predicts the time of fault occurrence based on the fault rate obtained through real-time monitoring;

[0053] The early warning prompt unit summarizes the fault prediction data of the cooling water jacket of each device into rows and constructs an annealing furnace fault prediction table. When the fault occurrence time in the annealing furnace fault prediction table is less than a preset fault time threshold, the fault prediction data of the cooling water jacket of the device is sent to the user display interface and an early warning prompt is issued.

[0054] Compared with the prior art, the present invention has the following beneficial effects:

[0055] 1. Achieved early prediction of faults: The existing technology usually takes measures to deal with the annealing furnace only after the fault occurs, which has a lag. However, the present invention can predict the impending fault in advance by collecting and analyzing the historical data of the equipment cooling water jacket, thus solving the lag of the traditional treatment method.

[0056] 2. Realization of intelligent early warning: The present invention constructs an annealing furnace fault prediction table and sorts the data in the fault prediction table from small to large according to the fault occurrence time, so that the staff can understand the fault risk level of the cooling water jacket of each equipment; the fault prediction data with a fault occurrence time less than the threshold is sent to the user display interface for early warning, thereby realizing the intelligent early warning function.

[0057] 3. Improved production efficiency: By providing early warning of impending failures, staff can take timely measures to avoid production interruptions caused by failures, thereby improving production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 Schematic diagram of the method flow of the annealing furnace fault prediction control system and method based on data mining of the present invention;

[0059] Figure 2 The diagram is a schematic diagram of the system structure of the annealing furnace fault prediction control system and method based on data mining of the present invention. DETAILED DESCRIPTION

[0060] Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of the present invention.

[0061] Example 1: Figure 1As shown, the present invention provides a technical solution, an annealing furnace fault prediction control method based on data mining, the control method comprising:

[0062] Step S100: obtaining a three-dimensional structural diagram of the annealing furnace, establishing a coordinate system in the three-dimensional structural diagram, capturing the position coordinates of the cooling water jacket of the equipment in the annealing furnace; collecting historical data of the cooling water jacket of the equipment in the annealing furnace, and classifying and storing the historical data;

[0063] Wherein, step S100 includes:

[0064] Step S101: In the three-dimensional structure diagram of the annealing furnace, a coordinate system is established with one corner of the bottom base of the annealing furnace as the coordinate origin, and the coordinates of the equipment cooling water jackets are marked in the three-dimensional structure diagram to extract the position coordinates W of each equipment cooling water jacket. i (X i ,Y i ,Z i ) ; Measure the actual distance D between each two cooling water jackets ij , in the three-dimensional structure diagram, the coordinate distance between each two cooling water jackets is calculated using the following formula:

[0065] ;

[0066] Among them, d ij Represents the position coordinate W i (X i ,Y i ,Z i ) corresponding to the cooling water jacket and position coordinate W j (X j ,Y j ,Z j ) The coordinate distance of the corresponding cooling water jacket in the three-dimensional structure diagram; D ij Represents the position coordinate W i (X i ,Y i ,Z i ) corresponding to the cooling water jacket and position coordinate W j (X j ,Y j ,Z j ) the actual distance between the corresponding cooling water jackets;

[0067] Step S102: Calculate distance deviation ; The distance deviation Distance deviation threshold from the preset Compare, if the distance deviation , then the origin position, coordinate axis direction and coordinate marking process of the coordinate system are corrected until the distance deviation ;

[0068] Step S103: extracting historical data of the cooling water jacket of the device corresponding to each position coordinate, classifying historical data corresponding to the same position coordinate into one category, and storing the historical data in a database according to the divided categories;

[0069] Step S200: extracting historical data corresponding to the cooling water jacket of each device, analyzing the change pattern of various parameters in the historical data of the cooling water jacket of each device, and obtaining the target time;

[0070] Wherein, step S200 includes:

[0071] Step S201: extract the historical data of the equipment cooling water jacket corresponding to each position coordinate, the historical data including the wall temperature of the equipment cooling water jacket, the equipment cooling water jacket inlet temperature, the equipment cooling water jacket outlet temperature, the scale thickness, the scale thickness in the equipment cooling water jacket, and the jacket thickness of the equipment cooling water jacket; sort the wall temperature of the cooling water jacket corresponding to each collected position coordinate according to the time series, and calculate the wall temperature change rate at two adjacent time points Among them, C i 、C i+1 Respectively represent the wall temperature of the cooling water jacket collected at the i-th and i+1-th sampling points, t i , t i+1 Respectively represent the i-th and i+1-th sampling times, V i Indicates the rate of change of wall temperature between the i-th sampling point and the i+1-th sampling point;

[0072] In the embodiment of the present invention, sampling point i: ti = 10:00h, Ci = 160°C; sampling point i+1: ti+1 = 10:15h, Ci+1 = 165°C; therefore, the wall temperature change rate Vi = (165-160) ÷ 0.25 = 20°C / h;

[0073] Step S202: Compare the wall temperature change rate with a preset wall temperature change rate threshold, extract the sampling time corresponding to the wall temperature change rate greater than the wall temperature change rate threshold, sort the sampling times in ascending order, and calculate the time interval between two adjacent sampling times. ; where t j , t j+1 Respectively represent the jth and j+1th sampling times extracted; if the time interval Greater than the time interval threshold , then mark the sampling time corresponding to the time interval, the first extracted sampling time, and the last extracted sampling time; count the number of sampling points in two adjacent marked time intervals; if the number of sampling points is less than the set error point number, remove the sampling points included in the time interval; and select the minimum sampling time among the remaining sampling points as the first start time;

[0074] Step S203: extract the inlet temperature C of the cooling water jacket of the equipment collected at all sampling points from the first starting time to the last sampling time. 入 , the outlet temperature of the cooling water jacket of the equipment C 出 , calculate the heat exchange efficiency of cooling water corresponding to each sampling point ; Among them, T 参 Indicates the reference temperature; calculates the mean heat exchange efficiency `η of all sampling points from the first start time to the current time; sets the heat exchange efficiency threshold η 阈 =`η·w; where w is the proportionality coefficient;

[0075] In an embodiment of the present invention, the inlet temperature of the cooling water jacket of a certain device is: Cin=150°C, the outlet temperature is: Cout=130°C, and the reference temperature Tref=200°C; therefore, η=(150-130)÷(200-130)=0.286;

[0076] Step S204: All sampling times of the sampling points are collected into a time axis in chronological order, and the time axis is divided into two time axis according to the time interval threshold. , divided into several time windows, and the number of times below the heat exchange efficiency threshold η in each time window is counted 阈 The number of sampling points is calculated to calculate the heat exchange efficiency threshold η in each window. 阈 Sampling point density ; Among them, r j Indicates that the heat exchange efficiency in the jth window is lower than the threshold η 阈 The sampling point density, K j Indicates that the heat exchange efficiency in the jth window is lower than the threshold η 阈 The number of sampling points, N j Represents the total number of all sampling points in the j-th window;

[0077] Step S205: Set the sampling point density threshold r 阈 , if the sampling point density r in the window j <r 阈 , then delete the sampling points in the corresponding window; if the sampling point density in the window is r j >r 阈 , then retain the sampling points in the corresponding window; filter out the minimum sampling time among all sampling points as the target time;

[0078] Step S300: Analyzing the historical data based on the target time, calculating the scale thickness threshold and wear characteristic value in the cooling water jacket of each device, and analyzing the fault degree threshold of the cooling water jacket of each device based on the calculated scale thickness threshold and wear characteristic value;

[0079] Wherein, step S300 includes:

[0080] Step S301: Using the target time as a reference, the time interval threshold is set as the backtracking time. The scale thickness covered in the cooling water jacket of each device in the time period is collected, and the collected scale thickness is sorted according to the sampling time to construct a scale thickness set D i ={d1,d2,…,d n}; where D i represents the scale thickness set collected from the cooling water jacket of the i-th device during the time period, and n represents the number of samples collected; calculate the scale thickness threshold corresponding to the cooling water jacket of each device ;in, represents the scale thickness threshold corresponding to the cooling water jacket of the i-th equipment, d j Denotes the scale thickness set D i The thickness of the jth scale in ;

[0081] Step S302: obtain a thermal image of the cooling water jacket of each device using a thermal imager during the time period, identify the high-temperature area on the cooling water jacket of each device using the thermal image, mark the high-temperature area on the three-dimensional structural diagram, and obtain the area of ​​the high-temperature area on the cooling water jacket of the device by establishing a coordinate system in the three-dimensional structural diagram; use sensing technology to collect the water jacket thickness of the high-temperature area on the cooling water jacket of each device during the time period, and calculate the wear characteristic value ; Where Pi represents the wear characteristic value of the cooling water jacket of the i-th equipment, T j represents the temperature of the jth high temperature area on the cooling water jacket of the i-th equipment, S j represents the high temperature area of ​​the jth high temperature area on the cooling water jacket of the i-th equipment, h j represents the thickness of the jth high-temperature area on the cooling water jacket of the i-th equipment, a represents the thickness influence coefficient, and n represents the total number of high-temperature areas on the cooling water jacket of the i-th equipment;

[0082] Step S303: Calculate the fault level threshold of each device cooling water jacket according to the scale thickness threshold and wear characteristic value corresponding to each device cooling water jacket. ;in, represents the fault degree threshold of the cooling water jacket of the i-th equipment, w1 represents the influence weight of scale on the fault degree of the equipment cooling water jacket, and w2 represents the influence weight of wear on the fault degree of the equipment cooling water jacket;

[0083] Step S400: monitoring the failure degree of each device cooling water jacket in real time and predicting the failure occurrence time of each device cooling water jacket;

[0084] Wherein, step S400 includes:

[0085] Step S401: Monitor the cooling water jacket of the equipment in real time, obtain the impurity content C of the cooling water in the equipment cooling water jacket, the wall temperature T of the cooling water jacket, and the equipment vibration frequency f, and calculate the failure rate based on the real-time monitoring data of the cooling water jacket of the equipment ; Among them, R i represents the failure rate of the cooling water jacket of the i-th device, C t represents the impurity content threshold of the cooling water in the i-th cooling water jacket, f t represents the vibration frequency threshold of the cooling water jacket of the i-th device, β1, β2, and β3 are all weight coefficients;

[0086] Step S402: Obtain the fault level G of each cooling water jacket in the current annealing furnace. i , based on the failure rate obtained by real-time monitoring, calculate the time required for the current failure level of each device cooling water jacket to reach the failure level threshold. The calculation formula is as follows:

[0087] ;

[0088] Among them, T i represents the failure time of the cooling water jacket of the i-th equipment, G i Indicates the current fault level of the cooling water jacket of the i-th device;

[0089] Step S403: The fault prediction data of each device cooling water jacket is aggregated into rows to construct an annealing furnace fault prediction table; the fault prediction information includes the location coordinates of the device cooling water jacket, the current fault severity, the fault severity threshold, and the fault occurrence time. Each row of fault prediction data in the annealing furnace fault prediction table is sorted in ascending order of fault occurrence time; when the fault occurrence time in the annealing furnace fault prediction table is less than a preset fault time threshold, the fault prediction data of the device cooling water jacket is sent to a user display interface and an early warning prompt is issued;

[0090] In an embodiment of the present invention, historical data of the cooling water jacket A of the equipment is collected and stored, the wall temperature of the cooling water jacket A of the equipment in the collected historical data is sorted according to the time series, and the rate of change of the wall temperature at two adjacent time points is calculated; for example, at sampling point i: ti =10:00h, C i =160℃; sampling point i+1: t i+1 =10:15h,C i+1 =165℃; therefore, the wall temperature change rate V i =(165-160)÷0.25=20℃ / h; the calculated temperature change rate is greater than the wall temperature change rate threshold, so the sampling time of the sampling point is extracted, the sampling times are sorted in ascending order, the time interval between two adjacent sampling times is calculated, and the sampling points are filtered by the time interval, and the minimum sampling time among the remaining sampling points is selected as the first start time; the equipment cooling water jacket inlet temperature C collected from all sampling points of the equipment cooling water jacket A from the first start time to the last sampling time in the historical data is extracted 入 , the outlet temperature of the cooling water jacket of the equipment C 出 , calculate the heat exchange efficiency of cooling water corresponding to each sampling point; for example, C 入 =150℃, outlet temperature: C 出 =130℃, reference temperature T 参 =200℃; therefore η=(150-130)÷(200-130)=0.286; combine the heat exchange efficiency threshold and the sampling point density threshold in each time window to determine the target time; use the target time as the benchmark to trace back The scale thickness set D of the cooling water jacket A of the equipment in the historical data is collected in a time period = {0.5, 0.6, 0.4, 0.7, 0.5} in mm. Therefore, the number of samples n = 5, the scale thickness threshold d of A 阈 =(0.5+0.6+0.4+0.7+0.5)÷5=0.54mm; The thermal imager identifies the high temperature area 1 in the cooling water jacket A of the equipment: T1=120℃, S1=2cm 2 , h1=1mm, high temperature area 2: T2=100℃, S2=1.5cm2, h2=0.8mm, thickness influence coefficient: a=0.1, therefore, P i =(120×200×1+100×150×1)÷(200×1+150×1)=111.43; set w1=0.6, w2=0.4, so the fault level threshold of the equipment cooling water jacket A is =0.6×0.54+0.4×111.43=44.896; the current state of the equipment cooling water jacket A is monitored in real time, and the impurity content C=30ppm, C t =50ppm, wall temperature T=80℃, equipment vibration frequency f=15HZ, f t=20HZ, weight coefficient: β1=0.4, β2=0.3, β3=0.3, current fault level G i =10; therefore, the failure rate of equipment cooling jacket A is R=0.4×(30÷50)+0.3×80+0.3×(15÷20)=24.465; therefore, the predicted failure time of equipment cooling water jacket A is T=(44.896-10)÷24.465=1.43 (days);

[0091] Example 2: Figure 2 As shown, in order to better implement the above method, an annealing furnace fault prediction control system based on data mining is also provided, which includes a data acquisition module, a data analysis module, a threshold calculation module, and a fault prediction module;

[0092] A data acquisition module is used to obtain historical data of the cooling water jacket of the equipment in the annealing furnace and classify and store the obtained historical data;

[0093] A data analysis module, used to analyze the patterns of the historical data and determine the target time;

[0094] A threshold calculation module is used to calculate the fault degree threshold of the cooling water jacket of each device;

[0095] A fault prediction module is used to monitor the cooling water jacket of the equipment in real time and predict the time of occurrence of a fault;

[0096] Among them, the data acquisition module includes a coordinate processing unit and a data storage unit;

[0097] a coordinate processing unit, configured to establish a coordinate system in the three-dimensional structural diagram of the annealing furnace, correct the established coordinate system, and obtain the position coordinates of the cooling water jacket of each device in the annealing furnace according to the corrected coordinate system;

[0098] A data storage unit, used to obtain historical data of the cooling water jacket of each device, and classify and store the obtained historical data of the cooling water jacket of each device according to the position coordinates of the cooling water jacket of each device;

[0099] Among them, the data analysis module includes a temperature analysis unit, a time processing unit, an efficiency calculation unit, and a window screening unit;

[0100] The temperature analysis unit is used to sort the obtained wall temperature of the cooling water jacket of each device according to the time series and calculate the wall temperature change rate at two adjacent time points;

[0101] a time processing unit, processing the sampling time to obtain a time interval according to the wall temperature change rate at two adjacent time points, and comparing the obtained time interval with a preset time interval threshold to obtain a first start time;

[0102] an efficiency calculation unit, which obtains all sampling points from the first start time to the last sampling time, calculates the heat exchange efficiency of the cooling water corresponding to each sampling point, and sets a heat exchange efficiency threshold;

[0103] a window screening unit that aggregates the acquired sampling points into a time axis in chronological order, divides the time axis into a plurality of time windows according to a time interval threshold, calculates the density of sampling points below the heat exchange efficiency threshold in each time window, and screens the time windows according to the sampling point density to determine the target time;

[0104] Among them, the threshold calculation module includes a scale threshold calculation unit, a wear calculation unit, and a fault threshold calculation unit;

[0105] The scale threshold calculation unit uses the target time as a reference and traces back a time period as the time interval threshold, collects the thickness of scale covered in the cooling water jacket of each device during the time period, and analyzes the collected scale thickness data to obtain the scale thickness threshold of the cooling water jacket of each device;

[0106] A wear calculation unit, which uses a thermal imager to obtain a thermal image of the cooling water jacket of each device, identifies and marks the high-temperature area on the cooling water jacket of each device based on the thermal image, and calculates the wear characteristic value of the cooling water jacket of each device;

[0107] A fault threshold calculation unit calculates the fault degree threshold of each device cooling water jacket according to the scale thickness threshold and wear characteristic value corresponding to each device cooling water jacket;

[0108] Among them, the fault prediction module includes: data monitoring unit, time prediction unit, and early warning prompt unit;

[0109] a data monitoring unit, which monitors the cooling water jacket of the equipment in real time, obtains real-time monitoring data of the cooling water jacket of the equipment, and calculates the failure rate of the cooling water jacket of the equipment based on the real-time monitoring data;

[0110] The time prediction unit obtains the fault level of the cooling water jacket of each device in the current annealing furnace and predicts the time of fault occurrence based on the fault rate obtained through real-time monitoring;

[0111] The early warning prompt unit summarizes the fault prediction data of the cooling water jacket of each device into rows and constructs an annealing furnace fault prediction table. When the fault occurrence time in the annealing furnace fault prediction table is less than a preset fault time threshold, the fault prediction data of the cooling water jacket of the device is sent to the user display interface and an early warning prompt is issued.

[0112] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions described in the aforementioned embodiments or substitute equivalents for some of the technical features. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. Annealing furnace fault prediction and control method based on data mining, characterized by: The control method includes: Step S100: obtaining a three-dimensional structural diagram of the annealing furnace, establishing a coordinate system in the three-dimensional structural diagram, capturing the position coordinates of the cooling water jacket of the equipment in the annealing furnace; collecting historical data of the cooling water jacket of the equipment in the annealing furnace, and classifying and storing the historical data; Step S200: extracting historical data corresponding to the cooling water jacket of each device, analyzing the change pattern of various parameters in the historical data of the cooling water jacket of each device, and obtaining the target time; Step S300: Analyzing the historical data based on the target time, calculating the scale thickness threshold and wear characteristic value in the cooling water jacket of each device, and analyzing the fault degree threshold of the cooling water jacket of each device based on the calculated scale thickness threshold and wear characteristic value; Step S400: monitoring the failure degree of each device cooling water jacket in real time and predicting the failure occurrence time of each device cooling water jacket; Step S300: Step S301: Using the target time as a reference, the time interval threshold is set as the backtracking time. The scale thickness covered in the cooling water jacket of each device in the time period is collected, and the collected scale thickness is sorted according to the sampling time to construct a scale thickness set D i ={d1,d2,…,d n }; where D i represents the scale thickness set collected from the cooling water jacket of the i-th device during the time period, and n represents the number of samples collected; calculate the scale thickness threshold corresponding to the cooling water jacket of each device ;in, represents the scale thickness threshold corresponding to the cooling water jacket of the i-th equipment, d j Denotes the scale thickness set D i The thickness of the jth scale in ; Step S302: obtain a thermal image of the cooling water jacket of each device using a thermal imager during the time period, identify the high-temperature area on the cooling water jacket of each device using the thermal image, mark the high-temperature area on the three-dimensional structural diagram, and obtain the area of ​​the high-temperature area on the cooling water jacket of the device by establishing a coordinate system in the three-dimensional structural diagram; use sensing technology to collect the water jacket thickness of the high-temperature area on the cooling water jacket of each device during the time period, and calculate the wear characteristic value ; Where Pi represents the wear characteristic value of the cooling water jacket of the i-th equipment, T j represents the temperature of the jth high temperature area on the cooling water jacket of the i-th equipment, S j represents the high temperature area of ​​the jth high temperature area on the cooling water jacket of the i-th equipment, h j represents the thickness of the jth high-temperature area on the cooling water jacket of the i-th equipment, a represents the thickness influence coefficient, and n represents the total number of high-temperature areas on the cooling water jacket of the i-th equipment; Step S303: Calculate the fault level threshold of each device cooling water jacket according to the scale thickness threshold and wear characteristic value corresponding to each device cooling water jacket. ;in, represents the fault degree threshold of the cooling water jacket of the i-th equipment, w1 represents the influence weight of scale on the fault degree of the equipment cooling water jacket, and w2 represents the influence weight of wear on the fault degree of the equipment cooling water jacket; Step S400: Step S401: Monitor the cooling water jacket of the equipment in real time, obtain the impurity content C of the cooling water in the equipment cooling water jacket, the wall temperature T of the cooling water jacket, and the equipment vibration frequency f, and calculate the failure rate based on the real-time monitoring data of the cooling water jacket of the equipment ; Among them, R i represents the failure rate of the cooling water jacket of the i-th device, C t represents the impurity content threshold of the cooling water in the i-th cooling water jacket, f t Represents the equipment vibration frequency threshold of the cooling water jacket of the i-th equipment; Step S402: Obtain the fault level G of each cooling water jacket in the current annealing furnace. i , based on the failure rate obtained by real-time monitoring, calculate the time required for the current failure level of each device cooling water jacket to reach the failure level threshold. The calculation formula is as follows: ； Among them, T i represents the failure time of the cooling water jacket of the i-th equipment, G i Indicates the current fault level of the cooling water jacket of the i-th device; Step S403: The fault prediction data of each device cooling water jacket are aggregated into rows to construct an annealing furnace fault prediction table; the fault prediction information includes the location coordinates of the device cooling water jacket, the current fault degree, the fault degree threshold, and the fault occurrence time, and each row of fault prediction data in the annealing furnace fault prediction table is sorted in ascending order according to the fault occurrence time; when the fault occurrence time in the annealing furnace fault prediction table is less than the preset fault time threshold, the fault prediction data of the device cooling water jacket is sent to the user display interface, and an early warning prompt is issued.

2. The annealing furnace fault prediction and control method based on data mining according to claim 1, characterized in that: Step S100: Step S101: In the three-dimensional structure diagram of the annealing furnace, a coordinate system is established with one corner of the bottom base of the annealing furnace as the coordinate origin, and the coordinates of the equipment cooling water jackets are marked in the three-dimensional structure diagram to extract the position coordinates W of each equipment cooling water jacket. i (X i ,Y i ,Z i ) ; Measure the actual distance D between each two cooling water jackets ij , calculate the coordinate distance d between each two cooling water jackets in the three-dimensional structure diagram ij ; Among them, d ij Represents the position coordinate W i (X i ,Y i ,Z i ) corresponds to the cooling water jacket and position coordinate W j (X j ,Y j ,Z j ) The coordinate distance of the corresponding cooling water jacket in the three-dimensional structure diagram; D ij Represents the position coordinate W i (X i ,Y i ,Z i ) corresponds to the cooling water jacket and position coordinate W j (X j ,Y j ,Z j ) the actual distance between the corresponding cooling water jackets; Step S102: Calculate distance deviation ; The distance deviation Distance deviation threshold from the preset Compare, if the distance deviation , then the origin position, coordinate axis direction and coordinate marking process of the coordinate system are corrected until the distance deviation ; Step S103: extracting historical data of the equipment cooling water jacket corresponding to each position coordinate respectively, classifying the historical data corresponding to the same position coordinate into one category, and storing the historical data into a database according to the divided categories.

3. The annealing furnace fault prediction and control method based on data mining according to claim 1, characterized in that: Step S200 includes: Step S201: extract the historical data of the equipment cooling water jacket corresponding to each position coordinate, the historical data including the wall temperature of the equipment cooling water jacket, the equipment cooling water jacket inlet temperature, the equipment cooling water jacket outlet temperature, the scale thickness, the scale thickness in the equipment cooling water jacket, and the jacket thickness of the equipment cooling water jacket; sort the wall temperature of the cooling water jacket corresponding to each collected position coordinate according to the time series, and calculate the wall temperature change rate at two adjacent time points Among them, C i 、C i+1 Respectively represent the wall temperature of the cooling water jacket collected at the i-th and i+1-th sampling points, t i , t i+1 Respectively represent the i-th and i+1-th sampling times, V i Indicates the rate of change of wall temperature between the i-th sampling point and the i+1-th sampling point; Step S202: Compare the wall temperature change rate with a preset wall temperature change rate threshold, extract the sampling time corresponding to the wall temperature change rate greater than the wall temperature change rate threshold, sort the sampling times in ascending order, and calculate the time interval between two adjacent sampling times. ; where t j , t j+1 Respectively represent the jth and j+1th sampling times extracted; if the time interval Greater than the time interval threshold , then mark the sampling time corresponding to the time interval, the first extracted sampling time, and the last extracted sampling time; count the number of sampling points in two adjacent marked time intervals; if the number of sampling points is less than the set error point number, remove the sampling points included in the time interval; and select the minimum sampling time among the remaining sampling points as the first start time; Step S203: Extract the inlet temperature C of the cooling water jacket of the equipment collected from the first start time to the last sampling time. 入 , the outlet temperature of the cooling water jacket of the equipment C 出 , calculate the heat exchange efficiency of cooling water corresponding to each sampling point ; Among them, T 参 Indicates the reference temperature; calculates the mean heat exchange efficiency `η of all sampling points from the first start time to the current time; sets the heat exchange efficiency threshold η 阈 =`η·w; where w is the proportionality coefficient; Step S204: All sampling times of the sampling points are collected into a time axis in chronological order, and the time axis is divided into two time axis according to the time interval threshold. , divided into several time windows, count the number of sampling points below the heat exchange efficiency threshold ηthreshold in each time window, and calculate the density of sampling points below the heat exchange efficiency threshold ηthreshold in each window ; Where rj represents the density of sampling points below the heat exchange efficiency threshold ηthreshold in the j-th window, Kj represents the number of sampling points below the heat exchange efficiency threshold ηthreshold in the j-th window, and Nj represents the total number of all sampling points in the j-th window; Step S205: Set a sampling point density threshold rthreshold. If the sampling point density rj in the window is less than rthreshold, delete the sampling points in the corresponding window. If the sampling point density rj in the window is greater than rthreshold, retain the sampling points in the corresponding window. Filter out the minimum sampling time among all sampling points as the target time.

4. An annealing furnace fault prediction control system based on data mining, used to implement the annealing furnace fault prediction control method based on data mining according to any one of claims 1 to 3, characterized in that: The control system includes: a data acquisition module, a data analysis module, a threshold calculation module, and a fault prediction module; The data acquisition module is used to obtain historical data of the cooling water jacket of the equipment in the annealing furnace and classify and store the obtained historical data; The data analysis module is used to analyze the patterns of the historical data and determine the target time; The threshold calculation module is used to calculate the fault degree threshold of the cooling water jacket of each device; The fault prediction module is used to monitor the cooling water jacket of the equipment in real time and predict the time when a fault occurs.

5. The annealing furnace fault prediction control system based on data mining according to claim 4, characterized in that: The data acquisition module includes a coordinate processing unit and a data storage unit; The coordinate processing unit is used to establish a coordinate system in the three-dimensional structural diagram of the annealing furnace, and to correct the established coordinate system, and to obtain the position coordinates of the cooling water jacket of each device in the annealing furnace according to the corrected coordinate system; The data storage unit is used to obtain the historical data of each device cooling water jacket, and classify and store the obtained historical data of the device cooling water jacket according to the position coordinates of each device cooling water jacket.

6. The annealing furnace fault prediction control system based on data mining according to claim 4, characterized in that: The data analysis module includes a temperature analysis unit, a time processing unit, an efficiency calculation unit, and a window screening unit; The temperature analysis unit is used to sort the obtained wall surface temperatures of the cooling water jacket of each device according to a time series and calculate the wall surface temperature change rate at two adjacent time points; The time processing unit processes the sampling time according to the wall temperature change rate at two adjacent time points to obtain a time interval, compares the obtained time interval with a preset time interval threshold, and obtains a first start time; The efficiency calculation unit obtains all sampling points from the first start time to the last sampling time, calculates the heat exchange efficiency of the cooling water corresponding to each sampling point, and sets a heat exchange efficiency threshold; The window screening unit aggregates the acquired sampling points into a time axis in chronological order, divides the time axis into a plurality of time windows according to a time interval threshold, calculates the density of sampling points below the heat exchange efficiency threshold in each time window, and screens the time windows according to the sampling point density to determine the target time.

7. The annealing furnace fault prediction control system based on data mining according to claim 4, characterized in that: The threshold calculation module includes a scale threshold calculation unit, a wear calculation unit, and a fault threshold calculation unit; The scale threshold calculation unit is configured to collect the thickness of scale covered in the cooling water jacket of each device within the time period from the target time as a reference and to analyze the collected scale thickness data to obtain the scale thickness threshold of the cooling water jacket of each device; The wear calculation unit obtains a thermal image of the cooling water jacket of each device through a thermal imager, identifies and marks the high-temperature area on the cooling water jacket of each device based on the thermal image, and calculates the wear characteristic value of the cooling water jacket of each device; The fault threshold calculation unit calculates the fault degree threshold of each device cooling water jacket according to the scale thickness threshold and wear characteristic value corresponding to each device cooling water jacket.

8. The annealing furnace fault prediction control system based on data mining according to claim 4, characterized in that: The fault prediction module includes: a data monitoring unit, a time prediction unit, and an early warning prompt unit; The data monitoring unit monitors the equipment cooling water jacket in real time, obtains real-time monitoring data of the equipment cooling water jacket, and calculates the failure rate of the equipment cooling water jacket based on the real-time monitoring data; The time prediction unit obtains the fault degree of each device cooling water jacket in the current annealing furnace and predicts the time of fault occurrence based on the fault rate obtained by real-time monitoring; The early warning prompt unit summarizes the fault prediction data of the cooling water jacket of each device into rows and constructs an annealing furnace fault prediction table. When the fault occurrence time in the annealing furnace fault prediction table is less than a preset fault time threshold, the fault prediction data of the cooling water jacket of the device is sent to the user display interface and an early warning prompt is issued.

Images