Bucket wheel machine thermotechnical measuring point digital twinborn modeling and fault visual presentation method

Abstract

The invention discloses a bucket wheel machine thermotechnical measuring point digital twin modeling and fault visual presentation method, and relates to the technical field of bucket wheel machines. The technical problems that in a traditional bucket wheel machine unattended system, due to the fact that a fault early warning function depends on a simple threshold value rule, a shallow diagnosis layer, a black box and response lag are caused are solved, electrical parameters are deeply analyzed to obtain a global state coefficient, then the three-dimensional laser scanning and radial basis function interpolation technology is utilized, and the fault early warning function is obtained. Reconstructing the temperature data of the limited measuring points into a continuous and high-fidelity three-dimensional temperature field so as to generate a visual thermal image evolution sequence; and finally, time sequence analysis is carried out based on the thermal image evolution sequence, sensitivity adjustment is carried out with the assistance of a global coefficient so as to output a temperature anomaly index, a composite type three-dimensional model is constructed according to the temperature anomaly index, the digital twin technology is successfully promoted to the engineering application level, and the intelligent level of bucket wheel machine state monitoring and fault early warning is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of bucket wheel machines, in particular to a bucket wheel machine thermal measurement point digital twin modeling and fault visualization presentation method. BACKGROUND

[0002] With the rapid development of the energy industry, especially the sustained demand for coal as one of the main energy sources, the bucket wheel stacker-reclaimer, as a key equipment for large-scale bulk material handling, plays a crucial role in coal ports, power plants and other scenarios. With the continuous progress of intelligent and automated technologies, the operation efficiency, safety and intelligent level of traditional bucket wheel machines face the urgent need for transformation and upgrading. The industry is actively promoting the technological transformation and intelligent upgrading of bucket wheel machines to improve operational efficiency, reduce operating costs, enhance safety performance, and achieve more efficient and environmentally friendly energy management.

[0003] In existing unattended systems for bucket wheel machines, fault warning functions mainly rely on simple judgment rules, such as temperature overrun alarms or current overrun alarms, and lack deep analysis of data, only outputting alarm results, and cannot transparently and visually present early fault signs, evolution processes and development forecasts, resulting in delayed response and inability to achieve true predictive maintenance. SUMMARY

[0004] Therefore, it is necessary to provide a bucket wheel machine thermal measurement point digital twin modeling and fault visualization presentation method to address the problems mentioned in the background.

[0005] The purpose of the present application can be achieved by the following technical solutions: a bucket wheel machine thermal measurement point digital twin modeling and fault visualization presentation method, comprising the following steps:

[0006] Step one, obtain the electrical parameters of the bucket wheel machine and perform deep analysis accordingly to output a global state coefficient representing the overall state of the bucket wheel machine;

[0007] Step two, use a cantilever laser scanning device to perform three-dimensional laser scanning and modeling on the bucket wheel machine body to obtain a three-dimensional model of the bucket wheel machine body; divide the bucket wheel machine body into several functional areas according to functional zones, construct a temperature field within the functional area based on the temperatures of the measurement points in each functional area of the bucket wheel machine, and accordingly produce a thermal image evolution sequence;

[0008] Step three, perform time series analysis on the functional area based on the thermal image evolution sequence, and supplement it with global state coefficient sensitivity adjustment to obtain a temperature anomaly index, construct a composite three-dimensional model based on the temperature anomaly index, and visually display it.

[0009] In some embodiments, the electrical parameters are deeply analyzed to output the global state coefficient:

[0010] Step 101, extract electrical parameters, specific electrical parameters include current, power, load, and record them as I(t), P(t) and F(t) respectively, where t is the index of the collection time, t = 1, 2, 3…T, T is the total number of current collection time; according to the formula The load coefficient K(t) is calculated, where I 额 is the rated current of the bucket wheel machine, P 额 is the rated power of the bucket wheel machine; α, β, γ are weight coefficients;

[0011] Step 102, construct a two-dimensional rectangular coordinate system with time as the horizontal coordinate and load coefficient as the vertical coordinate, draw points of K(t) in the coordinate system according to its corresponding collection time t and load coefficient, and connect them in turn with smooth curves to obtain the load coefficient curve; image feature analysis is carried out according to the load coefficient curve to extract feature parameters, including running smoothness, load health degree and overload impact degree;

[0012] Step 103, the running smoothness A, load health degree B and overload impact degree C are calculated by the formula to obtain the global state coefficient S, the specific calculation formula is as follows:

[0013]

[0014] Where η is the scaling factor of the global state coefficient, which is greater than zero, and controls the severity of the overall penalty.

[0015] In some embodiments, the extraction process of running smoothness is:

[0016] The load coefficient K(t) at the collection time is calculated according to the formula The standard deviation of the load coefficient curve is calculated, where is the average of the load coefficient K(t) at each collection time, and then the running smoothness is obtained by normalizing the standard deviation, and the formula for normalization is

[0017] In some embodiments, the extraction process of load health degree is:

[0018] A healthy range is preset. Two straight lines parallel to the horizontal axis are drawn on the load factor curve graph. The load factors of these lines represent the upper and lower limits of the healthy range, respectively. The load factor curve is divided into a downward segment, a middle segment, and an upward segment based on these two lines. The downward segment is the shaded area formed by the curve and the line representing the lower limit of the healthy range; the middle segment is the shaded area formed by the curve and the lines representing the lower and upper limits of the healthy range; and the downward segment is the shaded area formed by the curve and the line representing the upper limit of the healthy range. All upward, middle, and downward segments in the curve graph are summed to obtain the upward area, middle area, and downward area, respectively, and then M[i, t]. up M mid and M down Simultaneously, the areas of the upward, middle, and downward lines are summed to obtain the total area, denoted as M. tot According to the formula and Calculate the share F of the upstream area, midstream area, and downstream area respectively. up F mid and F down Then according to the formula The load health score B is calculated, where q, p, and r are the nonlinear amplification factors corresponding to the uplink share, midlink share, and downlink share, respectively, and ζ is a constant with a value of 10. -9 This is to prevent the denominator from being zero or very small, which could lead to unstable values.

[0019] In some embodiments, the process of extracting overload impact intensity is as follows:

[0020] Overload impact intensity is used to quantify the instantaneous impact and cumulative fatigue damage experienced by a bucket wheel excavator under overload conditions, i.e., K(t) > 1.1. The calculation formula is as follows:

[0021]

[0022] Where λ is the scaling factor for the overload impact, m is the amplitude amplification exponent, Δt is the sampling time interval, and M tot The total area is the sum of the areas of the upper row, the middle row, and the lower row, and ζ is a constant with a value of 10. -9 .

[0023] In some embodiments, a temperature field is constructed within the functional region:

[0024] Step 201: Each functional area has several measuring points, and a temperature sensor is installed at each measuring point to monitor the temperature at each point; the temperature corresponding to each measuring point in each functional area is recorded as T. i , where i = 1, 2, 3...N, i represents the index of any measurement point within the functional area, and N is the total number of measurement points within the functional area;

[0025] Step 202, separate the geometric model of the functional area from the overall three-dimensional model of the bucket wheel machine, generate a regular three-dimensional grid array inside the functional area, build a temperature field for extending the temperature of the discrete measuring points to the entire functional area, and calculate the weight of the grid point P affected by the measuring point i using a radial basis function distance weighting method, wherein the grid point P represents other grid points in the functional area except the measuring points;

[0026] Step 204, the base grid point P will be the basic unit for temperature field reconstruction and carry the calculated temperature value; the temperature value D(P) of each grid point P in the functional area is obtained by weighted average of the temperatures of all measuring points, and the calculation formula is:

[0027]

[0028] Step 203, convert the temperature values of the grid points in the functional area into intuitive visual display and make a thermal image evolution sequence.

[0029] In some embodiments, the weight calculation process of the grid point P affected by the measuring point i is:

[0030] The weight w i The calculation formula of D(P) is:

[0031]

[0032] Where Y i is the three-dimensional coordinates of the measuring point i, φ is the basis function, d is the distance between the point P and the measuring point, i.e. d = ||P-Y i ||.

[0033] In some embodiments, the process of making a thermal image evolution sequence is:

[0034] A preset temperature-color mapping table is used to convert the temperature of the grid point P and the temperature of the measuring point into a color value, the mapping color value is filled into the functional area according to the three-dimensional coordinates of the P point and the measuring point i, respectively, to generate a thermal map with smooth temperature gradient, thereby realizing the reconstruction of the discrete measuring point temperature into a continuous and intuitive three-dimensional temperature field, sorting the thermal maps of the bucket wheel machine at each collection time according to their corresponding collection time, and generating a thermal image evolution sequence of the bucket wheel machine in the time period T.

[0035] In some embodiments, the process of generating a composite three-dimensional model is:

[0036] Step 301, take any functional area, take any grid point Q in it, where P, i∈Q, that is, Q represents any grid point in the functional area, including the measuring point i and other grid points P; preset a pre-warning temperature corresponding to each functional area in the bucket wheel machine, extract the temperature corresponding to the grid point Q at the collection time t, and divide it by the pre-warning temperature corresponding to the functional area to obtain the contribution value of the abnormal temperature of the grid point Q, if the contribution value > 1, then the contribution value of the grid point is marked as an effective contribution value, and the sum of all effective contribution values in the functional area is calculated to obtain the total abnormal temperature contribution value of the functional area at the collection time t, denoted as Where n = 1, 2, 3, …, n is a positive integer, representing the index of any functional area in the bucket wheel machine;

[0037] Step 302, construct a two-dimensional rectangular coordinate system with time as the horizontal coordinate and the total abnormal temperature contribution value as the vertical coordinate, and input the temperature abnormal contribution total value at each collection time into the coordinate system to form several contribution points, connect the contribution points in turn with a smooth curve to obtain an abnormal temperature contribution total value curve, and perform time series analysis on the abnormal temperature contribution total value curve, and adjust the sensitivity with the aid of the global state coefficient to obtain a temperature abnormality index;

[0038] Step 303, preset an abnormal interval, when the temperature abnormality index is greater than the upper limit of the abnormal interval, the corresponding position of the functional area in the composite three-dimensional model is marked red, and a temperature warning is issued; when the temperature abnormality index is within the abnormal interval, the corresponding position of the functional area in the composite three-dimensional model is marked yellow; when the temperature abnormality index is less than the lower limit of the abnormal interval, the corresponding position of the functional area in the composite three-dimensional model is marked green; the thermal image evolution sequence of the bucket wheel machine and the composite three-dimensional model after coloring are visualized and displayed, and the abnormal temperature contribution total value curve, the maximum abnormal temperature contribution total value, the average abnormal contribution rate, the contribution slope and the temperature abnormality index of each functional area are stored as attribute data of each functional area in the corresponding position of the composite three-dimensional model.

[0039] In some embodiments, the temperature abnormality index generation process is as follows:

[0040] In the abnormal temperature contribution total value curve, the maximum abnormal temperature contribution total value is extracted, the area enclosed by the abnormal temperature contribution total value curve and the horizontal axis is calculated using calculus to obtain a contribution area, and then the contribution area is divided by the total time of the abnormal temperature contribution total value curve to obtain an average abnormal contribution rate; linear fitting is performed on all the contribution points in the abnormal temperature contribution total value curve to obtain a fitting straight line, and the contribution slope of the fitting straight line is calculated;

[0041] The maximum abnormal temperature contribution total value Average abnormal contribution rate H n , contribution slope k n and global state coefficient S are calculated and analyzed to output the temperature anomaly index of the functional area, so that the temperature anomaly index of the bucket wheel machine in each functional area can be obtained; the specific calculation formula is:

[0042]

[0043] Among them is the peak contribution reference value of the functional area, which represents the upper limit of the acceptable peak contribution in the application scenario; is the average contribution rate reference value of the functional area, which represents the upper limit of the allowable average contribution rate in the application scenario; is the slope change reference value of the functional area, which represents the upper limit of the allowable slope in the application scenario.

[0044] Compared with the prior art, the beneficial effects of the present application are:

[0045] 1、The present application extracts three indexes of running smoothness, load health degree and overload impact degree by deeply analyzing electrical parameters, and calculates and fuses them to generate a global state coefficient, which provides a basis for subsequent steps, so that the overall state background of the bucket wheel machine in which the local anomaly occurs can be better understood.

[0046] 2、The present application uses three-dimensional laser scanning and radial basis function interpolation algorithm to reconstruct the limited and discrete distributed temperature data of the measuring points into continuous and intuitive three-dimensional temperature field, and generates thermal image evolution sequence; the temperature distribution of each functional area of the bucket wheel machine is clearly presented, and the comprehensiveness and accuracy of temperature state monitoring are improved; at the same time, the generated thermal image evolution sequence provides a dynamic perspective for fault tracing for operation and maintenance personnel, and clearly shows the whole process of generation, development and diffusion of temperature anomaly.

[0047] 3、The present application extracts the maximum abnormal temperature contribution total value, average abnormal contribution rate and contribution slope by analyzing the change trend and change degree of temperature abnormal contribution, and adjusts the sensitivity through the global state coefficient to calculate and analyze the temperature anomaly index, and finally generates a composite three-dimensional model; the multi-dimensional characteristics of the local overheating of the bucket wheel machine are accurately evaluated, and the local and global correlation is realized through the global coefficient, that is, when the overall state of the equipment is not good, a higher level of early warning is issued for the local anomaly, which is more in line with the physical law of fault chain occurrence; the intelligent, dynamic and visual precise early warning of the local fault risk of the bucket wheel machine is realized. BRIEF DESCRIPTION OF DRAWINGS

[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, the accompanying drawings in the following description only constitute some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without any creative effort.

[0049] Figure 1 A schematic diagram of the present application;

[0050] Figure 2 A segmentation diagram of the load coefficient curve of the present application. DETAILED DESCRIPTION

[0051] In order to make the above-mentioned objects, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings. In the following description, a large number of specific details are set forth in order to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar improvements without departing from the spirit of the present application, so the present application is not limited to the specific embodiments disclosed below.

[0052] In the existing bucket wheel machine unattended system, the fault early warning function mainly depends on simple judgment rules (such as temperature overrun alarm or current overrun alarm), which is too simple and lacks deep analysis of data, can only output alarm results, and cannot transparently and visually present early fault signs, evolution process and development prediction, resulting in response lag and inability to realize real predictive maintenance; in order to solve this technical problem, the present application adopts the following way, it should be noted that the implementation process of the present application is in the operation process of the bucket wheel machine;

[0053] As shown in Figure 1 , the bucket wheel machine thermal measurement point digital twin modeling and fault visualization presentation method comprises the following steps:

[0054] Step one, obtain the electrical parameters of the bucket wheel machine, the specific electrical parameters including current, power, load (in percentage form); according to the electrical parameters, the running state of the bucket wheel machine is comprehensively quantified by deep analysis, and the global state coefficient is output; specifically:

[0055] Extract electrical parameters, including current, power, load, and denote them as I(t), P(t) and F(t) respectively, where t is the index of the collection time, t = 1, 2, 3 … T, T is the total number of the current collection time; according to the formula The load coefficient K(t) is calculated, where I 额 is the rated current of the bucket wheel machine, P 额is the rated power of the bucket wheel machine; a, b, g are weight coefficients, the specific size of which can be determined by subjective assignment method, and a+b+g=1 is satisfied, for example, the current I(t) is the most real-time reflection of the torque change, and the weight is set to 0.5; the power P(t) is a comprehensive performance index, and the weight is 0.3; the load F(t) is usually related to the current, and can be verified, and the weight can be lower, and the value is 0.2; when K(t)=1, it indicates that the bucket wheel machine is running under the rated condition at the current time, when K(t)>1, it indicates that the bucket wheel machine is running under overload at the current time, and when K(t)<1, it indicates that the bucket wheel machine is running under light load.

[0056] A two-dimensional rectangular coordinate system is constructed with time as the horizontal coordinate and load coefficient as the vertical coordinate, K(t) is plotted according to its corresponding collection time t and load coefficient in the coordinate system, and a smooth curve is used to connect in turn to obtain a load coefficient curve; image feature analysis is performed according to the load coefficient curve to extract feature parameters, and formula calculation analysis is performed according to the feature parameters to output a global state coefficient S, wherein the feature parameters include running smoothness, load health degree and overload impact degree; specifically:

[0057] (1) Running smoothness A: according to the formula the standard deviation of the load coefficient curve is calculated, wherein is the mean value of the load coefficient K(t) at each collection time, and then the running smoothness is obtained by normalizing it, and the formula for normalizing is It should be noted that the value of the running smoothness A is closer to 1, indicating that the load coefficient curve is smoother, the bucket wheel machine runs more smoothly, and the impact and fatigue are smaller; the smaller the value, the more intense the load fluctuation, and the bucket wheel machine may be troubled by problems such as jamming, impact load or unstable control.

[0058] (2) Load health degree B: based on safety and efficiency, a health interval is preset, and the health interval is set to [0.8, 1.1] by those skilled in the art, within this interval, the bucket wheel machine runs close to its rated capacity, which means that the energy, loss (depreciation) and labor cost invested obtains the highest output return; when the lower limit 0.8 of the health interval is exceeded, the capacity of the bucket wheel machine is not fully utilized, although the mechanical wear and failure risk is low, but the running efficiency is low; when the upper limit 1.1 of the health interval is exceeded, it indicates that the bucket wheel machine is running under load exceeding its design capacity, which leads to overheating of motor winding, overload wear of gear and bearing, increase of structural stress, etc., which will significantly accelerate the aging of equipment, shorten its service life, and greatly increase the probability of sudden failure; for example Figure 2As shown, two straight lines parallel to the horizontal axis are drawn on the load factor curve graph. The load factors of the lines represent the upper and lower limits of the healthy range, respectively. The load factor curve is divided into a downward part, a middle part, and an upward part based on the two straight lines. The downward part is the shaded area formed by the curve and the line representing the lower limit of the healthy range; the middle part is the shaded area formed by the curve and the lines representing the lower and upper limits of the healthy range; and the downward part is the shaded area formed by the curve and the line representing the upper limit of the healthy range. All upward, middle, and downward parts in the curve graph are summed to obtain the upward area, the middle area, and the downward area, respectively, and M is calculated. up M mid and M down Simultaneously, the areas of the upward, middle, and downward lines are summed to obtain the total area, denoted as M. tot According to the formula

[0059] and Calculate the share F of the upstream area, midstream area, and downstream area respectively. up F mid and F down Then according to the formula The load health score B is calculated, where q, p, and r are the nonlinear amplification factors corresponding to the uplink share, midlink share, and downlink share, respectively, to control the amplification effect of each share. q is set to 1.4; when q > 1, the impact of the uplink share on the denominator is amplified by the nonlinear factor q, resulting in a stronger penalty effect. r is set to 1.0; when p > 1, it indicates a larger F... mid This will be further amplified, causing the load health B to approach 1 more quickly; its value is 1.1. ζ is a constant with a value of 10. -9 To prevent the denominator from being zero or very small, which would lead to numerical instability; as can be seen from the calculation formula of load health B, the closer the load health is to 1, the more it indicates that the load distribution is mainly in the middle range (i.e., the healthy range), and the better the balance between capacity, energy consumption and wear and tear is achieved; the closer the load health is to 0, the more it indicates that the proportion is dominated by upward or downward, especially upward, which will significantly reduce the healthy load health B.

[0060] (3) Overload impact degree C: The overload impact degree is used to quantify the instantaneous impact and cumulative fatigue damage degree of the bucket wheel excavator under overload conditions, i.e., K(t) > 1.1. The calculation formula is as follows:

[0061]

[0062] wherein λ is a scaling factor of overload impact degree, taking a constant greater than zero, used to control the rate of change of the entire exponential function, ensuring that the overload impact degree C value can be reasonably distributed in the range of 0 to 1; m is the amplitude amplification index, taking 2, ensuring that the high amplitude overload will have an impact on the result, effectively capturing the impact load; Δt is the sampling time interval, (K(t)-1.1) represents the amplitude of overload, ζ is a constant, taking 10 -9 ; the formula calculates the relative proportion of the weighted overload area to the total running area, and maps it to 0-1 through the exponential function, and the weighted overload area is obtained by the product of the high power of the overload amplitude (K(t)-1.1) m and time, emphasizing the punishment of high amplitude overload; when there is no overload (i.e. all K(t)≤1.1, the sum term is 0, and the overload impact degree C=0, indicating no overload impact; when there is overload, the sum term The overload impact degree C value starts to increase; when the overload is very serious (large amplitude, long time), the exponential term tends to 0, and the overload impact degree C value tends to 1, indicating that it has suffered extremely serious overload impact; through the double nonlinear mapping of the exponential term m and the exponential function exp(...), m amplifies the contribution of high amplitude overload, while the exponential function ensures that the overload impact degree C value is very sensitive to the change of cumulative overload, and grows faster at the beginning of the cumulative damage, and gradually tends to 1 (saturation) as the cumulative damage intensifies, which conforms to the physical law of device damage accumulation.

[0063] The running smoothness, load health degree and overload impact degree are calculated by the formula to obtain the global state coefficient S, and the specific calculation formula is as follows:

[0064]

[0065] wherein η is a scaling factor of the global state coefficient, taking a value greater than zero, which controls the severity of the overall punishment, and the greater η, the more sensitive S is to state changes; the natural exponential function exp(...) is used to perfectly map the global state coefficient S to 0-1, accurately outputting the global state coefficient; the formula fuses the key indicators of three different dimensions of the state of the bucket wheel machine, and responds more and more intensively to the degradation process of the bucket wheel machine through the square term and the exponential function, which conforms to the development law of device failure;

[0066] By deeply analyzing the electrical parameters to extract the running smoothness, load health degree and overload impact degree, and finally calculating and fusing the global state coefficient, it can provide an important basis for the subsequent steps, so that the overall state background of the bucket wheel machine where the local anomaly occurs can be understood.

[0067] Step two, use cantilever laser scanning equipment to scan the body of the bucket wheel machine to obtain high-precision point cloud data, and model the body of the bucket wheel machine according to the high-precision point cloud data to obtain the three-dimensional modeling of the body of the bucket wheel machine; the body of the bucket wheel machine is divided into several functional areas according to the functional areas, and the specific functional areas include: bucket wheel driving motor area, bucket wheel reduction box area, suspension belt driving motor area, luffing hydraulic area, slewing bearing area, cable joint area and the like; the temperature field in the functional area is constructed according to the corresponding temperature of the measuring point in each functional area of the bucket wheel machine, and the thermal image evolution sequence is made accordingly, which is specifically:

[0068] Step 201, a plurality of measuring points are arranged in each part functional area, and a person skilled in the art sets a plurality of measuring points for each functional area according to the importance of the corresponding function of each functional area; it should be noted that the importance of the function responsible by different functional areas is different, for example, the bucket wheel reduction box area is a key equipment of the bucket wheel machine, and when there is a serious problem in this functional area, it will directly lead to equipment downtime, cause significant asset loss or cause safety accidents; a temperature sensor is installed at each measuring point to monitor the temperature of each measuring point; temperature is one of the most direct and important parameters for representing the running state of mechanical equipment, and most mechanical and electrical faults are accompanied by abnormal temperature rise, especially for large, expensive and continuous running equipment such as the bucket wheel machine, temperature is a very important monitoring parameter; the temperature corresponding to each measuring point of each functional area is denoted as T i , wherein i = 1, 2, 3, …, N, i represents the index of any one measuring point in the functional area, and N is the total number of measuring points in the functional area;

[0069] Step 202, separate the geometric model of the functional area (such as the reduction box body area) from the whole three-dimensional model of the bucket wheel machine, the surface of the model is usually composed of a large number of small triangular patches; a regular three-dimensional grid array is generated inside the functional area, in order to extend the temperature of the discrete measuring point to the whole functional area to construct the temperature field, the influence domain of each measuring point needs to be defined, that is, the influence range and degree of the temperature value of the measuring point on the surrounding area; the radial basis function distance weighting method is used to achieve this goal, for the grid point P other than the measuring point in the area, the weight w i of the influence of the measuring point i on the grid point P is calculated as follows:

[0070]

[0071] , wherein Y i is the three-dimensional coordinates of the measuring point i, φ is the basis function, d is the distance between the point P and the measuring point, that is, d = ||P-Y i ||; the relative influence degree of each measuring point on the grid point P is calculated, the sum of the weights of all measuring points on the same grid point is always 1, which ensures the rationality of the temperature calculation;

[0072] Step 204, based on the weight w of the above grid P point i (P) calculation, the grid point P will be as the basic unit of temperature field reconstruction, carrying the calculated temperature value; the temperature value D(P) of each grid point P in the functional area can be obtained by weighted average of all measured point temperatures:

[0073]

[0074] As can be known from the calculation process of each grid point D(P), in the area near the measuring point, the temperature value is close to the actual measured value of the measuring point, in the area between two measuring points, the temperature presents smooth transition, and in the area far away from all measuring points, the temperature value is the average value of all measuring point temperatures;

[0075] Step 203, converting the temperature value of each grid point in the functional area into intuitive visual display: first color mapping, using a preset temperature-color mapping table to convert the temperature D(P) of the grid point P and the temperature T i of the measuring point i into a color value, usually using a gradient color band from blue to red, wherein blue represents low temperature, green represents normal temperature, yellow represents pre-warning temperature, orange represents alarm temperature, and red represents dangerous high temperature; filling the color value obtained by mapping into the functional area according to the three-dimensional coordinates of P point and measuring point i, to generate a thermal map with smooth temperature gradient, thereby realizing the reconstruction of discrete measuring point temperature into continuous and intuitive three-dimensional temperature field, making the temperature distribution and hot spot position clear at a glance, and improving the intuitiveness and accuracy of the state monitoring and fault diagnosis of the bucket wheel machine; sorting the thermal maps of the bucket wheel machine at each collection time according to their corresponding collection time, to generate a thermal image evolution sequence of the bucket wheel machine in the time period T, which clearly shows the process of dynamic change, development and evolution of the temperature field of each functional area of the bucket wheel machine over time;

[0076] By using three-dimensional laser scanning and radial basis function interpolation algorithm, the limited and discrete distributed measuring point temperature data is reconstructed into continuous and intuitive three-dimensional temperature field, and a thermal image evolution sequence is generated; the temperature distribution of each functional area of the bucket wheel machine can be clearly presented, and the comprehensiveness and accuracy of temperature state monitoring are improved; at the same time, the generated thermal image evolution sequence provides a dynamic perspective for fault tracing for the operation and maintenance personnel, and can clearly show the whole process of generation, development and diffusion of temperature anomaly.

[0077] Step three, temperature change trend analysis based on the thermal image evolution sequence, and auxiliary sensitivity adjustment of the global state coefficient of the bucket wheel machine to generate a temperature anomaly index representing the temperature risk of each functional area, and generate a composite three-dimensional model according to the temperature anomaly index of each functional area of the bucket wheel machine; specifically:

[0078] Step 301, take any functional area, take any grid point Q in it, where P, i∈Q, that is, Q represents any grid point in the functional area, including the measuring point i and other grid points P; preset a pre-warning temperature corresponding to each functional area in the bucket wheel machine (that is, the maximum allowable working temperature of each functional area), and it needs to be noted that the pre-warning temperature is set by the person skilled in the art on the basis of the design limit of the equipment or component corresponding to each functional area (for example, motor insulation grade, bearing lubrication limit, sealing material temperature resistance, etc.) and the physical failure mode of each functional area (the sensitivity of different functional areas to temperature and the damage mechanism of the parts are different); the temperature corresponding to the grid point Q at the collection time t is extracted, and the temperature is divided by the pre-warning temperature corresponding to the functional area to obtain the contribution value of the abnormal temperature of the grid point, if the contribution value > 1, then the contribution value of the grid point is marked as an effective contribution value, and the sum of all effective contribution values in the functional area is calculated to obtain the abnormal temperature contribution total value of the functional area at the collection time t, denoted as Where n = 1, 2, 3, …, n is a positive integer, representing the index of any functional area in the bucket wheel machine; it needs to be noted that, as can be seen from the calculation process of the abnormal temperature contribution total value, is a dimensionless constant, the greater the value, the more serious the temperature abnormality of the functional area at this collection time;

[0079] Step 302, take time as the horizontal coordinate, and take the abnormal temperature contribution total value as the vertical coordinate to construct a two-dimensional rectangular coordinate system, and input the temperature abnormality contribution total value at each collection time into the coordinate system to form a plurality of contribution points, and connect the contribution points in turn by using a smooth curve to obtain an abnormal temperature contribution total value curve; extract the maximum abnormal temperature contribution total value from the abnormal temperature contribution total value curve, denoted as The maximum abnormal temperature contribution total value represents the most serious instantaneous overheating intensity that the functional area has suffered during this period, and a very high peak usually means that the functional area has suffered a great heat stress for a short time, which may be a precursor of a serious failure; the area surrounded by the abnormal temperature contribution total value curve and the horizontal axis (that is, the time axis) is calculated by using calculus to obtain the contribution area, and then the contribution area is divided by the total time (T) of the abnormal temperature contribution total value curve to obtain the average abnormal contribution rate, denoted as H n The average abnormal contribution rate represents the abnormal contribution degree per unit time on average, which quantifies the persistence and cumulative effect of heat stress; the greater the abnormal contribution degree, the longer the device is in a long-term and stable overheating state, even if the peak is not high, it will also cause material aging and lubrication failure due to the continuous action; linear fitting is performed on all the contribution points in the abnormal temperature contribution total value curve to obtain a fitting straight line, and the contribution slope of the fitting straight line is calculated, denoted as k n The contribution slope k nindicates the trend and speed of the abnormal contribution; k n > 0, indicating that the overheating situation is accelerating deterioration, k n = 0, indicating that the overheating situation is in a stable state, k n < 0, indicating that the overheating situation is mitigating; the maximum abnormal temperature contribution total value average abnormal contribution rate H n , contribution slope k n and global state coefficient S are calculated and analyzed to output the temperature abnormality index of the functional area, so that the temperature abnormality index of the bucket wheel machine in each functional area can be obtained, and the composite three-dimensional model is generated by labeling each functional area in the three-dimensional model of the bucket wheel machine; the specific calculation formula is:

[0080]

[0081] wherein is the peak contribution reference value of the functional area, indicating the upper limit of the acceptable peak contribution in the application scenario; is the average contribution rate reference value of the functional area, indicating the upper limit of the allowable average contribution rate in the application scenario; is the slope change reference value of the functional area, indicating the upper limit of the allowable slope in the application scenario; this formula simultaneously considers the abnormal peak value, average level and change trend to avoid misjudgment of a single indicator; the multiplication structure is adopted, the peak intensity, persistence and development rate are regarded as mutual amplification evidence, the smaller the global state coefficient S is, the worse the overall state of the bucket wheel machine is, and in this background, the same local abnormality is amplified (more likely to cause a chain failure), this formula integrates the global state coefficient of the bucket wheel machine, making a leap from isolated local judgment to system global linkage judgment, and realizing dynamic perception of the local temperature abnormality risk;

[0082] Step 303, preset an abnormal interval, which is taken as [1.2, 1.8] in the art; when the temperature abnormality index is greater than the upper limit of the abnormal interval, it indicates that the temperature of the functional area is seriously abnormal, then the corresponding position of the functional area in the composite three-dimensional model is marked red, and a temperature alarm is issued; when the temperature abnormality index is within the abnormal interval, the corresponding position of the functional area in the composite three-dimensional model is marked yellow; when the temperature abnormality index is less than the lower limit of the abnormal interval, the corresponding position of the functional area in the composite three-dimensional model is marked green; the thermal image evolution sequence of the bucket wheel machine and the composite three-dimensional model after coloring processing are visually displayed, and the abnormal temperature contribution total value curve, the maximum abnormal temperature contribution total value average abnormal contribution rate H n , contribution slope k n and temperature abnormality index V nThe attribute data of each functional area is stored in the corresponding position of each functional area of the composite three-dimensional model; through the visualization of the state of the bucket wheel machine and the deep fusion of data, the thermal image evolution sequence and the composite three-dimensional model based on the color mapping of the temperature anomaly index are synchronously visualized and displayed, and an intuitive spatial distribution and time-space evolution process of the health state of the bucket wheel machine are provided.

[0083] The maximum abnormal temperature contribution total value, the average abnormal contribution rate and the contribution slope are extracted by analyzing the change trend and the change degree of the temperature abnormality contribution, a global state coefficient is introduced for sensitivity adjustment, the temperature anomaly index is calculated and analyzed, and finally the composite three-dimensional model is generated; the multi-dimensional characteristics of the local overheating of the bucket wheel machine are accurately evaluated, the global coefficient is introduced, the local and the whole are associated, that is, when the overall state of the equipment is not good, a higher level of early warning is issued for the local abnormality, which is more in line with the physical law of fault chain occurrence; and intelligent, dynamic and visual accurate early warning of the local fault risk of the bucket wheel machine is realized.

[0084] From the specific implementation process of the above examples, it can be known that: first, the global state coefficient representing the overall operation state of the bucket wheel machine is generated by deeply analyzing the electrical parameters to extract three indexes of operation stability, load health degree and overload impact degree; the temperature data of the limited measuring points is reconstructed into a continuous and high-fidelity three-dimensional temperature field by using three-dimensional laser scanning and radial basis function interpolation technology to generate an intuitive thermal image evolution sequence, so as to solve the problem of blind area and incomplete monitoring of temperature monitoring; second, the temperature field of each functional area is analyzed in time sequence, the maximum abnormal temperature contribution total value, the average abnormal contribution rate and the contribution slope are extracted, the global state coefficient is introduced for sensitivity adjustment, the temperature anomaly index which can dynamically reflect the local fault risk is calculated and analyzed, not only considering the local overheating itself, but also considering whether the overall bucket wheel machine can withstand such overheating, realizing risk perception based on the whole and the local; finally, all analysis results are integrated into a composite three-dimensional model: the risk level is intuitively marked by red, yellow and green, and the process data and result data are bound as attribute data with the model to form a digital twin which can be intuitively interacted and contains rich data kernel.

[0085] The technical features of the above-described embodiments can be combined in any manner, and to make the description concise, not all possible combinations of the technical features in the above-described embodiments are described, however, as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present disclosure.

[0086] The above embodiments only express several implementation manners of the present application, and the description is more specific and detailed, but it should not be understood as a limitation on the scope of the patent. It should be noted that for ordinary skilled persons in the art, without departing from the concept of the present application, several modifications and improvements can be made, which are within the scope of protection of the present application. Therefore, the protection scope of the patent of the present application should be subject to the appended claims.

Claims

1. A method for digital twin modeling and fault visualization presentation of thermal measurement points of a bucket wheel machine, characterized in that, The method comprises the following steps: Step one, obtaining the electrical parameters of the bucket wheel machine, and performing in-depth analysis based on the same to output a global state coefficient representing the overall state of the bucket wheel machine; Step two, performing three-dimensional laser scanning on the bucket wheel machine body using a cantilever laser scanning device and modeling to obtain a three-dimensional model of the bucket wheel machine body; dividing the bucket wheel machine body into a plurality of functional regions according to functional areas, constructing a temperature field in the functional region according to the temperatures of the measuring points in each functional region of the bucket wheel machine, and making a thermal image evolution sequence based on the same; Step three, performing time sequence analysis on the functional region based on the thermal image evolution sequence, and supplementing the same with the global state coefficient to adjust the sensitivity, obtaining a temperature anomaly index, constructing a composite three-dimensional model according to the temperature anomaly index, and visualizing the same.

2. The method of claim 1, wherein, In-depth analysis of electrical parameters to output global state coefficient: Step 101, extract electrical parameters, including current, power, load, and its value is recorded as I(t), P(t) and F(t) respectively, wherein t is the index of the collection time, t = 1, 2, 3, …, T, T is the total number of the current collection time; according to the formula The load coefficient K(t) is calculated, wherein I 额 is the rated current of the bucket wheel machine, P 额 is the rated power of the bucket wheel machine; α, β, γ are weight coefficients; Step 102, constructing a two-dimensional rectangular coordinate system with time as the horizontal coordinate and load coefficient as the vertical coordinate, dotting K(t) in the coordinate system according to its corresponding collection time t and load coefficient, and connecting the dots in sequence using a smooth curve to obtain a load coefficient curve; performing image feature analysis on the load coefficient curve to extract feature parameters, wherein the feature parameters include running smoothness, load health degree, and overload impact degree; Step 103, obtaining the global state coefficient by formula calculation of the running smoothness, load health degree, and overload impact degree.

3. The method of claim 2, wherein, Extraction process of running smoothness: The load coefficient K(t) at the collection time is calculated according to the formula The standard deviation of the load coefficient curve is calculated, wherein The mean value of the load coefficient K(t) at each collection time is calculated, and then normalized to obtain the operation stability A. The formula for the normalization processing is 4. The method of claim 3, wherein, Extraction process of load health degree: A healthy interval is preset, two straight lines parallel to the horizontal coordinate are drawn in the load coefficient curve, and the load coefficients of the straight lines are the upper limit and the lower limit of the healthy interval, respectively; the load coefficient curve is divided into a downward portion, a middle portion, and an upward portion according to the two straight lines; the downward portion is the shadow portion formed by the curve and the straight line representing the lower limit of the healthy interval, the middle portion is the shadow portion formed by the curve and the straight lines representing the lower limit and the upper limit of the healthy interval, and the upward portion is the shadow portion formed by the curve and the straight line representing the upper limit of the healthy interval; the sum of all upward portions, middle portions, and downward portions in the curve image is calculated to obtain upward area, middle area, and downward area, respectively, and the sum of upward area, middle area, and downward area is calculated to obtain total area; The upward area, middle area, and downward area are divided by the total area to obtain the proportions of the upward area, middle area, and downward area, respectively, and formula calculation analysis is performed to obtain the load health degree.

5. The method of claim 4, wherein, Extraction process of overload impact degree: The overload impact degree is used to quantify the instantaneous impact and cumulative fatigue damage degree of the bucket wheel machine in the overload state, i.e., K(t)>1.1, and the calculation formula is as follows: where λ is a scaling factor for the overload impact, m is a magnitude amplification exponent, Δt is the sampling time interval, M tot is the total area, and ζ is a constant.

6. The method of claim 1, wherein, Constructing a temperature field in the functional region: Step 201, each functional area of each site corresponds to a plurality of measuring points, and a temperature sensor is installed at each measuring point to monitor the temperature of each measuring point; the temperature corresponding to each measuring point of each functional area is recorded as T i where i = 1, 2, 3, …, N, i represents the index of any one measuring point in the functional area, and N is the total number of measuring points in the functional area; Step 202, separating the geometric model of the functional region from the overall three-dimensional model of the bucket wheel machine, generating a regular three-dimensional grid array inside the functional region, extending the temperature of the discrete measuring points to the entire functional region to construct a temperature field, and calculating the weight of the grid point P affected by the measuring point i using a radial basis function distance weighting method, wherein the grid point P represents other grid points in the functional region except the measuring points; Step 204, the base grid point P will be as the basic unit of temperature field reconstruction, carrying the calculated temperature value; the temperature value of each grid point P in the functional area is obtained by weighted average of all measured point temperatures; Step 203, the temperature value of each grid point in the functional area is converted into intuitive visual display, and a thermal image evolution sequence is made.

7. The method of claim 6, wherein, The weight calculation process of the grid point P affected by the measured point i is: Weights w i (P) The calculation formula is: where Y i is the three-dimensional coordinate of the measurement point i, φ is the orientation of the measurement point i, d is the distance between the point P and the measurement point, i.e. d = ||P - Y i ||.

8. The method of claim 7, wherein, The process of making the thermal image evolution sequence is: The temperature and measured point temperature of the grid point P are converted into color values by using the preset temperature-color mapping table, the mapping color values are filled into the functional area according to the three-dimensional coordinates of the P point and the measured point i respectively, and the thermal map with smooth temperature gradient is generated, so as to realize the reconstruction of discrete measured point temperature into continuous and intuitive three-dimensional temperature field, and the thermal map of the bucket wheel machine at each collection time is sorted according to the corresponding collection time, and the thermal image evolution sequence of the bucket wheel machine in the time period T is generated.

9. The method of claim 1, wherein, The process of generating the composite three-dimensional model is: Step 301, randomly select a functional area, and randomly select any grid point Q in the functional area, wherein P, i Q, that is, Q represents any grid point in the functional area, including the measured point i and other grid points P; A preset warning temperature is set for each functional area of the bucket wheel machine, the temperature corresponding to the grid point Q in the functional area at the collection time t is extracted, and the abnormal temperature contribution value of the grid point Q is obtained by dividing the temperature by the warning temperature corresponding to the functional area; if the contribution value is greater than 1, the contribution value of the grid point is marked as an effective contribution value, and the sum of all effective contribution values in the functional area is calculated to obtain the total abnormal temperature contribution value of the functional area at the collection time t; Step 302, a two-dimensional rectangular coordinate system is constructed with time as the horizontal coordinate and the total abnormal temperature contribution value as the vertical coordinate, and the total abnormal temperature contribution values at each collection time are input into the coordinate system to form a plurality of contribution points, a smooth curve is used to connect the contribution points in sequence to obtain an abnormal temperature contribution total value curve, and the abnormal temperature contribution total value curve is analyzed in time sequence, and the sensitivity is adjusted by using a global state coefficient to obtain a temperature abnormality index; Step 303, a preset abnormal interval is set, when the temperature abnormality index is greater than the upper limit of the abnormal interval, the corresponding position of the functional area in the composite three-dimensional model is marked red, and a temperature alarm is issued; when the temperature abnormality index is within the abnormal interval, the corresponding position of the functional area in the composite three-dimensional model is marked yellow; when the temperature abnormality index is less than the lower limit of the abnormal interval, the corresponding position of the functional area in the composite three-dimensional model is marked green; The thermal image evolution sequence of the bucket wheel machine and the composite three-dimensional model after coloring treatment are visually displayed, and the total abnormal temperature contribution value curve, the maximum abnormal temperature contribution value, the average abnormal contribution rate, the contribution slope and the temperature abnormality index of each functional area are stored as the attribute data of each functional area in the corresponding position of the composite three-dimensional model.

10. The method of claim 9, wherein, The process of generating the temperature abnormality index is: The maximum abnormal temperature contribution total value is extracted from the abnormal temperature contribution total value graph, the area surrounded by the abnormal temperature contribution total value graph and the horizontal axis is obtained by using calculus, the average abnormal contribution rate is obtained by dividing the contribution area by the total time of the abnormal temperature contribution total value graph; the linear fitting is performed on all the contribution points in the abnormal temperature contribution total value graph to obtain a fitting straight line, and the contribution slope of the fitting straight line is calculated; The maximum abnormal temperature contribution total value, the average abnormal contribution rate, the contribution slope and the global state coefficient are calculated and analyzed by formula to output the temperature abnormality index of the functional area, and thus the temperature abnormality index of the bucket wheel machine in each functional area can be obtained.

Images