Electrodeposition coating equipment, electrodeposition coating method, trained model, method for calculating optimal operating conditions for electrodeposition using the trained model

Abstract

To provide an electrodeposition coating facility in which loss of an electrodeposition paint can be reduced by detecting the shortage or excess of thickness of an electrodeposition paint film caused by unintended fluctuation of conditions, and by automatically updating an optimal operating condition.SOLUTION: An electrodeposition coating facility 10 of the present invention is a facility for performing electrodeposition coating to a vehicle body W1 immersed into an electrodeposition paint. The electrodeposition coating facility 10 includes film thickness measurement means 51 and operating condition calculation means. The film thickness measurement means 51 is arranged between an electrodeposition tank and a drying furnace to measure a wet film thickness of an electrodeposition paint film formed on a surface of the vehicle body W1 for each vehicle body W1. The operating condition calculation means includes: calculating an optimal operating condition for achieving a target value of a dry film thickness of the electrodeposition paint film based on a measurement value of the wet film thickness; and outputting the calculated optimal operating condition to an electrodeposition tank control device.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to electrodeposition coating equipment that performs electrodeposition coating on a vehicle body immersed in an electrodeposition coating material to . Background Art

[0002] Conventionally, electrodeposition coating equipment that performs electrodeposition coating on a vehicle body immersed in an electrodeposition coating material in an electrodeposition tank is known (see, for example, Patent Document 1). Electrodeposition coating using such electrodeposition coating equipment is widely employed in the undercoating process of vehicle body coating for the purpose of imparting rust prevention performance. The film thickness of the electrodeposition coating film formed on the surface of the vehicle body is obtained by determining the optimum value of electrode voltage (optimum operating conditions) based on equipment design conditions, vehicle conditions (conditions such as vehicle shape) and conditions of the electrodeposition coating material (bath liquid) (bath conditions), and the equipment is operated and managed under settings determined in advance through experiments. Prior Art Documents Patent Documents

[0003] Patent Document 1 Japanese Unexamined Patent Application Publication No. 2020-132903 (Figures 1, 4, etc.) Summary of the Invention Problems to be Solved by the Invention

[0004] By the way, in electrodeposition coating, the above-mentioned various conditions are rarely maintained constant, and change from moment to moment, for example, due to fluctuations in bath conditions over time. However, every time bath conditions fluctuate, advanced knowledge and know-how are required for an operator to determine the optimum value and change the equipment settings. Therefore, conventionally, in consideration of insufficient or excessive film thickness of the electrodeposition coating film caused by fluctuations in bath conditions, an excessive set value with a high safety factor is used as the optimum value. However, on the other hand, since the set value is not reviewed for a long period of time, there is a problem that loss is likely to occur in the usage amount of the electrodeposition coating material and the like.

[0005] This invention was made in view of the above-mentioned problems, the The objective is to reduce the loss of electrodeposited paint in electrodeposition coating equipment by detecting insufficient or excessive film thickness in the electrodeposition coating film caused by unintended fluctuations in conditions and automatically updating the optimal operating conditions. of The purpose is to provide. [Means for solving the problem]

[0006] To solve the above problems, the invention described in claim 1 is an electrodeposition coating apparatus comprising an electrodeposition tank for storing electrodeposition paint, a drying oven for drying the vehicle body that has been transported out of the electrodeposition tank, and an electrodeposition tank control device for setting the operating conditions of the electrodeposition tank and controlling the electrodeposition tank, wherein electrodeposition coating is performed on the vehicle body that has been immersed in the electrodeposition paint, and the apparatus comprises a film thickness measuring means disposed between the electrodeposition tank and the drying oven for measuring the wet film thickness of the electrodeposition coating film formed on the surface of the vehicle body for each vehicle body, The system includes a model building means for constructing a trained model that stores the measured wet film thickness, the adjustment factors during operation of the electrodeposition tank, and the composition condition factors of the electrodeposition paint as input information, and outputs a target value for the dry film thickness of the electrodeposition coating, and for achieving the target dry film thickness calculated using the trained model. The system includes an operating condition calculation means that outputs optimal operating conditions to the electrodeposition tank control device. The operating condition calculation means monitors the error between the measured value of the wet film thickness and the target value of the dry film thickness, and the model building means automatically updates the trained model when the error between the measured value of the wet film thickness and the target value of the dry film thickness exceeds an acceptable range. The essence of this invention is an electrodeposition coating apparatus characterized by the following features.

[0007] In the invention described in claim 1, even if the electrodeposited coating film becomes under-filmed or over-filmed due to unintended fluctuations in conditions, the film thickness measuring means measures the wet film thickness, and the operating condition calculation means calculates the optimal operating conditions for the electrodeposited tank to achieve the target dry film thickness of the electrodeposited coating film based on the wet film thickness measurement. As a result, the optimal operating conditions are automatically updated, and the operating condition calculation means outputs the calculated optimal operating conditions to the electrodeposited tank control device, which then controls the electrodeposited tank. This control allows for constant optimization of the dry film thickness of the electrodeposited coating film, improving the paint quality of the vehicle body and minimizing the loss of electrodeposited paint.

[0008] While cationic and anionic electrodeposition coatings are available, cationic electrodeposition coatings are preferable from the standpoint of rust prevention.

[0009] In the invention described in claim 1 above,The operating condition calculation means includes a model building means that stores the measured value of the wet film thickness, the adjustment factors during operation of the electrodeposition tank, and the composition condition factors of the electrodeposition paint as input information, and builds a learned model that outputs a target value for the dry film thickness, and outputs the optimal operating conditions calculated using the learned model to the electrodeposition tank control device.

[0010] Therefore, By accumulating wet film thickness measurements, adjustment factors during electrodeposition tank operation, and compositional condition factors of the electrodeposition paint for a trained model, the accuracy of the target dry film thickness can be improved. Therefore, by using the trained model to calculate the optimal operating conditions for the electrodeposition tank, and having the electrodeposition tank control device control the tank based on these calculated optimal operating conditions, the target dry film thickness can be obtained with high accuracy.

[0011] Factors that can be adjusted during the operation of the electrodeposition tank include the current value of the voltage applied to the vehicle body, the liquid temperature of the electrodeposition paint, and the voltage application time. Factors that can be adjusted for the composition of the electrodeposition paint include acid concentration (MEQ), NV (non-volatile content), ash content, paint conductivity, and Coulomb efficiency.

[0012] In the invention described in claim 1 above, The operating condition calculation means monitors the error between the measured wet film thickness and the target dry film thickness, and the model building means automatically updates the trained model when the error between the measured wet film thickness and the target dry film thickness exceeds an acceptable range. 。

[0013] Therefore,The learned model is automatically updated when the error between the measured wet film thickness and the target dry film thickness exceeds the acceptable range, i.e., when it is anticipated that the dry film thickness of the electrodeposited coating formed on the vehicle body surface will be either insufficient or excessive. In this case, the learned model can be updated and the target dry film thickness changed while the above error has been exceeding the acceptable range for very little time, thus preventing the error from deviating significantly from the acceptable range before the target value is changed. As a result, the dry film thickness, which fluctuates according to the wet film thickness, can be set to the desired thickness, improving the paint quality of the vehicle body.

[0014] Claim 2 The invention described in claim 1 comprises a storage means for storing a learned model that stores the measured value of the wet film thickness, the adjustment factors during operation of the electrodeposition tank, and the composition condition factors of the electrodeposition paint as input information and outputs a target value of the dry film thickness, and an optimization calculation means for inputting the target value of the dry film thickness to the learned model and performing an optimization calculation to calculate the optimal operating conditions that minimize the error with the target value of the dry film thickness and output them from the learned model.

[0015] Claim 2 In the invention described above, the optimization calculation means calculates the optimal operating conditions that minimize the error with the target value of the dry film thickness and outputs them from the trained model, thereby obtaining the optimal operating conditions. Therefore, if the electrodeposition tank control device controls the electrodeposition tank based on the calculated optimal operating conditions, the target dry film thickness can be obtained with high accuracy.

[0016] Claim 3 The invention described in the claim 1 In this configuration, the film thickness measuring means measures the wet film thickness of the electrodeposited coating formed on the outer panel of the vehicle body, and the learned model is used to calculate the optimal operating conditions for achieving a target value for the dry film thickness of the electrodeposited coating formed on the inner panel of the vehicle body.

[0017] Claim 3In the invention described in , the film thickness measuring means measures the wet film thickness of an electrodeposition coating film formed on the outer panel of a vehicle body, so measurement can be performed more easily and quickly than when targeting the wet film thickness of the inner panel. Furthermore, since there is no need to extend the film thickness measuring means over a long distance and insert it into the vehicle body, collision between the film thickness measuring means and the vehicle body can be prevented. In addition, the trained model is used to calculate optimal operating conditions for achieving the target dry film thickness of an electrodeposition coating film formed on the inner panel of the vehicle body (i.e., a region where electrodeposition paint is difficult to flow into and uncoated areas are likely to occur). Therefore, the electrodeposition coating film can be reliably formed over the entire surface of the vehicle body including the inner panel.

[0018] Claim 4 The invention described in 1 according to Claim , the gist thereof is that the operating condition calculation means calculates a set value of voltage applied to an electrode as the optimal operating condition, and outputs the calculated set value of electrode voltage to the electrodeposition tank control device.

[0019] Claim 4 In the invention described in , even if insufficient film thickness or excessive film thickness occurs in the electrodeposition coating film due to unintended condition fluctuations, the operating condition calculation means calculates the set value of voltage applied to the electrode as the optimal operating condition of the electrodeposition tank for achieving the target dry film thickness of the electrodeposition coating film. As a result, the optimal operating conditions are automatically updated. Therefore, the operating condition calculation means outputs the calculated electrode voltage set value to the electrodeposition tank control device, and the electrodeposition tank control device controls the electrodeposition tank, whereby the target value of the dry film thickness can always be achieved.

[0020] Claim 5 The invention described in 4 according to Claim , the gist thereof is that the operating condition calculation means performs optimization calculation for calculating the electrode voltage set value with quality conditions of the electrodeposition coating film as constraint conditions.

[0021] Claim 5In the invention described above, the operating condition calculation means calculates the electrode voltage setting value so as to satisfy the quality conditions (constraints) of the electrodeposited coating film, thereby improving the quality of the electrodeposited coating film while achieving the target value of the dry film thickness.

[0022] Furthermore, constraints include electrode voltage settings that prevent gas pinholes from forming within the electrodeposited coating, and electrode voltage settings that ensure good coverage of the electrodeposited paint without leaving any uncoated areas.

[0023] Claim 6 The invention described is Electrodeposition coating equipment comprising: an electrodeposition tank for storing electrodeposition paint; a drying oven for drying a vehicle body that has been transported out of the electrodeposition tank; a transport means for transporting the vehicle body; and an electrodeposition tank control device for setting the operating conditions of the electrodeposition tank and controlling the electrodeposition tank, wherein electrodeposition coating is performed on a vehicle body immersed in the electrodeposition paint, and further comprising: a film thickness measuring means disposed between the electrodeposition tank and the drying oven for measuring the wet film thickness of the electrodeposition coating film formed on the surface of the vehicle body for each vehicle body; and an operating condition calculation means for calculating optimal operating conditions to achieve a target value for the dry film thickness of the electrodeposition coating film based on the measured wet film thickness, and outputting the calculated optimal operating conditions to the electrodeposition tank control device, wherein the transport means comprises a plurality of hanger rails for transporting the vehicle body while suspending it and immersing it in the electrodeposition tank, and a drop lifter for transferring the vehicle body, which has been lifted out of the electrodeposition tank, from the hanger rails to a trolley, and the film thickness measuring means is disposed at the location where the drop lifter is installed. This is the gist of it.

[0024] Generally, after a vehicle is removed from an electrodeposition tank, any excess electrodeposited paint adhering to the surface is washed off in a water-washing area, and then baked dry in a drying oven. However, it is not possible to place a film thickness measuring device for measuring the wet film thickness of the electrodeposited coating within a water-washing area with a shower or inside a drying oven that reaches high temperatures. Therefore, the claim 6 In the invention described above, the film thickness measuring means is located at the drop lifter installation site, which is separate from the washing area and the drying oven. This will produce the effects described in paragraph 0007, The film thickness measurement can be performed reliably and quickly using the film thickness measurement means.

[0025] Reference 1 The invention relates to an electrodeposition coating method for performing electrodeposition coating on a vehicle body immersed in electrodeposition paint, using an electrodeposition coating apparatus comprising an electrodeposition tank for storing electrodeposition paint and an electrodeposition tank control device for setting operating conditions for the electrodeposition tank and controlling the electrodeposition tank, the method comprising: a film thickness measurement step of measuring the wet film thickness of the electrodeposition coating film formed on the surface of the vehicle body for each vehicle body; an operating condition calculation step of calculating optimal operating conditions to achieve a target value for the dry film thickness of the electrodeposition coating film based on the measured wet film thickness; and an operating condition output step of outputting the calculated optimal operating conditions to the electrodeposition tank control device.

[0026] Reference to the first point aboveIn this invention, even if the electrodeposited coating film becomes under-filmed or over-filmed due to unintended condition fluctuations, the wet film thickness is measured in the film thickness measurement step, and the optimal operating conditions for the electrodeposited tank to achieve the target dry film thickness of the electrodeposited coating are calculated in the operating condition calculation step based on the wet film thickness measurement. As a result, the optimal operating conditions are automatically updated, and in the operating condition output step, the calculated optimal operating conditions are output to the electrodeposited tank control device. The electrodeposited tank control device then controls the electrodeposited tank, thereby constantly optimizing the dry film thickness of the electrodeposited coating and improving the paint quality of the vehicle body. Moreover, since it is not necessary to set the optimal operating conditions to have an excessively high safety margin to account for unintended condition fluctuations, the loss of electrodeposited paint caused by the optimal operating conditions not being reviewed for a long period of time can be reduced.

[0027] Second reference The gist of the invention is a trained model characterized by accumulating measured values ​​of the wet film thickness of the electrodeposited coating film formed on the surface of a vehicle body, adjustment factors during operation of the electrodeposited tank in the electrodeposited coating equipment, and composition condition factors of the electrodeposited paint as input information to a neural network, and causing the computer to function so that it outputs a target value for the dry film thickness of the electrodeposited coating film from the neural network.

[0028] Reference to the second point above In this invention, by having the computer function to accumulate wet film thickness measurements, adjustment factors during electrodeposition tank operation, and compositional condition factors of the electrodeposition coating in a neural network, the accuracy of the target value of the dry film thickness output from the neural network is improved. Therefore, by using the trained model, it becomes possible to optimize the film thickness of the electrodeposition coating.

[0029] A neural network is a model of the network formed by connecting neurons in the human brain's nervous system with synapses. Representative neural networks include convolutional neural networks (CNNs) and recurrent neural networks (RNNs).

[0030] Third reference The gist of the invention is a method for calculating optimal operating conditions for electrodeposition using a trained model, which involves using a trained model that stores measured values ​​of the wet film thickness of an electrodeposition coating film formed on the surface of a vehicle body, adjustment factors for the operation of the electrodeposition tank in the electrodeposition coating equipment, and composition condition factors of the electrodeposition paint as input information to a neural network, and causing the computer to function so that it outputs a target value for the dry film thickness of the electrodeposition coating film from the neural network, and inputting the target value for the dry film thickness to the trained model to calculate the optimal operating conditions for the electrodeposition tank to achieve the target value for the dry film thickness.

[0031] Reference to the third point above In this invention, by having the computer function to accumulate wet film thickness measurements, adjustment factors during electrodeposition tank operation, and compositional condition factors of the electrodeposition coating in a neural network, the accuracy of the target value of the dry film thickness output from the neural network is improved. Therefore, by inputting the target value of the dry film thickness into a trained model and having it calculate the optimal operating conditions for the electrodeposition tank, it becomes possible to optimize the film thickness of the electrodeposition coating. [Effects of the Invention]

[0032] As detailed above, claims 1 to 11 According to the invention described above, by detecting insufficient or excessive film thickness of the electrodeposited coating due to unintended fluctuations in conditions and automatically updating the optimal operating conditions, the loss of electrodeposited coating can be reduced. [Brief explanation of the drawing]

[0033] [Figure 1] A schematic cross-sectional view showing the electrodeposition coating equipment in this embodiment. [Figure 2] A schematic plan view showing the electrodeposition coating equipment. [Figure 3] A schematic cross-sectional view showing the arrangement of the wet film thickness sensor. [Figure 4] A schematic diagram illustrating a neural network. [Figure 5]An explanatory diagram showing the mathematical model (equations) of the objective function and constraints. [Figure 6] An explanatory diagram showing an optimization calculation method using a neural network. [Modes for carrying out the invention]

[0034] Hereinafter, one embodiment embodying the present invention will be described in detail with reference to the drawings.

[0035] As shown in Figure 1, the electrodeposition coating equipment 10 of this embodiment is equipment for performing electrodeposition coating on a vehicle body W1 immersed in electrodeposition paint P1. The electrodeposition coating equipment 10 includes an electrodeposition tank 11 for storing the electrodeposition paint P1. The vehicle body W1 is transported in the electrodeposition tank 11 while immersed in the electrodeposition paint P1. The electrodeposition tank 11 is composed of an upper part 12 that constitutes the ceiling of the electrodeposition tank 11, a lower part 13 that constitutes the floor of the electrodeposition tank 11, and two side walls 14. The electrodeposition tank 11 also has an entrance 15 for bringing the vehicle body W1 into the electrodeposition tank 11 and an exit 16 for bringing the vehicle body W1 out of the electrodeposition tank 11. The electrodeposition paint P1 of this embodiment is, for example, a paint that uses a cationic electrolytic resin as the main component of the vehicle.

[0036] Furthermore, the electrodeposition coating equipment 10 is equipped with a conveyor 21 (conveying means) that transports the vehicle body W1 in the transport direction (to the right in Figure 1). The conveyor 21 lowers the vehicle body W1 and transports it into the electrodeposition tank 11 through the entrance 15, and then raises the vehicle body W1 and transports it out of the electrodeposition tank 11 through the exit 16. The conveyor 21 is equipped with a rail 22 that extends in the transport direction and a plurality of hanger rails 23 provided on the rail 22 that suspend the vehicle body W1 and transport it while immersing it in the electrodeposition tank 11. In addition, a plurality of electrodes 31 are arranged inside the electrodeposition tank 11. Each electrode 31 is strip-shaped and is arranged at intervals along the transport direction of the vehicle body W1.

[0037] As shown in Figures 2 and 3, the electrodeposition coating equipment 10 further includes a washing area (not shown), a drop lifter 41, a preheating area 42, and a drying oven 43. The washing area is an area with a shower that washes away excess electrodeposition paint P1 adhering to the surface of the car body W1 lifted from the electrodeposition tank 11. The drop lifter 41 is for transferring the car body W1 that has passed through the washing area from the hanger rail 23 to the bogie 45 on the conveyor 44. The preheating area 42 is an area for evaporating the solvent component of the electrodeposition paint P1 by applying hot air to the car body W1 that has been transferred to the bogie 45. The drying oven 43 then applies hot air to the car body W1 that has passed through the preheating area 42 to bake and dry the electrodeposition paint P1, thereby forming an electrodeposition coating film.

[0038] Furthermore, a wet film thickness sensor 51, which is a means for measuring film thickness, is positioned between the electrodeposition tank 11 and the drying oven 43 (more precisely, between the electrodeposition tank 11 and the preheating area 42). Specifically, the wet film thickness sensor 51 is positioned (fixed) so as to protrude laterally from the installation location of the drop lifter 41 (in this embodiment, the side wall of the drop lifter 41). The wet film thickness sensor 51 is for measuring the wet film thickness (film thickness of the electrodeposition coating before drying) of the electrodeposition coating formed on the surface of the vehicle body W1 for each vehicle body W1. The wet film thickness sensor 51 is also designed to measure the wet film thickness of the electrodeposition coating formed on the outer panel portion of the vehicle body W1 (for example, the outer panel portion of the door). In this embodiment, the wet film thickness sensor 51 is a non-contact type sensor that measures the wet film thickness by irradiating the vehicle body W1 with a laser, and is positioned so that the distance between the tip surface of the wet film thickness sensor 51 and the surface of the vehicle body W1 is, for example, 100 mm or more and 130 mm or less. The wet film thickness sensor 51 is connected to the personal computer 60 via a cable (not shown). In this embodiment, a laser-type wet film thickness sensor 51 is used, but other types of sensors may be used as long as they are non-contact.

[0039] Next, the electrical configuration of the electrodeposition coating equipment 10 will be described.

[0040] As shown in Figure 1, the electrodeposition coating equipment 10 is equipped with a personal computer 60, which in turn is equipped with a control device 61 for comprehensively controlling the entire equipment. The control device 61 consists of a CPU 62, ROM 63, RAM 64, input / output circuits, etc. The CPU 62 is electrically connected to the keyboard 65 and the display 66 of the personal computer 60. Furthermore, the CPU 62 is electrically connected to the conveyors 21, 44, each electrode 31, the drop lifter 41, and the wet film thickness sensor 51, and controls them with various drive signals.

[0041] Next, the electrodeposition coating method using the electrodeposition coating equipment 10 will be explained.

[0042] First, the CPU 62 outputs a drive signal to the conveyor 21, causing the car body W1 (hanger rail 23) to be continuously fed into the electrodeposition tank 11 via the input 15 and continuously discharged from the electrodeposition tank 11 via the output 16. When the car body W1 that has been fed into the electrodeposition tank 11 is immersed in the electrodeposition paint P1, current is applied to the car body W1 from each electrode 31 located in the electrodeposition tank 11, and electrodeposition coating is performed on the car body W1. As a result, an electrodeposition coating film is formed on the surface of the car body W1.

[0043] After the electrodeposition coating is completed, the car body W1 is lifted out of the electrodeposition tank 11 and moved into the washing area. In the washing area, any excess electrodeposition paint P1 adhering to the surface of the car body W1 is washed away by a shower. Furthermore, after passing through the washing area, the car body W1 is transferred from the hanger rail 23 to the bogie 45 by the drop lifter 41. Subsequently, when the car body W1 is moved into the preheating area 42, the solvent in the electrodeposition coating evaporates, and when it is moved into the drying oven 43, the electrodeposition coating is baked and dried.

[0044] Incidentally, since the conditions (bath conditions) of the electrodeposition coating P1 change moment by moment, if electrodeposition coating is continued, the dry film thickness of the electrodeposition coating (film thickness of the electrodeposition coating after drying) may fluctuate. For this reason, the CPU 62 performs processing to achieve the target value of the dry film thickness of the electrodeposition coating P1.

[0045] First, the CPU 62 performs a film thickness measurement step before the vehicle body W1 is transported to the preheating area 42 and the drying oven 43. In the film thickness measurement step, a drive signal is output to the wet film thickness sensor 51, and the wet film thickness of the electrodeposited coating formed on the surface of the vehicle body W1 is measured (inline) for each vehicle body W1. Here, the wet film thickness is measured at only one representative control measurement point (in this embodiment, a control measurement point set on the outer door panel) out of several control measurement points (approximately 30 to 40) that can be set on the surface of the vehicle body W1. The wet film thickness measurement is performed within a few seconds while the vehicle body W1 is stopped, for example, after being transferred to the drop lifter 41. The data related to the wet film thickness (wet film thickness data) is transmitted to the personal computer 60 (CPU 62) via a cable and stored in the RAM 64 for each vehicle body W1. By repeating this film thickness measurement step, the wet film thickness data is converted into big data.

[0046] In the subsequent operating conditions calculation step, the CPU 62 calculates the optimal operating conditions to achieve the target value of the dry film thickness of the electrodeposited coating based on the measured wet film thickness. In other words, the CPU 62 functions as an "operating conditions calculation means". Specifically, first, the CPU 62 constructs a trained model. In other words, the CPU 62 functions as a "model construction means". The trained model is used to calculate the optimal operating conditions to achieve the target value of the dry film thickness of the electrodeposited coating formed on the inner panel of the vehicle body W1 (in this embodiment, the inner panel of the door). More specifically, the trained model stores input information X1 to X10 for the neural network 71 shown in Figure 4, and causes the control device 61, which is a computer, to function so that the neural network 71 outputs the target value of the dry film thickness (Y = inner panel film thickness).

[0047] Of the input information X1 to X10, input information X1 to X4 are information that serves as adjustment factors during the operation of the electrodeposition tank 11. Input information X1 is the voltage application time, i.e., the takt (takt time), for one vehicle body W1. Input information X2 is the voltage value applied to the electrodes 31 located in the first half (left half in Figure 1) in the transport direction (rightward in Figure 1) (first half electrode voltage), and input information X3 is the voltage value applied to the electrodes 31 located in the second half (right half in Figure 1) in the transport direction (second half electrode voltage). Input information X4 is the liquid temperature of the electrodeposition paint P1.

[0048] Furthermore, input information X5 to X10 can be set as needed. In this embodiment, input information X5 is set to the measured wet film thickness (wet film thickness data), input information X6 to the acid concentration (MEQ), input information X7 to NV (non-volatile content), input information X8 to the ash content, input information X9 to the paint conductivity, and input information X10 to the Coulomb efficiency. Of the input information X5 to X10, input information X6 to X10 are information that serves as a composition condition factor for the electrodeposited paint P1.

[0049] The CPU 62 stores the constructed, trained model in the RAM 64. In other words, the RAM 64 functions as a "memory."

[0050] Then, as shown in Figure 6, the CPU 62 inputs a target value for the dry film thickness to a trained model (neural network 71) stored in RAM 64 and performs an optimization calculation to calculate the optimal operating conditions that minimize the error with the target dry film thickness and output them from the trained model. In other words, the CPU 62 functions as an "optimization calculation means". Specifically, the CPU 62 calculates the set value of the voltage applied to the electrode 31 (electrode voltage set value) as the optimal operating conditions. The CPU 62 also performs an optimization calculation to calculate the electrode voltage set value with the quality conditions of the electrodeposited coating film as constraints. In this embodiment, the constraints are that no gas pins (gas pinholes) occur in the electrodeposited coating film, the electrodeposited paint P1 has good coverage and no uncoated areas are left uncoated, and an electrode voltage set value that satisfies these conditions is calculated.

[0051] As shown in Figure 4, the output from the neural network 71 is a black-box function (NN regression model). In this case, we would like to calculate the optimal operating conditions (electrode voltage setting values) using a prediction model (trained model), but since the shape of the black-box function is unknown, we cannot obtain the optimal operating conditions by manipulating the regression equation as in linear regression.

[0052] Therefore, in this embodiment, the black-box function is formulated as a mathematical optimization problem to find a solution that minimizes a predetermined objective function under the constraints described above. As shown in Figure 5, the objective function mathematical model (equation) indicates that the error between the measured value of the wet film thickness (y) and the target value of the dry film thickness (y') is minimized.

[0053] As a result, the neural network 71 is given a target value for the dry film thickness (X = inner plate film thickness (target value)), and it is possible to input this value and have the neural network 71 output the electrode voltage setting value that minimizes the error with the target value, thereby obtaining an electrode voltage setting value that satisfies the constraints. At this time, information regarding the error is transmitted from the output side to the input side, in other words, from right to left in Figure 6. Furthermore, the electrode voltage setting value is simulated (calculated) while the cycle time and the liquid temperature of the electrodeposition coating P1 are fixed to their current values.

[0054] Subsequently, in the operating condition output step, the CPU 62 outputs (feeds back) the optimal operating conditions (electrode voltage setting values) calculated using the trained model (neural network 71) to the control device 61. In other words, the control device 61 also functions as an "electrodeposition tank control device" that sets the operating conditions of the electrodeposition tank 11 and controls the electrodeposition tank 11.

[0055] Then, the settings of the electrodeposition coating equipment 10 are changed based on the calculated optimal operating conditions, and electrodeposition coating is performed in this state. More specifically, the CPU 62 of the control device 61 applies voltage from the electrode 31 to the vehicle body W1 based on the calculated electrode voltage setting value. This adjusts the dry film thickness of the electrodeposition coating formed on the surface of the vehicle body W1 to approach the target value. It is generally known that the dry film thickness tends to increase as the voltage (electrode voltage setting value) increases. Therefore, it can be said that there is a positive correlation between voltage and dry film thickness.

[0056] Furthermore, the CPU 62 monitors the error between the measured wet film thickness and the target dry film thickness, based on the wet film thickness measured by the wet film thickness sensor 51 when the electrode voltage setting value has been changed (updated). If the error between the measured wet film thickness and the target dry film thickness exceeds the acceptable range, the CPU 62 automatically updates the trained model (neural network 71).

[0057] For example, if the error between the measured wet film thickness and the target dry film thickness is higher than the upper limit of the acceptable range, the CPU 62 updates the trained model and causes the neural network 71 to calculate a lower electrode voltage setting than the current value. This prevents the electrodeposited coating (dry film thickness) from becoming excessively thick. On the other hand, if the error between the measured wet film thickness and the target dry film thickness is lower than the lower limit of the acceptable range, the CPU 62 updates the trained model and causes the neural network 71 to calculate a higher electrode voltage setting than the current value. This prevents the electrodeposited coating (dry film thickness) from becoming insufficiently thick.

[0058] Therefore, according to this embodiment, the following effects can be obtained.

[0059] (1) In this embodiment, even if the electrodeposited coating film becomes under-filmed or over-filmed due to unintended fluctuations in conditions, the wet film thickness sensor 51 measures the wet film thickness, and the CPU 62 calculates the optimal operating conditions (electrode voltage setting) for the electrodeposition tank 11 to achieve the target dry film thickness of the electrodeposited coating film based on the wet film thickness measurement. As a result, the voltage applied to the electrode 31 (electrode voltage setting) is automatically updated. Therefore, by controlling the electrodeposition tank 11 based on the calculated electrode voltage setting, the CPU 62 can constantly optimize the dry film thickness of the electrodeposited coating film, improving the painting quality of the vehicle body W1 and minimizing the loss of electrodeposited paint P1. Thus, the amount of electrodeposited paint P1 used can be minimized.

[0060] (2) In this embodiment, the wet film thickness sensor 51 measures the wet film thickness of the electrodeposited coating formed on the outer panel portion of the vehicle body W1 (specifically, the outer panel portion of the door). The CPU 62 also estimates (calculates) the electrode voltage setting value to achieve the target value of the dry film thickness of the electrodeposited coating formed on the inner panel portion of the vehicle body W1 based on the measured wet film thickness. In other words, by making an estimation regarding the inner panel portion, which is a part where the electrodeposited paint P1 does not easily wrap around and is prone to being left uncoated, the electrodeposited coating can be reliably formed on the entire surface of the vehicle body W1, including the inner panel portion. Furthermore, the wet film thickness sensor 51 measures the wet film thickness at only one representative control measurement point (specifically, the control measurement point set on the outer panel portion of the door) out of a plurality of control measurement points that may be set on the surface of the vehicle body W1. Therefore, the cost and time required for measurement can be reduced compared to when the wet film thickness sensor 51 measures the wet film thickness at each control measurement point.

[0061] (3) Generally, after the vehicle body W1 is lifted from the electrodeposition tank 11, excess electrodeposition paint P1 adhering to the surface is washed off in the water washing area, the solvent in the electrodeposition coating is evaporated in the preheating area 42, and the electrodeposition coating is baked dry in the drying oven 43. However, the wet film thickness sensor 51, which measures the wet film thickness of the electrodeposition coating, cannot be placed in the water washing area with a shower, or in the preheating area 42 and drying oven 43, which become hot. Therefore, in this embodiment, the wet film thickness sensor 51 is placed at the location where the drop lifter 41 is installed, which is separate from the water washing area, preheating area 42, and drying oven 43. This makes it possible to reliably measure with the wet film thickness sensor 51 without putting a load on the wet film thickness sensor 51.

[0062] (4) Conventionally, after the vehicle body W1 was baked and dried in a drying oven, the thickness of the electrodeposited coating film formed on the surface of the vehicle body W1 (dry thickness) was measured using a contact-type electromagnetic film thickness gauge. Furthermore, conventionally, the film thickness was measured only on representative vehicle bodies W1 selected from a large number of vehicles (for example, about 10,000 vehicles per week). However, if defects such as insufficient or excessive film thickness occur in the electrodeposited coating film of vehicle bodies W1 whose film thickness has not been measured, there is a problem that defective products will end up on the market.

[0063] On the other hand, in this embodiment, before the baking and drying stage, the wet film thickness of the electrodeposited coating formed on the surface of the vehicle body W1 is measured (in-line) for each vehicle body W1 (100% inspection). Therefore, vehicle bodies W1 with defects such as insufficient or excessive film thickness in the electrodeposited coating can be reliably detected.

[0064] (5) Conventionally, the thickness of the electrodeposited coating (dry thickness) was measured after the vehicle body W1 was baked and dried for a predetermined time, so it took a considerable amount of time to determine the quality of the electrodeposited coating (whether or not defects such as insufficient or excessive thickness occurred).

[0065] In contrast, in this embodiment, the film thickness (wet film thickness) of the electrodeposited coating is measured immediately after electrodeposition coating is performed on the vehicle body W1, allowing for rapid determination of the quality of the electrodeposited coating in near real-time. Therefore, if a defect in the quality of the electrodeposited coating is detected, the electrode voltage setting can be immediately reflected (updated), and the film thickness (dry film thickness) of the electrodeposited coating can be corrected in a short time, thereby minimizing the number of defective vehicle bodies W1. As a result, the loss of vehicle bodies W1 being scrapped can be reduced.

[0066] The above embodiment may be modified as follows.

[0067] • In the above embodiment, the wet film thickness sensor 51 was positioned (fixed) at the installation location of the drop lifter 41 (in this embodiment, the side wall of the drop lifter 41), but it may be positioned at other locations. For example, a movable robot may be positioned at the installation location of the drop lifter 41, and the wet film thickness sensor 51 may be mounted on the robot and the robot may be driven to change the measurement position. In addition, multiple wet film thickness sensors 51 may be provided. For example, a pair of wet film thickness sensors 51 may be provided so as to sandwich the vehicle body W1.

[0068] In the above embodiment, the wet film thickness sensor 51 measured the wet film thickness of the electrodeposited coating formed on the surface of the vehicle body W1 at only one representative control measurement point out of a plurality of control measurement points (approximately 30 to 40) that could be set on the surface of the vehicle body W1. However, the wet film thickness sensor 51 may measure the wet film thickness at two or more control measurement points selected from each control measurement point, or it may measure the wet film thickness at all control measurement points.

[0069] In the above embodiment, the adjustment factors for the operation of the electrodeposition tank 11, which are stored as input information in the trained model (neural network 71), were the cycle time (input information X1), which is the time for which the voltage applied to the vehicle body W1 is applied; the voltage values ​​of the voltage applied to the vehicle body W1, namely the front electrode voltage and the rear electrode voltage (input information X2, X3); and the liquid temperature of the electrodeposition paint P1 (input information X4). However, the adjustment factor may be at least one selected from the application time (input information X1), voltage values ​​(input information X2, X3), and liquid temperature (input information X4). Generally, it is known that the dry film thickness tends to increase as the application time increases. Therefore, it can be said that there is a positive correlation between the application time and the dry film thickness. Also, it is known that the dry film thickness tends to increase as the liquid temperature increases. Therefore, it can be said that there is a positive correlation between the liquid temperature and the dry film thickness. In the above embodiment, the voltage value was divided into two voltage values ​​(front electrode voltage and rear electrode voltage), but it may also be combined into a single voltage value.

[0070] In the above embodiment, the compositional condition factors of the electrodeposited coating P1 stored as input information in the trained model (neural network 71) were acid concentration (MEQ) (input information X6), NV (non-volatile content) (input information X7), ash content (input information X8), coating conductivity (input information X9), and Coulomb efficiency (input information X10). However, the compositional condition factors may be at least one selected from acid concentration (MEQ), NV (non-volatile content), ash content, coating conductivity, and Coulomb efficiency.

[0071] Generally, as the acid concentration decreases and the amount of acid decreases, the repulsion between paint particles contained in the electrodeposited paint P1 weakens, making them more prone to aggregation. As a result, compared to when the acid concentration is high, the Coulomb efficiency (amount deposited per Coulomb) increases, making it easier to deposit the electrodeposited paint P1 on the surface of the vehicle body W1 with less electricity. Therefore, it is known that under the same electrical conditions, the dry film thickness tends to increase. Thus, it can be said that there is a negative correlation between acid concentration and dry film thickness. Furthermore, it is known that as NV increases, the amount of resin component remaining in the electrodeposited paint film during drying increases, so the dry film thickness tends to increase. Thus, it can be said that there is a positive correlation between NV and dry film thickness. In addition, it is known that as the ash content increases, the pigment content of the electrodeposited paint P1 increases, increasing film resistance and improving adhesion, so the dry film thickness tends to increase. Thus, it can be said that there is also a positive correlation between ash content and dry film thickness. Furthermore, it is known that as the conductivity of the paint increases, the electrodeposited coating film is more easily formed, and therefore the dry film thickness tends to increase. Therefore, it can be said that there is a positive correlation between the conductivity of the paint and the dry film thickness. Also, as the Coulomb efficiency increases, it becomes easier to deposit the electrodeposited paint P1 on the surface of the vehicle body W1 with a small amount of electricity, as described above, and therefore the dry film thickness tends to increase. Therefore, it can be said that there is a positive correlation between the Coulomb efficiency and the dry film thickness.

[0072] Next, in addition to the technical ideas described in the claims, the technical ideas that can be grasped by the embodiments described above are listed below.

[0073] (1) The electrodeposition coating equipment according to any one of claims 2 to 5, characterized in that the adjustment factor during operation of the electrodeposition tank is at least one selected from the voltage value of the voltage applied to the vehicle body, the liquid temperature of the electrodeposition paint, and the voltage application time.

[0074] (2) An electrodeposition coating apparatus according to any one of claims 2 to 5, characterized in that the composition condition factor of the electrodeposition coating is at least one selected from acid concentration (MEQ), NV (non-volatile content), ash content, coating conductivity, and Coulomb efficiency.

[0075] (3) The electrodeposition coating apparatus according to claim 7, characterized in that the constraint is an electrode voltage setting value such that gas pinholes do not occur in the electrodeposition coating film and the electrodeposition paint has good coverage and no uncoated areas occur. [Explanation of Symbols]

[0076] 10… Electrodeposition coating equipment 11...electrodeposition bath 21... Conveyor as a means of transport 23... Hanger rail 31...Electrode 41… Drop lifter 43…Drying oven 45... Dolly 51...Wet film thickness sensor as a means of measuring film thickness 61... Electrodeposition tank control device and control device as a computer 62...CPU as a means for calculating operating conditions, a means for building a model, and a means for optimizing calculations 64...RAM as a means of storage 71...Neural Networks P1…electrodeposition paint W1...vehicle body X1~X10... Input Information

Claims

1. An electrodeposition coating apparatus comprising an electrodeposition tank for storing electrodeposition paint, a drying oven for drying the vehicle body after it has been removed from the electrodeposition tank, and an electrodeposition tank control device for setting the operating conditions of the electrodeposition tank and controlling the electrodeposition tank, wherein electrodeposition coating is performed on the vehicle body immersed in the electrodeposition paint, A film thickness measuring means is disposed between the electrodeposition tank and the drying oven and measures the wet film thickness of the electrodeposition coating formed on the surface of the vehicle body for each vehicle body, The system includes a model building means for constructing a trained model that stores the measured wet film thickness, the adjustment factors during operation of the electrodeposition tank, and the composition condition factors of the electrodeposition paint as input information, and outputs a target value for the dry film thickness of the electrodeposition coating, and an operating condition calculation means for outputting the optimal operating conditions for achieving the target dry film thickness calculated using the trained model to the electrodeposition tank control device. Equipped with, The aforementioned operating condition calculation means monitors the error between the measured value of the wet film thickness and the target value of the dry film thickness. The model building means automatically updates the trained model when the error between the measured value of the wet film thickness and the target value of the dry film thickness exceeds an acceptable range. An electrodeposition coating apparatus characterized by the following features.

2. The aforementioned means for calculating operating conditions is: A storage means for storing a learned model that stores the measured value of the wet film thickness, the adjustment factors during the operation of the electrodeposition tank, and the composition condition factors of the electrodeposition paint as input information, and outputs a target value for the dry film thickness. An optimization calculation means performs an optimization calculation on the trained model by inputting a target value for the dry film thickness, calculating the optimal operating conditions that minimize the error with the target value for the dry film thickness, and outputting them from the trained model. The electrodeposition coating equipment according to claim 1, characterized by comprising the following features.

3. The aforementioned film thickness measuring means measures the wet film thickness of the electrodeposited coating film formed on the outer panel of the vehicle body. The trained model is used to calculate the optimal operating conditions for achieving the target value of the dry film thickness of the electrodeposited coating formed on the inner panel of the vehicle body. The electrodeposition coating equipment according to feature 1.

4. The electrodeposition coating equipment according to claim 1, characterized in that the operating condition calculation means calculates a set value for the voltage applied to the electrodes as the optimal operating condition and outputs the calculated electrode voltage set value to the electrodeposition tank control device.

5. The electrodeposition coating equipment according to claim 4, characterized in that the operating condition calculation means performs an optimization calculation to calculate the electrode voltage setting value with the quality conditions of the electrodeposition coating film as constraints.

6. An electrodeposition coating apparatus comprising: an electrodeposition tank for storing electrodeposition paint; a drying oven for drying the vehicle body after it has been transported out of the electrodeposition tank; a transport means for transporting the vehicle body; and an electrodeposition tank control device for setting the operating conditions of the electrodeposition tank and controlling the electrodeposition tank, wherein electrodeposition coating is performed on the vehicle body immersed in the electrodeposition paint, A film thickness measuring means is disposed between the electrodeposition tank and the drying oven and measures the wet film thickness of the electrodeposition coating formed on the surface of the vehicle body for each vehicle body, An operating condition calculation means calculates the optimal operating conditions for achieving the target value of the dry film thickness of the electrodeposited coating based on the measured wet film thickness, and outputs the calculated optimal operating conditions to the electrodeposited tank control device. Equipped with, The transport means comprises a plurality of hanger rails for transporting the vehicle body while suspending it and immersing it in the electrodeposition tank, and a drop lifter for transferring the vehicle body, which has been lifted out of the electrodeposition tank, from the hanger rails to a trolley. The aforementioned film thickness measuring means is located at the installation location of the drop lifter. An electrodeposition coating apparatus characterized by the following features.

7. The means for calculating the operating conditions is: The system includes a model building means for constructing a trained model that stores the measured value of the wet film thickness, the adjustment factors during the operation of the electrodeposition tank, and the composition condition factors of the electrodeposition paint as input information, and outputs a target value for the dry film thickness. The optimal operating conditions calculated using the trained model are output to the electrodeposition tank control device. The electrodeposition coating equipment according to feature 6.

8. The operating condition calculation means monitors the error between the measured value of the wet film thickness and the target value of the dry film thickness, The model building means automatically updates the trained model when the error between the measured value of the wet film thickness and the target value of the dry film thickness exceeds an acceptable range. The electrodeposition coating equipment according to feature 7.

9. The means for calculating the operating conditions is: A storage means for storing a learned model that stores the measured value of the wet film thickness, the adjustment factors during the operation of the electrodeposition tank, and the composition condition factors of the electrodeposition paint as input information, and outputs a target value for the dry film thickness. An optimization calculation means performs an optimization calculation on the trained model by inputting a target value for the dry film thickness, calculating the optimal operating conditions that minimize the error with the target value for the dry film thickness, and outputting them from the trained model. The electrodeposition coating equipment according to claim 6, characterized by comprising the following features.

10. The film thickness measuring means measures the wet film thickness of the electrodeposited coating film formed on the outer panel of the vehicle body, The trained model is used to calculate the optimal operating conditions for achieving the target value of the dry film thickness of the electrodeposited coating formed on the inner panel of the vehicle body. The electrodeposition coating equipment according to feature 7.

11. The electrodeposition coating equipment according to claim 6, characterized in that the operating condition calculation means calculates a set value for the voltage applied to the electrodes as the optimal operating condition and outputs the calculated electrode voltage set value to the electrodeposition tank control device.

Images