Method for determining hardening layer depth of speed governor rope sheave subjected to laser quenching treatment

Abstract

The invention provides a method for determining the depth of a hardening layer of a speed limiter rope wheel subjected to laser quenching treatment, and relates to the technical field of speed limiter rope wheel machining, and the method comprises the specific steps that the hardness of the hardening layer at different depths and laser equipment, machining and environment parameters influencing the hardness of the hardening layer are collected through a system, and corresponding influence coefficients are generated; building a quenching environment characteristic model and training to obtain a real-time quenching environment influence coefficient; meanwhile, data processing is conducted on the laser equipment and the machining environment, and influence coefficients are generated; generating correction coefficients by comprehensively analyzing the influence coefficients, dividing the depth of the hardening layer into a shallow type, a medium type and a deep type, and respectively distributing corresponding correction coefficients; and drawing a depth-hardness curve, and obtaining the depth of the hardening layer according to the curve. According to the method, through multi-level systematic analysis and fine treatment, the precision of the depth and hardness of the hardening layer is remarkably improved, and the deviation between an actual measured value and an expected target is reduced.

Description

Technical Field

[0001] The invention relates to the technical field of speed governor rope wheel processing, in particular to a method for determining the depth of a hardened layer of a speed governor rope wheel processed by laser quenching. Background Art

[0002] The method for determining the depth of the hardened layer of the speed limiter rope sheave treated by laser quenching has been continuously developed in terms of technology, measurement, materials and intelligence, and has become an important research direction in modern manufacturing and materials engineering. In the future, with the further development of laser technology and materials science, more breakthroughs are expected in accuracy, efficiency and application scope.

[0003] After analysis and sorting, it was found that in the laser quenching process of the existing technology, the depth and hardness of the hardened layer usually depend on multiple variables, including laser equipment parameters, processing environment and quenching environment. However, the existing methods often lack a systematic analysis of these parameters, resulting in inconsistency in the depth of the hardened layer and deviations in hardness; in addition, in the existing methods, the hardened layer is often regarded as a whole, and the depth of the hardened layer is not divided into multiple layers. There is a lack of detailed analysis of different depth levels, which leads to significant deviations between the actual measured hardened layer depth and hardness values ​​and the expected targets when determining the depth of the hardened layer.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0005] The object of the present invention is to provide a method for determining the depth of the hardened layer of a speed governor rope wheel treated by laser quenching, so as to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for determining the depth of a hardened layer of a speed governor rope wheel treated by laser quenching, the specific steps comprising:

[0008] S1. Collect the original hardness of the hardened layer at different depths and the laser equipment parameters, processing parameters and environmental parameters that affect the hardness of the hardened layer of the speed limiter rope wheel during laser quenching treatment. The laser equipment parameters include laser power, focal length, spot diameter and pulse frequency. The processing environment parameters include laser irradiation time, scanning speed and coolant flow rate. The quenching environment parameters include quenching temperature, quenching humidity and oxygen content.

[0009] S2. Construct a quenching environment characteristic model, take the previous quenching temperature, quenching humidity and oxygen content as input, determine the influence coefficient of the quenching environment based on the expert scoring method, take the corresponding quenching environment influence coefficient as a label training model, train the quenching environment characteristic model, and obtain a trained quenching environment characteristic model;

[0010] S3. Input the current quenching temperature, quenching humidity and oxygen content into the trained quenching environment characteristic model to obtain the influence coefficient of the quenching environment at the current moment;

[0011] S4. Process the laser power, focal length, spot diameter and pulse frequency collected at the current moment to generate the influence coefficient of the laser equipment that affects the hardness of the hardened layer of the speed limiter rope wheel, and process the laser irradiation time, scanning speed and coolant flow rate at the current moment to generate the influence coefficient of the processing environment that affects the hardness of the hardened layer of the speed limiter rope wheel;

[0012] S5. Perform data processing and correlation analysis on the influence coefficient of the current quenching environment, the influence coefficient of the laser equipment and the influence coefficient of the processing environment to generate a comprehensive coefficient for correcting the hardness of the hardened layer. According to the range of the hardened layer depth, the hardened layer depth is divided into three types of hardened layers: shallow, medium and deep. According to the proportion coefficients of the three hardened layers, corresponding correction coefficients are allocated to the three hardened layers.

[0013] S6. Perform data processing on the correction coefficients assigned to the three hardened layers and the corresponding original hardness of the hardened layers to generate correction values ​​of the three hardnesses of the hardness of the hardened layers. Use the three hardened layer depths as the horizontal coordinates and the correction value of the hardness of the hardened layer as the vertical coordinate to draw a depth-hardness curve, and set a hardness threshold for the hardness of the hardened layer. According to the depth-hardness curve, record the point where the corrected hardness value intersects with the hardness threshold. The depth value of the intersection is the depth of the hardened layer.

[0014] Furthermore, the quenching environment characteristic model is composed of a deep neural network based on a multilayer perceptron, and the deep neural network of the multilayer perceptron includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, and the first hidden layer, the second hidden layer and the third hidden layer each have at least two neurons, and each uses ReLU as an activation function;

[0015] The process of training the quenching environment characteristic model is as follows:

[0016] The quenching temperature, quenching humidity and oxygen content are used as input, and the influence coefficient of the quenching environment is used as the output label for training. The mean square error is used as the loss function. When the mean square error is in the range of [0, 0.1], the training of the quenching environment characteristic model is completed.

[0017] Further, based on the correlation between laser power, focal length, spot diameter, pulse frequency, and the hardness of the hardened layer, the laser power, focal length, spot diameter, and pulse frequency collected at the current moment are processed to generate the influence coefficient of the laser device on the hardness of the governor rope wheel. The formula is as follows:

[0018]

[0019] Among them, XS SB is the influence coefficient of the laser device, P is the laser power, F is the focal length, D is the spot diameter, f is the pulse frequency, a is the influence coefficient of laser power on hardness, 0.1 < a < 10, b is the correction coefficient of focal length on hardness, 0.1 < b < 1, c is a constant, 0 < c < 5, α is the exponential influence coefficient of spot diameter, 1 < α < 2, β is the exponential decay coefficient of pulse frequency, 0.1 < β < 1.

[0020] Further, based on the correlation between laser irradiation time, scanning speed, coolant flow rate, and the hardness of the hardened layer, the laser irradiation time, scanning speed, and coolant flow rate at the current moment are processed to generate the influence coefficient of the processing environment on the hardness of the governor rope wheel. The formula is as follows:

[0021]

[0022] Among them, XS JG is the influence coefficient of the processing environment, T is the laser irradiation time, S is the scanning speed, C is the coolant flow rate, B is the adjustment coefficient of scanning speed, 0.1 < B < 10, Q is the influence coefficient of coolant flow rate, 0.1 < Q < 5.

[0023] Further, the influence coefficient of the quenching environment, the influence coefficient of the laser device, and the influence coefficient of the processing environment at the current moment are processed and correlation analyzed to generate a comprehensive coefficient for correcting the hardness of the hardened layer. The formula is as follows:

[0024] ZS = ω 1 ·XS CH + ω 2 ·XS SB + ω 3 ·XS JG

[0025] Among them, ZS is the comprehensive coefficient, XS CH is the influence coefficient of the quenching environment, ω 1 is the weight coefficient of the influence coefficient of the quenching environment, ω 2 is the weight coefficient of the influence coefficient of the laser device, ω 3 is the weight coefficient of the influence coefficient of the processing environment, 0 < ω 3<ω 2 <ω 1 <1, and ω 3 +ω 2 +ω 1 =1.

[0026] Furthermore, according to the range of the hardened layer depth, the hardened layer depth is divided into three types of hardened layers: shallow, medium and deep, and according to the proportion coefficients of the three hardened layers, corresponding correction coefficients are allocated to the three hardened layers. The specific process is as follows:

[0027] According to the range of the hardened layer, it is divided into shallow, middle and deep layers:

[0028] Shallow hardened layer: Depth SD sh ∈[0,d y1 ];

[0029] Middle hardened layer: Depth SD me ∈(d y1 ,d y2 ];

[0030] Deep hardened layer: Depth SD de ∈(d y2 ,d max ];

[0031] Among them, d y1 is the first depth threshold, d y2 is the second depth threshold, d max is the maximum hardened layer depth;

[0032] The proportional coefficients of the shallow hardened layer, the medium hardened layer, and the deep hardened layer are calculated based on the following formula:

[0033]

[0034] Among them, ZB 1 is the proportion coefficient of the shallow hardened layer, ZB 2 is the proportion coefficient of the hardened layer, ZB 3 is the proportion coefficient of deep hardened layer;

[0035] The correction coefficients for the shallow hardened layer, the medium hardened layer, and the deep hardened layer are calculated based on the following formula:

[0036] QS=ZB 1 ×ZS

[0037] MS=ZB 2 ×ZS

[0038] SS=ZB 3 ×ZS

[0039] Among them, QS is the correction coefficient of the shallow hardened layer, MS is the proportion coefficient of the medium hardened layer, and SS is the proportion coefficient of the deep hardened layer.

[0040] Furthermore, the correction coefficients assigned to the three hardened layers and the corresponding original hardness of the hardened layers are processed to generate correction values ​​of the hardness of the three hardened layers according to the following formula:

[0041]

[0042] Among them, H sh H is the hardness correction value of the shallow hardened layer, me H is the hardness correction value of the middle hardened layer, de is the hardness correction value of the deep hardened layer, is the original hardness of the shallow hardened layer, is the original hardness of the middle hardened layer, It is the original hardness of the deep hardened layer.

[0043] Furthermore, according to the depth-hardness curve, the point where the corrected hardness value intersects with the hardness threshold is recorded. The depth value of the intersection is the depth of the hardened layer. The specific process is as follows:

[0044] In order to accurately determine the intersection point, linear interpolation is used to calculate the intersection point, set at depth d 1 The hardness value is H 1 , at depth d 2 The hardness value is H 2 , and H 1 <H min <H 2 ;

[0045] The depth value of the intersection is calculated using the linear interpolation formula:

[0046]

[0047] Among them, H min is the minimum hardness, and d is the depth of the intersection, that is, the depth of the hardened layer.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] The present invention collects the original hardness of the hardened layer at different depths and the laser equipment parameters, processing parameters and environmental parameters that affect the hardness of the hardened layer of the speed limiter rope wheel during laser quenching treatment, generates the influence coefficients of the quenching environment, laser equipment and processing environment, performs data processing and correlation analysis on the influence coefficients of the quenching environment, laser equipment and processing environment, generates a comprehensive coefficient for correcting the hardness of the hardened layer, divides the hardened layer depth into three hardened layers of shallow, medium and deep according to the range of the hardened layer depth, and allocates corresponding correction coefficients to the three hardened layers according to the proportion coefficients of the three hardened layers, performs data processing on the correction coefficients allocated to the three hardened layers and the corresponding original hardened layer hardness, generates the correction values ​​of the three hardened layer hardnesses, draws the depth-hardness curve, and obtains the depth of the hardened layer according to the depth-hardness curve. Therefore, through the systematic analysis of the influencing parameters, and dividing the hardened layer depth into multiple layers, the different depth levels are finely analyzed and processed, the accuracy of the hardened layer depth and hardness is improved, so that the deviation between the actual measured hardened layer depth and hardness value and the expected target is significantly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic diagram of the overall method flow of the present invention. DETAILED DESCRIPTION

[0051] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0052] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0053] Embodiment 1:

[0054] See also Figure 1 , the present invention provides a technical solution:

[0055] A method for determining the depth of a hardened layer of a speed governor rope wheel treated by laser quenching, the specific steps comprising:

[0056] S1. Collect the original hardness of the hardened layer at different depths and the laser equipment parameters, processing parameters and environmental parameters that affect the hardness of the hardened layer of the speed limiter rope wheel during laser quenching treatment. The laser equipment parameters include laser power, focal length, spot diameter and pulse frequency. The processing environment parameters include laser irradiation time, scanning speed and coolant flow rate. The quenching environment parameters include quenching temperature, quenching humidity and oxygen content.

[0057] S2. Construct a quenching environment characteristic model, take the previous quenching temperature, quenching humidity and oxygen content as input, determine the influence coefficient of the quenching environment based on the expert scoring method, take the corresponding quenching environment influence coefficient as a label training model, train the quenching environment characteristic model, and obtain a trained quenching environment characteristic model;

[0058] S3. Input the current quenching temperature, quenching humidity and oxygen content into the trained quenching environment characteristic model to obtain the influence coefficient of the quenching environment at the current moment;

[0059] S4. Process the laser power, focal length, spot diameter and pulse frequency collected at the current moment to generate the influence coefficient of the laser equipment that affects the hardness of the hardened layer of the speed limiter rope wheel, and process the laser irradiation time, scanning speed and coolant flow rate at the current moment to generate the influence coefficient of the processing environment that affects the hardness of the hardened layer of the speed limiter rope wheel;

[0060] S5. Perform data processing and correlation analysis on the influence coefficient of the current quenching environment, the influence coefficient of the laser equipment and the influence coefficient of the processing environment to generate a comprehensive correction coefficient for correcting the hardness of the hardened layer. According to the range of the hardened layer depth, the hardened layer depth is divided into three types of hardened layers: shallow, medium and deep. According to the proportion coefficients of the three hardened layers, corresponding correction coefficients are allocated to the three hardened layers.

[0061] S6. Perform data processing on the correction coefficients assigned to the three hardened layers and the corresponding original hardness of the hardened layers to generate correction values ​​of the three hardnesses of the hardness of the hardened layers. Use the three hardened layer depths as the horizontal coordinates and the correction value of the hardness of the hardened layer as the vertical coordinate to draw a depth-hardness curve, and set a hardness threshold for the hardness of the hardened layer. According to the depth-hardness curve, record the point where the corrected hardness value intersects with the hardness threshold. The depth value of the intersection is the depth of the hardened layer.

[0062] Based on the above embodiment, the quenching environment characteristic model is composed of a deep neural network based on a multilayer perceptron, and the deep neural network of the multilayer perceptron includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer. The first hidden layer, the second hidden layer and the third hidden layer each have at least two neurons, and each uses ReLU as an activation function.

[0063] In this embodiment, the input features of the deep neural network of the multilayer perceptron include three features: quenching temperature, quenching humidity and oxygen content.

[0064] The structure of a deep neural network with a multilayer perceptron is:

[0065] Input layer: receives input of 3 features;

[0066] The first hidden layer has 64 neurons and uses ReLU as the activation function.

[0067] The second hidden layer has 128 neurons and also uses the ReLU activation function.

[0068] The third hidden layer has 64 neurons and uses the ReLU activation function.

[0069] Output layer: has a single neuron and the influence coefficient of the quenching environment at the current moment.

[0070] The greater the influence coefficient of the quenching environment, the better the quenching effect and the deeper the hardened layer formed.

[0071] The process of training the quenching environment characteristic model is as follows:

[0072] The quenching temperature, quenching humidity and oxygen content are used as input, and the influence coefficient of the quenching environment is used as the output label for training. The mean square error is used as the loss function. When the mean square error is in the range of [0, 0.1], the training of the quenching environment characteristic model is completed.

[0073] Based on the above embodiments, within a certain range, the correlation between laser power, focal length, spot diameter, pulse frequency and hardness of the hardened layer is as follows:

[0074] The laser power is positively correlated with the hardness of the hardened layer.

[0075] Because the higher the laser power, the stronger the heating effect on the material surface, and the harder the hardened layer produced.

[0076] The focal length is negatively correlated with the hardness of the hardened layer.

[0077] Because the shorter the focal length, the smaller the spot size, the higher the energy density, and the more concentrated the local heating, the harder the hardened layer will be.

[0078] The spot diameter is negatively correlated with the hardness of the hardened layer.

[0079] As the spot diameter increases, the energy of the laser is dispersed over a larger surface area, resulting in a decrease in the energy density per unit area. This leads to uneven heating, a slower rate of increase in surface temperature, the formation of a shallower hardened layer, and a decrease in hardness due to the reduced cooling rate.

[0080] The pulse frequency and the hardness of the hardened layer are negatively correlated.

[0081] Because the higher the pulse frequency, the shorter the interval between laser pulses, and the material does not have enough time to cool sufficiently after each pulse heating. In this case, the surface temperature remains at a relatively high level, leading to grain growth and incomplete phase transformation, resulting in a lower hardness of the hardened layer.

[0082] Based on the correlations between the laser power, focal length, spot diameter, pulse frequency, and the hardness of the hardened layer, the laser power, focal length, spot diameter, and pulse frequency collected at the current moment are processed to generate the influence coefficient of the laser device on the hardness of the governor rope wheel. The formula is as follows:

[0083]

[0084] Where, XS SB is the influence coefficient of the laser device, P is the laser power, F is the focal length, D is the spot diameter, f is the pulse frequency, a is the influence coefficient of the laser power on the hardness, 0.1 < a < 10, b is the correction coefficient of the focal length on the hardness, 0.1 < b < 1, c is a constant, 0 < c < 5, α is the exponential influence coefficient of the spot diameter, 1 < α < 2, β is the exponential decay coefficient of the pulse frequency, 0.1 < β < 1.

[0085] The influence coefficient XS of the laser device SB has a value range of [0,1].

[0086] When XS SB is close to 0, it means that the influence of the parameter combination of the laser device is small;

[0087] When XS SB is close to 1, it means that the influence of the parameter combination of the laser device is large.

[0088] Then, the larger the influence coefficient XS of the laser device SB , the greater the influence of the parameter combination of the laser device.

[0089] The reasons for using the above functional form to express the laser power, focal length, spot diameter, pulse frequency, and the influence coefficient of the laser device are as follows:

[0090] First, laser power is the direct source of laser beam energy. The higher the power, the more energy is transferred to the material per unit time, resulting in a higher thermal effect, which directly affects the formation of the hardened layer.

[0091] Use square P 2 This is because the effect of laser power on material hardness is usually nonlinear. Higher power will lead to more significant temperature increase and stronger phase change effect, so the square term can better describe this acceleration effect.

[0092] Second, the focal length affects the focusing degree of the laser beam. A shorter focal length can form a higher energy density on the material surface, resulting in faster heating and a deeper hardened layer.

[0093] The correction factor b is introduced to adjust the effect of focal length on hardness and ensure that the model can maintain consistency under different focal lengths. The coefficient b can be determined experimentally to reflect the effect of focal length on laser energy distribution.

[0094] Third, the spot diameter determines the distribution range of laser energy. A larger spot diameter means energy dispersion, resulting in lower energy density per unit area and a weaker effect on hardness.

[0095] The exponent α is used because the effect of spot diameter on hardness is not linear. As the spot diameter increases, the effect may drop significantly, so choosing an exponent greater than 1 better captures this change in effect.

[0096] Fourth, the pulse frequency affects the heat accumulation effect of the material. A higher frequency may cause excessive heating of the material surface, thus affecting the formation of hardness.

[0097] Use exponential decay -βf This is to reflect the decreasing effect of frequency on hardness. When the pulse frequency increases, the effective utilization of energy may decrease, so the introduced exponential decay can effectively express this.

[0098] Fifth, the constant c is introduced to adjust the model of the effect of focal length on hardness so that when the focal length is 0, the model will not produce unreasonable results (such as division by zero, etc.); using the form of min(1,…) to limit the influence coefficient to the range of [0,1], it is easier to understand the influence.

[0099] To increase the accuracy of calculating the influence coefficient of a laser device on the hardened layer hardness of a governor rope pulley, a data set including the laser power, focal length, spot diameter, and pulse frequency of the set is constructed. The entire data set is divided into a training set and a validation set. 80% of the data is used for training, and 20% of the data is used for validation. By using the laser power, focal length, spot diameter, and pulse frequency of the set as inputs and the corresponding influence coefficient of the laser device as the output label, the laser device model of the governor rope pulley is trained. The backpropagation algorithm is used to optimize a, b, α, and β corresponding to the influence coefficient of the laser device respectively. The mean square error is used as the loss function. When the mean square error is within the range of [0, 0.1], the training of the laser device model of the governor rope pulley is completed. Once the training is completed, the obtained influence coefficient a, correction coefficient b, exponential influence coefficient α, and exponential decay coefficient β are used to predict the influence coefficient of a new laser device.

[0100] On the basis of the above embodiments, within a certain range, the correlations between the laser irradiation time, scanning speed, coolant flow rate, and the hardness of the hardened layer are as follows:

[0101] The laser irradiation time is positively correlated with the hardness of the hardened layer.

[0102] Because a longer irradiation time allows the material to fully heat up to the austenitizing temperature, thus promoting the formation of martensite and ultimately increasing the hardness.

[0103] The scanning speed is negatively correlated with the hardness of the hardened layer.

[0104] Because a slower scanning speed allows more energy to be concentrated on the material surface, increasing its energy density and promoting austenitization, thereby increasing the hardness.

[0105] The coolant flow rate is negatively correlated with the hardness of the hardened layer.

[0106] Because a higher cooling rate can effectively lock the martensite transformation and form a hardened layer with high hardness.

[0107] Based on the correlations between the laser irradiation time, scanning speed, coolant flow rate, and the hardness of the hardened layer, the laser irradiation time, scanning speed, and coolant flow rate at the current moment are processed to generate the influence coefficient of the processing environment affecting the hardness of the hardened layer of the governor rope pulley. The formula is as follows:

[0108]

[0109] Among them, XS JG is the influence coefficient of the processing environment, T is the laser irradiation time, S is the scanning speed, C is the coolant flow rate, B is the adjustment coefficient of the scanning speed, 0.1 < B < 10, Q is the influence coefficient of the coolant flow rate, 0.1 < Q < 5.

[0110] Processing environment influence coefficient XS JG The value range of is [0,1].

[0111] When XS JG When it is close to 0, it means that the parameter combination of the processing environment has little influence;

[0112] When XS JG When it is close to 1, it means that the parameter combination of the processing environment has a great influence.

[0113] Then the influence coefficient of the processing environment is XS JG The larger it is, the greater the influence of the parameter combination of the processing environment.

[0114] The reasons for using the above function form to express the influence coefficients of laser irradiation time, scanning speed, coolant flow rate and processing environment are as follows:

[0115] First, the longer the laser irradiation time T, the higher the temperature of the processing area, and the greater the possibility of material melting and recrystallization. However, once the material reaches a certain temperature, too long a time may cause the heat-affected zone to expand and reduce material performance.

[0116] The exponential function can simulate the nonlinear characteristics of this heating process, that is, after heating to a certain temperature, the marginal effect of continuing heating decreases.

[0117] Second, when the scanning speed S increases, the laser stays on the material for a shorter time, the heat input decreases, and the melting and phase change of the material also decrease. At this time, the cooling rate of the material increases.

[0118] A higher scanning speed S will rapidly reduce the influence coefficient XS JG , which can be effectively captured by an exponential decay function.

[0119] Third, the coolant flow rate C directly affects the speed of heat removal. The faster the flow rate C, the faster the heat is removed, and the better the cooling performance of the material. Especially after laser irradiation, the change of coolant flow rate C can significantly affect the quenching effect.

[0120] Using Q as the influence coefficient, the intensity of this cooling effect can be quantified, especially at high scanning speeds, the influence of the coolant flow rate C is more significant.

[0121] Fourth, in laser processing, the interaction between laser irradiation time T, scanning speed S and coolant flow rate C is often complex. For example, when the scanning speed S increases, although the laser irradiation time T is relatively short, if the coolant flow rate C is also increased accordingly, it means that the material may still be able to obtain enough energy to melt within the shorter laser irradiation time T, and the increase in coolant flow rate C can quickly take away excess heat and prevent the material from overheating. The exponential function form e -Q·S Able to flexibly express this nonlinear relationship.

[0122] In order to increase the accuracy of calculating the influence coefficient of the processing environment that affects the hardness of the hardened layer of the speed limiter rope pulley, a data set including laser irradiation time, scanning speed and coolant flow rate is constructed, and the entire data set is divided into a training set and a validation set. 80% of the data is used for training and 20% of the data is used for validation. The processing environment model of the speed limiter rope pulley is trained by taking the laser irradiation time, scanning speed and coolant flow rate as inputs and the corresponding processing environment influence coefficients as output labels. The back propagation algorithm is used to optimize B and Q corresponding to the processing environment influence coefficients respectively. The mean square error is used as the loss function. When the mean square error is in the range of [0,0.1], the training of the processing environment model of the speed limiter rope pulley is completed. Once the training is completed, the obtained adjustment coefficient B and influence coefficient Q are used to predict the influence coefficient of the new processing environment.

[0123] On the basis of the above embodiment, the influence coefficient of the current quenching environment, the influence coefficient of the laser equipment and the influence coefficient of the processing environment are subjected to data processing and correlation analysis to generate a comprehensive coefficient for correcting the hardness of the hardened layer, according to the following formula:

[0124] ZS=ω 1 ·XS CH +ω 2 ·XS SB +ω 3 ·XS JG

[0125] Among them, ZS is the comprehensive coefficient, XS CH is the influence coefficient of quenching environment, ω 1 is the weight coefficient of the influence coefficient of the quenching environment, ω 2 is the weight coefficient of the influence coefficient of the laser equipment, ω 3 is the weight coefficient of the influence coefficient of the processing environment.

[0126] The influence coefficient of quenching environment mainly involves the following factors:

[0127] Quenching temperature: determines the phase change and hardness of the material. Too high or too low will directly affect the formation of the final hardness.

[0128] Quenching humidity: affects the cooling effect and quenching rate. Too high or too low humidity may lead to inconsistent hardness.

[0129] Oxygen content: High oxygen content may cause oxidation of the material and affect the surface hardness.

[0130] The quenching environment has a direct and significant impact on material properties, so 1 It is the largest of the three impact coefficients.

[0131] Although laser equipment parameters are crucial in the processing process, their influence is usually reflected through the quenching environment. Relatively speaking, the influence of laser equipment is smaller than that of quenching environment. 2 <ω 1 .

[0132] Although the parameters of the processing environment also affect the hardness, they are usually considered to be secondary factors because their influence is often indirect and their fluctuations have little effect on the hardness. 3 Minimum.

[0133] In summary, the weight coefficient of the influence coefficient of the quenching environment is greater than the weight coefficient of the influence coefficient of the laser equipment, and the weight coefficient of the influence coefficient of the laser equipment is greater than the weight coefficient of the influence coefficient of the processing environment, that is, 0<ω 3 <ω 2 <ω 1 <1, and ω 3 +ω 2 +ω 1 =1.

[0134] On the basis of the above embodiment, according to the range of the hardened layer depth, the hardened layer depth is divided into three types of hardened layers: shallow, medium and deep, and according to the proportion coefficients of the three hardened layers, corresponding correction coefficients are allocated to the three hardened layers. The specific process is as follows:

[0135] According to the range of the hardened layer, it is divided into shallow, middle and deep layers:

[0136] Shallow hardened layer: Depth SD sh ∈[0,d y1 ];

[0137] Middle hardened layer: Depth SD me ∈(d y1 ,d y2 ];

[0138] Deep hardened layer: Depth SD de ∈(d y2 ,d max ];

[0139] Among them, dy1 is the first depth threshold, d y2 is the second depth threshold, d max is the maximum hardened layer depth;

[0140] The proportional coefficients of the shallow hardened layer, the medium hardened layer, and the deep hardened layer are calculated based on the following formula:

[0141]

[0142] Among them, ZB 1 is the proportion coefficient of the shallow hardened layer, ZB 2 is the proportion coefficient of the middle hardened layer, ZB 3 is the proportion coefficient of deep hardened layer;

[0143] The correction coefficients for the shallow hardened layer, the medium hardened layer, and the deep hardened layer are calculated based on the following formula:

[0144] QS=ZB 1 ×ZS

[0145] MS=ZB 2 ×ZS

[0146] SS=ZB 3 ×ZS

[0147] Among them, QS is the correction coefficient of the shallow hardened layer, MS is the proportion coefficient of the medium hardened layer, and SS is the proportion coefficient of the deep hardened layer.

[0148] On the basis of the above embodiment, the correction coefficients assigned to the three hardened layers and the corresponding original hardness of the hardened layers are processed to generate correction values ​​of the hardness of the three hardened layers according to the following formula:

[0149]

[0150] Among them, H sh H is the hardness correction value of the shallow hardened layer, me H is the hardness correction value of the middle hardened layer, de is the hardness correction value of the deep hardened layer, is the original hardness of the shallow hardened layer, is the original hardness of the middle hardened layer, It is the original hardness of the deep hardened layer.

[0151] On the basis of the above embodiment, according to the depth-hardness curve, the point where the corrected hardness value intersects with the hardness threshold is recorded, and the depth value of the intersection is the depth of the hardened layer. The specific process is as follows:

[0152] In order to accurately determine the intersection point, linear interpolation is used to calculate the intersection point, set at depth d 1The hardness value is H 1 , at depth d 2 The hardness value is H 2 , and H 1 <H min <H 2 ;

[0153] The depth value of the intersection is calculated using the linear interpolation formula:

[0154]

[0155] Among them, H min is the minimum hardness, and d is the depth of the intersection, that is, the depth of the hardened layer.

[0156] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0157] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by computer software, electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0158] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0159] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

[0160] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0161] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for determining the depth of the hardened layer of a speed governor sheave treated by laser quenching, characterized in that: The specific steps include: S1. Collect the original hardness of the hardened layer at different depths and the laser equipment parameters, processing parameters and environmental parameters that affect the hardness of the hardened layer of the speed limiter rope wheel during laser quenching treatment. The laser equipment parameters include laser power, focal length, spot diameter and pulse frequency. The processing environment parameters include laser irradiation time, scanning speed and coolant flow rate. The quenching environment parameters include quenching temperature, quenching humidity and oxygen content. S2. Construct a quenching environment characteristic model, take the previous quenching temperature, quenching humidity and oxygen content as input, determine the influence coefficient of the quenching environment based on the expert scoring method, take the corresponding quenching environment influence coefficient as a label training model, train the quenching environment characteristic model, and obtain a trained quenching environment characteristic model; S3. Input the current quenching temperature, quenching humidity and oxygen content into the trained quenching environment characteristic model to obtain the influence coefficient of the quenching environment at the current moment; S4. Process the laser power, focal length, spot diameter and pulse frequency collected at the current moment to generate the influence coefficient of the laser equipment that affects the hardness of the hardened layer of the speed limiter rope wheel, and process the laser irradiation time, scanning speed and coolant flow rate at the current moment to generate the influence coefficient of the processing environment that affects the hardness of the hardened layer of the speed limiter rope wheel; S5. Perform data processing and correlation analysis on the influence coefficient of the current quenching environment, the influence coefficient of the laser equipment and the influence coefficient of the processing environment to generate a comprehensive coefficient for correcting the hardness of the hardened layer. According to the range of the hardened layer depth, the hardened layer depth is divided into three types of hardened layers: shallow, medium and deep. According to the proportion coefficients of the three hardened layers, corresponding correction coefficients are allocated to the three hardened layers. S6. Perform data processing on the correction coefficients assigned to the three hardened layers and the corresponding original hardness of the hardened layers to generate correction values ​​of the three hardnesses of the hardness of the hardened layers. Use the three hardened layer depths as the horizontal coordinates and the correction value of the hardness of the hardened layer as the vertical coordinate to draw a depth-hardness curve, and set a hardness threshold for the hardness of the hardened layer. According to the depth-hardness curve, record the point where the corrected hardness value intersects with the hardness threshold. The depth value of the intersection is the depth of the hardened layer.

2. The method for determining the depth of the hardened layer of the speed governor rope sheave treated by laser quenching according to claim 1 is characterized in that: The quenching environment characteristic model is composed of a deep neural network based on a multilayer perceptron, and the deep neural network of the multilayer perceptron includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, and the first hidden layer, the second hidden layer and the third hidden layer each have at least two neurons, and each uses ReLU as an activation function; The process of training the quenching environment characteristic model is as follows: The quenching temperature, quenching humidity and oxygen content are used as input, and the influence coefficient of the quenching environment is used as the output label for training. The mean square error is used as the loss function. When the mean square error is in the range of [0, 0.1], the training of the quenching environment characteristic model is completed.

3. The method for determining the depth of the hardened layer of the speed governor rope sheave treated by laser quenching according to claim 2 is characterized in that: According to the correlation between laser power, focal length, spot diameter, pulse frequency and hardness of the hardened layer, the laser power, focal length, spot diameter and pulse frequency collected at the current moment are processed to generate the influence coefficient of the laser equipment affecting the hardness of the hardened layer of the speed limiter rope wheel, based on the following formula: Among them, XS SB is the influence coefficient of the laser equipment, P is the laser power, F is the focal length, D is the spot diameter, f is the pulse frequency, a is the influence coefficient of laser power on hardness, 0.1<a<10, b is the correction coefficient of the focal length on hardness, 0.1<b<1, c is a constant, 0<c<5, α is the exponential influence coefficient of the spot diameter, 1<α<2, β is the exponential attenuation coefficient of the pulse frequency, 0.1<β<1.

4. The method for determining the depth of the hardened layer of the speed governor rope sheave treated by laser quenching according to claim 3 is characterized in that: According to the correlation between laser irradiation time, scanning speed, coolant flow rate and hardness of the hardened layer, the laser irradiation time, scanning speed and coolant flow rate at the current moment are processed to generate the influence coefficient of the processing environment that affects the hardness of the hardened layer of the speed limiter rope wheel. The formula is as follows: Among them, XS JG is the influence coefficient of the processing environment, T is the laser irradiation time, S is the scanning speed, C is the coolant flow rate, B is the adjustment coefficient of the scanning speed, 0.1<B<10, Q is the influence coefficient of the coolant flow rate, 0.1<Q<5.

5. The method for determining the depth of the hardened layer of the speed governor rope sheave treated by laser quenching according to claim 4 is characterized in that: The influence coefficients of the current quenching environment, the laser equipment and the processing environment are processed and correlated to generate a comprehensive coefficient for correcting the hardness of the hardened layer. The formula is as follows: ZS=ω1·XS CH + ω2·XS SB + ω3·XS JG Among them, ZS is the comprehensive coefficient, XS CH is the influence coefficient of quenching environment, ω1 is the weight coefficient of the influence coefficient of quenching environment, ω2 is the weight coefficient of the influence coefficient of laser equipment, ω3 is the weight coefficient of the influence coefficient of processing environment, 0<ω3<ω2<ω1<1. And ω3+ω2+ω1=1.

6. The method for determining the depth of the hardened layer of the speed governor sheave treated by laser quenching according to claim 5, characterized in that: According to the range of the hardened layer depth, the hardened layer depth is divided into shallow, medium and deep hardened layers, and according to the proportion coefficients of the three hardened layers, the corresponding correction coefficients are allocated to the three hardened layers. The specific process is as follows: According to the range of the hardened layer, it is divided into shallow, middle and deep layers: Shallow hardened layer: Depth SD sh ∈[0,d y1 ]; Middle hardened layer: Depth SD me ∈(d y1 , d y2 ]; Deep hardened layer: Depth SD de ∈(d y2 , d max ]; Among them, d y1 is the first depth threshold, d y2 is the second depth threshold, d max is the maximum hardened layer depth; The proportional coefficients of the shallow hardened layer, the medium hardened layer, and the deep hardened layer are calculated based on the following formula: Among them, ZB1 is the proportion coefficient of the shallow hardened layer, ZB2 is the proportion coefficient of the medium hardened layer, and ZB3 is the proportion coefficient of the deep hardened layer; The correction coefficients for the shallow hardened layer, the medium hardened layer, and the deep hardened layer are calculated based on the following formula: QS=ZB1×ZS MS=ZB2×ZS SS=ZB3×ZS Among them, QS is the correction coefficient of the shallow hardened layer, MS is the proportion coefficient of the medium hardened layer, and SS is the proportion coefficient of the deep hardened layer.

7. The method for determining the depth of the hardened layer of the speed governor rope sheave treated by laser quenching according to claim 6, characterized in that: The correction coefficients assigned to the three hardened layers and the corresponding original hardness of the hardened layers are processed to generate correction values ​​of the hardness of the three hardened layers according to the following formula: Among them, H sh H is the hardness correction value of the shallow hardened layer, me H is the hardness correction value of the middle hardened layer, de is the hardness correction value of the deep hardened layer, is the original hardness of the shallow hardened layer, is the original hardness of the middle hardened layer, It is the original hardness of the deep hardened layer.

8. The method for determining the depth of the hardened layer of the speed governor rope sheave treated by laser quenching according to claim 7, characterized in that: According to the depth-hardness curve, record the point where the corrected hardness value intersects with the hardness threshold. The depth value of the intersection is the depth of the hardened layer. The specific process is as follows: In order to accurately determine the intersection point, linear interpolation is used to calculate the intersection point. It is set at depth d1, the hardness value is H1, at depth d2, the hardness value is H2, and H1<H min <H2; The depth value of the intersection is calculated using the linear interpolation formula: Among them, H min is the minimum hardness, and d is the depth of the intersection, that is, the depth of the hardened layer.

Images