Intelligent temperature control automobile part ultrasonic cleaning machine and ultrasonic cleaning method
By using temperature control devices and sensor networks in ultrasonic cleaning machines, the temperature of the cleaning liquid is adjusted, the cavitation shielding problem is solved, the cleaning range is expanded and the efficiency is improved, and the high cleaning requirements of automotive parts are met.
Patent Information
- Application Number
- CN202411248179.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-06
- Publication Date
- 2024-12-13
AI Technical Summary
The existing ultrasonic cleaning technology has cavitation shielding phenomenon, resulting in a small cleaning range and low efficiency. It is impossible to expand the cavitation range by increasing the sound power, and it is difficult to meet the high cleanliness requirements of automotive parts.
Using an intelligent temperature-controlled ultrasonic cleaning machine for automotive parts, the cooling and heating devices are set up in the cleaning tank, combined with multiple sound pressure and temperature sensors, and the temperature of the cleaning liquid is adjusted using PID control and deep neural network algorithm to form a temperature gradient and weaken the air. The cavitation energy is evenly distributed.
The cleaning area range of ultrasonic cleaning is effectively expanded, the cleaning uniformity and efficiency are improved, the cavitation energy of the cleaning liquid is uniform, and the high cleanliness standards are met.
Smart Images
Figure CN119114517A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to an intelligent temperature-controlled automobile parts ultrasonic cleaning machine and an ultrasonic cleaning method, belonging to the technical field of ultrasonic cleaning. Background Art
[0002] The quality management of automotive core components and related accessories is crucial to their performance and lifespan. These components have complex production processes and diverse materials. They are often exposed to harsh environments during driving and are susceptible to oil, dust, rust, corrosion and other pollution. Therefore, fine cleaning is required to remove pollutants to ensure product quality and environmental protection of the process, which has also caused many cleaning problems. Traditional cleaning technologies, such as immersion, steam cleaning, brushing, pressure spraying, etc., have low cleaning efficiency, are difficult to clean every detail of automotive parts, and are prone to human pollution, making it difficult to meet the requirements of automotive industry cleanliness testing standards such as VDA-19 and ISO-16232.
[0003] At present, ultrasonic cleaning based on cavitation effect is becoming the mainstream cleaning technology for automotive parts. In terms of meeting cleanliness standards, ultrasonic cleaning has unique advantages: high cleaning depth (micron-level particle size), no damage, fast speed, no dead angle, can adapt to assembly line automated production, green and environmentally friendly, and can meet the cleanliness requirements of various new energy vehicle parts. Ultrasonic cleaning uses an ultrasonic transducer to convert the sound energy of a power ultrasonic frequency source (ultrasonic generator) into mechanical vibration, radiating sound waves into the cleaning tank. At this time, the liquid pressure in the cleaning tank will be disturbed. When the sound pressure is lower than a certain value, the tiny gas nuclei in the cleaning liquid will expand violently, collapse sharply, and collapse. This phenomenon is called ultrasonic cavitation. The series of physical processes such as high temperature, high pressure, micro jet, shock wave, etc. generated by bubbles in the dynamic process are collectively referred to as cavitation effect. The cavitation effect plays a vital role in ultrasonic cleaning.
[0004] Ultrasonic cleaning technology belongs to physical cleaning with bubble cavitation effect as the core mechanism. However, these bubbles are not evenly distributed in the cleaning tank, and the intensity and spatial distribution of cavitation energy are crucial to improving the cleaning effect. In the near-field area near the ultrasonic transducer, due to the large sound power, more cavitation bubbles are generated. The cavitation bubbles will absorb the sound wave energy for its own radial oscillation, and reflect and scatter the sound waves, so that the sound waves are confined to this area, hindering their propagation to the far field. This phenomenon is called cavitation shielding. Cavitation shielding makes the cavitation energy mainly concentrated in the near-field area near the transducer. When the input sound power of the cleaning liquid is low, increasing the input sound power can expand the range of the cavitation occurrence area to a certain extent. However, since cavitation bubbles will shield the sound propagation, the greater the input sound power, the more cavitation bubbles will be generated near the transducer, and the stronger the shielding effect of these cavitation bubbles on the sound propagation. When the input sound power reaches a certain level, the cavitation bubbles near the ultrasonic transducer will almost completely prevent the sound waves from continuing to propagate forward. At this time, cavitation occurs only in the ultrasonic near-field area near the transducer, which means that the ultrasonic cavitation range cannot be expanded by continuously increasing the ultrasonic input power, and if the input sound power of the cleaning liquid is reduced, the cavitation range will be reduced. This makes ultrasonic cleaning technology always have many problems such as a small cleaning range, fewer parts to be cleaned at the same time, and low cleaning efficiency. Summary of the invention
[0005] In order to overcome the above-mentioned defects of the prior art, the present invention provides an intelligent temperature-controlled ultrasonic cleaning machine and an ultrasonic cleaning method for automobile parts, which can weaken the cavitation shielding phenomenon in the ultrasonic near-field area, expand the cavitation occurrence area, make the cavitation energy distribution uniform, and improve the cleaning efficiency and effect.
[0006] The technical solution for achieving the above-mentioned purpose of the present invention is: an intelligent temperature-controlled ultrasonic cleaning machine for automobile parts, comprising a cleaning tank, an ultrasonic transducer is provided on the tank wall of the cleaning tank, a refrigeration device is provided in the ultrasonic near-field area of the cleaning tank, and a heating device is provided in the ultrasonic far-field area of the cleaning tank.
[0007] Preferably, a plurality of sound pressure sensors (or hydrophones) are provided in the cleaning tank along the ultrasonic propagation direction from the ultrasonic near-field region to the ultrasonic far-field region, and the plurality of sound pressure sensors are evenly distributed in the cleaning tank along the ultrasonic propagation direction, and the signal outputs of the plurality of sound pressure sensors are connected to a data acquisition card, and the sound pressure signal output of the data acquisition card is connected to a main control computer, and the temperature control signal output of the main control computer is connected to a temperature controller (or temperature control board), and the operation of the refrigeration device and the heating device is controlled by the temperature controller.
[0008] Preferably, a plurality of temperature sensors are provided in the cleaning tank along the ultrasonic propagation direction from the ultrasonic near-field region to the ultrasonic far-field region, the plurality of temperature sensors are evenly distributed in the cleaning tank along the ultrasonic propagation direction, the temperature signal outputs of the plurality of temperature sensors are connected to the data acquisition card, and the temperature signal outputs of the data acquisition card are connected to the main control computer.
[0009] Preferably, the sound pressure sensor and the temperature sensor are both arranged in a linear distribution along the ultrasonic wave propagation direction in the cleaning tank.
[0010] Preferably, the number of the sound pressure sensors and the temperature sensors is the same, and the positions along the ultrasonic wave propagation direction are arranged in a one-to-one correspondence.
[0011] Preferably, the ultrasonic transducer is arranged at the center of the bottom wall of the cleaning tank, the refrigeration device is arranged on the bottom side wall of the cleaning tank, and the heating device is arranged at the upper center of the cleaning tank. It can be fixedly connected to the upper side wall of the cleaning tank by a bracket, or fixed by hanging. The sound pressure sensor and the temperature sensor are distributed in the cleaning tank along the vertical center line of the cleaning tank.
[0012] The refrigeration device can adopt semiconductor refrigeration plates, and the number of the semiconductor refrigeration plates can be multiple, which are evenly distributed along the circumference of the bottom of the cleaning tank and fixed on the outer wall of the cleaning tank. The heating device can adopt a heating rod, which is vertically inserted into the cleaning tank from top to bottom.
[0013] An intelligent temperature-controlled ultrasonic cleaning method for automobile parts adopts any one of the intelligent temperature-controlled ultrasonic cleaning machines for automobile parts disclosed in the present invention, and heats and / or cools the cleaning liquid in different areas of the cleaning tank, so that the cleaning liquid forms a temperature gradient with gradually increasing temperature in the ultrasonic propagation direction from the ultrasonic near-field area to the ultrasonic far-field area, thereby increasing the cavitation threshold of the cleaning liquid in the ultrasonic near-field area, weakening the cavitation shielding phenomenon in the ultrasonic near-field area, reducing the cavitation threshold of the cleaning liquid in the ultrasonic far-field area, expanding the cavitation occurrence area, and making the cavitation energy in the cleaning tank evenly distributed.
[0014] Preferably, a PID control algorithm is used to control the heating and / or cooling of the cleaning liquid in different areas of the cleaning tank, and the control parameters of the PID control algorithm are calculated based on multiple sound pressure signals collected in the ultrasonic propagation direction from the ultrasonic near-field area to the ultrasonic far-field area.
[0015] Furthermore, the overall cavitation energy uniformity in the cleaning fluid is calculated based on multiple sound pressure signals collected in the ultrasonic propagation direction from the ultrasonic near-field region to the ultrasonic far-field region; based on the cavitation energy uniformity in the cleaning fluid, a preset cavitation energy uniformity threshold and the difference between the two, a deep neural network algorithm is used to calculate the control parameters of the PID control algorithm.
[0016] Preferably, the cavitation energy uniformity in the cleaning liquid, the preset cavitation energy uniformity threshold and the difference between the two are input into the deep neural network to obtain the control parameters of the PID control algorithm, and thereby control the heating and / or cooling of the cleaning liquid in different areas of the cleaning tank, and calculate the difference between the new cavitation energy uniformity after the cleaning liquid is heated and / or cooled and the preset cavitation energy uniformity. If the difference is within the set allowable range, the calculation is terminated; if the difference is not within the set allowable range, the deep neural network automatically adjusts the weights of each layer according to the difference, and continues the calculation in a cycle until the difference between the cavitation energy uniformity in the cleaning liquid and the preset cavitation energy uniformity threshold is within the set allowable range.
[0017] The beneficial effects of the present invention are as follows: the present invention cools the cleaning liquid in the ultrasonic near-field area of the cleaning tank and / or heats the cleaning liquid in the ultrasonic far-field area of the cleaning tank, so that the cleaning liquid in the cleaning tank forms a temperature gradient with gradually increasing temperature in the ultrasonic propagation direction from the ultrasonic near-field area to the ultrasonic far-field area, thereby forming a gradual cavitation threshold in the ultrasonic propagation direction, weakening the cavitation shielding phenomenon in the ultrasonic near-field area and expanding the cavitation occurrence range in the ultrasonic far-field area. The cleaning liquid in a large area of the cleaning tank can be cavitated for a long time without human intervention, and the cavitation energy in the cleaning tank is evenly distributed, thereby expanding the cleaning area range of ultrasonic cleaning, improving the cleaning uniformity, cleaning efficiency and effect. The deep neural network combined with the PID control strategy realizes the intelligent control of the heating and cooling of the cleaning liquid, making the temperature regulation (cavitation threshold) of the cleaning liquid more accurate and ensuring the uniformity of the cavitation energy of the cleaning liquid. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] Figure 1 It is a structural schematic diagram of an embodiment of the intelligent temperature-controlled automotive parts ultrasonic cleaning machine of the present invention;
[0019] Figure 2 It is a block diagram of a PID control system of the intelligent temperature-controlled automobile parts ultrasonic cleaning method of the present invention;
[0020] Figure 3 is a flow chart of the present invention for calculating control parameters of a PID control algorithm using a deep neural network;
[0021] Figure 4It is a structural diagram of the deep neural network of the present invention. DETAILED DESCRIPTION
[0022] See also Figure 1 The present invention discloses an intelligent temperature-controlled ultrasonic cleaning machine for automobile parts, comprising a cleaning tank 1, an ultrasonic transducer 2 is provided on the tank wall of the cleaning tank, and is used to radiate ultrasonic waves to the cleaning liquid in the cleaning tank, a refrigeration device 3 is provided in the ultrasonic near-field region of the cleaning tank, and is used to cool the cleaning liquid in the ultrasonic near-field region, and a heating device 4 is provided in the ultrasonic far-field region of the cleaning tank, and is used to heat the cleaning liquid in the ultrasonic far-field region. By cooling the cleaning liquid in the ultrasonic near-field region of the cleaning tank and / or heating the cleaning liquid in the ultrasonic far-field region of the cleaning tank, a temperature gradient with a gradually increasing temperature is formed in the ultrasonic propagation direction from the ultrasonic near-field region to the ultrasonic far-field region of the cleaning tank, thereby forming a gradual cavitation threshold in the ultrasonic propagation direction.
[0023] Preferably, a plurality of sound pressure sensors (or hydrophones) 5 are provided in the cleaning tank along the ultrasonic propagation direction from the ultrasonic near-field region to the ultrasonic far-field region, for collecting sound pressure signals at different positions of the cleaning liquid in the cleaning tank. The plurality of sound pressure sensors are evenly distributed in the cleaning tank along the ultrasonic propagation direction. The signal outputs of the plurality of sound pressure sensors are connected to a data acquisition card (for example, Lingzhi Electronics DAQ122, Anyibo Electronic Technology USB2DaqsA, Zishu Technology DAQ-USB4010, etc.) 6. The sound pressure signal output sampled by the data acquisition card is connected to a main control computer 7 for analysis and processing. The temperature control signal output of the main control computer is connected to a temperature controller (or temperature control board, such as MAX31865 or Gongsheng Instrument GSC-J1440WR4, etc.) 8. The temperature controller controls the circuit connection and disconnection of the refrigeration device and the heating device to adjust the temperature of the cleaning liquid in the cleaning tank.
[0024] Preferably, a plurality of temperature sensors 9 are provided in the cleaning tank along the ultrasonic propagation direction from the ultrasonic near-field region to the ultrasonic far-field region, for collecting temperature signals of the cleaning liquid at different positions in the cleaning tank. The plurality of temperature sensors are evenly distributed in the cleaning tank along the ultrasonic propagation direction. The signal outputs of the plurality of temperature sensors are connected to the data acquisition card, and the temperature signal output sampled by the data acquisition card is connected to the main control computer.
[0025] The sound pressure sensors and the temperature sensors are preferably arranged in a linear distribution along the ultrasonic wave propagation direction in the cleaning tank, and the number of the sound pressure sensors and the temperature sensors are preferably the same, and the positions along the ultrasonic wave propagation direction are arranged one by one so that the positions in the cleaning liquid targeted by the collected sound pressure signals and temperature signals match.
[0026] The ultrasonic transducer is arranged at the center of the bottom wall of the cleaning tank, and is connected to the main control computer through the ultrasonic power supply 10. The refrigeration device is arranged on the bottom side wall of the cleaning tank. The heating device is arranged at the upper center of the cleaning tank, and can be fixedly connected to the upper side wall of the cleaning tank through a bracket, or can be fixed by hanging. The sound pressure sensor and the temperature sensor are distributed and arranged in the cleaning tank along the vertical center line of the cleaning tank. The ultrasonic power supply and the ultrasonic transducer constitute the ultrasonic generating system of the ultrasonic cleaning machine; the refrigeration device, the heating device and the temperature controller constitute the temperature control system of the ultrasonic cleaning machine; the sound pressure sensor, the temperature sensor and the data acquisition card constitute the data monitoring system of the ultrasonic cleaning machine, and each system is uniformly controlled by the main control computer. The refrigeration device can use a semiconductor cooling plate, and the number of the semiconductor cooling plates can be multiple, which are evenly distributed and fixed on the outer wall of the cleaning tank along the circumference of the bottom of the cleaning tank. The heating device can use a heating rod, which is vertically inserted into the cleaning tank from top to bottom.
[0027] The present invention also discloses an intelligent temperature-controlled ultrasonic cleaning method for automobile parts. By using any one of the intelligent temperature-controlled ultrasonic cleaning machines for automobile parts disclosed in the present invention, the cleaning liquid in different areas of the cleaning tank is heated and / or cooled (the temperature of the cleaning liquid in the ultrasonic near-field area is reduced, and the temperature of the cleaning liquid in the ultrasonic far-field area is increased), so that the cleaning liquid forms a temperature gradient with gradually increasing temperature in the ultrasonic propagation direction from the ultrasonic near-field area to the ultrasonic far-field area. On the one hand, the cavitation threshold of the cleaning liquid in the ultrasonic near-field area is increased, and the cavitation shielding phenomenon is weakened; on the other hand, the cavitation threshold of the cleaning liquid in the ultrasonic far-field area is reduced, and the cavitation occurrence area is expanded, and finally the cavitation energy in the cleaning tank is evenly distributed.
[0028] The multiple sound pressure sensors and the multiple temperature sensors are arranged in the cleaning tank in a one-to-one correspondence, and correspond to the multiple data monitoring points in the cleaning tank. For example, the number of the sound pressure sensors and the temperature sensors are four, that is, four monitoring points are evenly set on the vertical center line (vertical center line) of the cleaning tank (when setting other numbers of monitoring points, the principle and calculation method are the same), and each monitoring point is provided with a sound pressure sensor and a temperature sensor. The four monitoring points from the bottom to the top of the cleaning tank are marked as A1, A2, A3, and A4, respectively, wherein A1 and A2 are located in the lower cleaning area (ultrasonic near field area), and A3 and A4 are located in the upper cleaning area (ultrasonic far field area). The control method of the cleaning liquid temperature is: the time domain signal of the sound pressure of the corresponding monitoring point detected by each sound pressure sensor is transmitted to the main control computer, which is Fourier transformed by the main control computer into a frequency domain signal, and then the cavitation energy value at each monitoring point is obtained by integration, which is recorded as W1, W2, W3, and W4 respectively. The average cavitation energy of the cleaning zone (ultrasonic near field area) at the bottom of the cleaning tank is defined as W 12 =(W1+W2) / 2, the average cavitation energy in the upper cleaning area (ultrasonic far field area) of the cleaning tank is W 34 =(W3+W4) / 2, the average value of the overall cavitation energy in the cleaning tank is W ave =(W1+W2+W3+W4) / 4, the cavitation energy variance between the lower cleaning zone and the upper cleaning zone in the cleaning tank is W var =[(W 12 -W ave ) 2 +(W 34 -W ave ) 2 ] / 2, the variance W var As an indicator of the uniformity of cavitation energy in the cleaning tank. var When the cavitation energy uniformity is greater than a certain threshold (the preset cavitation energy uniformity threshold), it is considered that the cavitation energy distribution in the cleaning tank is uneven. At this time, the cavitation energy uniformity definition formula W is used. uni =W 12 -W 34 , to calculate the uniformity of cavitation energy. var If the threshold value exceeds the previously set value, the main control computer will continue to calculate the cavitation energy uniformity W uni , according to W uni value, calls the control algorithm, outputs a control instruction to the temperature controller, controls the working state of the refrigeration device and / or the heating device, changes the temperature of the cleaning fluid in the ultrasonic near field and ultrasonic far field areas, and then changes the temperature gradient of the cleaning fluid along the ultrasonic propagation direction, and finally makes the cavitation energy uniformity in the cleaning fluid reach the expected level.
[0029] In actual application, the main control computer sends instructions to the ultrasonic power supply through the USB interface, and the ultrasonic power supply loads the AC voltage to the ultrasonic transducer according to the instructions. The ultrasonic transducer radiates ultrasonic waves into the cleaning liquid in the cleaning tank under the drive of the voltage. The cleaning liquid in the lower cleaning area of the cleaning tank is cooled by the refrigeration device, and the cleaning liquid in the upper cleaning area of the cleaning tank is heated by the heating device. In this way, the temperature of the cleaning liquid in the ultrasonic near field area of the cleaning tank is relatively low, and the temperature of the cleaning liquid in the ultrasonic far field area is relatively high, so that the cleaning liquid in the cleaning tank forms a temperature gradient with a gradual temperature change. The main control computer can control the switch and power-on time of the refrigeration device and the heating device through the thermostat to control the cooling and heating treatment of the cleaning liquid, thereby changing the temperature distribution of the cleaning liquid. The temperature sensor can monitor the temperature of each monitoring point in the cleaning liquid in real time, and feed back the temperature signal to the main control computer after sampling through the data acquisition card. The data acquisition card can sample the sound pressure analog signal collected by the sound pressure sensor and send the data to the main control computer, which analyzes the digital signal collected by the sound pressure sensor to obtain cavitation energy uniformity information. When the cavitation energy uniformity of the cleaning liquid exceeds the set threshold, the automatic control algorithm in the main control computer will control the circuit switch and power-on time of the refrigeration device and the heating device based on this information, change the temperature distribution in the cleaning liquid, and ensure that the cavitation energy uniformity of the cleaning liquid returns to the preset range.
[0030] The specific control strategies are:
[0031] The refrigeration device and the heating device are controlled by using a deep neural network combined with a PID control method. The control system block diagram is shown in the following figure. Figure 2 shown.
[0032] The cavitation energy uniformity threshold is set to R(s), the cavitation energy uniformity actually measured is Y(s), the error (or difference) E(s) = R(s)-Y(s), and the system disturbance is D(s). p1 , K i1 , K d1 are the proportional, integral and differential coefficients of the heating device, K p2 , K i2 , K d2 They are the proportional, integral and differential coefficients of the refrigeration device respectively. The control strategy consists of two parts: traditional PID control and deep neural network control algorithm:
[0033] Part 1: Traditional PID control. Closed-loop control is performed directly on the refrigeration device and the heating device. Two sets of parameters K p1 , K i1, K d1 and K p2 , K i2 , K d2 This is the online setting method.
[0034] Part 2: Deep neural network algorithm. According to the real-time sound field distribution in the cleaning tank, the parameters of PID control are automatically adjusted to control the refrigeration device and the heating device, so that the output layer neurons of the deep neural network correspond to the two sets of adjustable parameters K of PID control. p1 , K i1 , K d1 and K p2 , K i2 , K d2 , and then continuously train the deep neural network and adjust the weights, so that its stable state finally corresponds to the control parameters of PID under the optimal control law.
[0035] The process of PID control is as follows Figure 3 As shown, a certain heating temperature and cooling temperature are first given to form a temperature gradient in the liquid, and then ultrasonic cavitation treatment is started. During the treatment process, the main control computer calculates the cavitation energy uniformity y(k) in the cleaning liquid in real time based on the signal collected by the sound pressure sensor, and calculates the error e(k) between it and the set threshold r(k). Input y(k), r(k) and e(k) into the deep neural network to obtain K p1 , K i1 , K d1 , K p2 , K i2 and K d2 , obtain the output u(k), and use it as a basis to control the refrigeration device and the heating device to change the cleaning liquid temperature. Calculate the cavitation energy uniformity after the cleaning liquid temperature changes and the new error between it and the set threshold, and calculate whether the new error is within the allowable range. If so, end this temperature adjustment. If not, the deep neural network automatically adjusts the weights of each layer according to the error value, and keeps the above cycle until the error between the cavitation energy uniformity index and the set threshold is lower than the preset value. ( Figure 2 The variables in the frequency domain are expressed in the frequency domain, the frequency domain Y(s) corresponds to the time domain y(k), the frequency domain R(s) corresponds to the time domain r(k), the frequency domain E(s) corresponds to the time domain e(k), and the frequency domain U(s) corresponds to the time domain u(k)).
[0036] See also Figure 4The input layer of the deep neural network has three nodes, which are the cavitation energy uniformity threshold r, the actual measured value of the cavitation energy uniformity index y, and the error e between the two. There are two hidden layers, each with 20 nodes. The output layer has 6 nodes, which correspond to the PID parameters (K p1 , K i1 , K d1 , K p2 , K i2 and K d2 ).
[0037] Unless otherwise specified or one preferred or optional technical means is a further limitation of another technical means, all preferred and optional technical means disclosed in the present invention can be arbitrarily combined to form several different technical solutions.
Claims
1. An intelligent temperature-controlled ultrasonic cleaning machine for automobile parts, comprising a cleaning tank, wherein an ultrasonic transducer is arranged on the wall of the cleaning tank, characterized in that The ultrasonic near-field region of the cleaning tank is provided with a refrigeration device, and the ultrasonic far-field region of the cleaning tank is provided with a heating device.
2. The intelligent temperature-controlled ultrasonic cleaning machine for automobile parts according to claim 1, characterized in that A plurality of sound pressure sensors are arranged in the cleaning tank along the ultrasonic propagation direction from the ultrasonic near-field region to the ultrasonic far-field region. The plurality of sound pressure sensors are evenly distributed in the cleaning tank along the ultrasonic propagation direction. The sound pressure signal outputs of the plurality of sound pressure sensors are connected to a data acquisition card. The sound pressure signal output of the data acquisition card is connected to a main control computer. The temperature control signal output of the main control computer is connected to a temperature controller. The operation of the refrigeration device and the heating device is controlled by the temperature controller.
3. The intelligent temperature-controlled ultrasonic cleaning machine for automobile parts according to claim 2, characterized in that A plurality of temperature sensors are provided in the cleaning tank along the ultrasonic propagation direction from the ultrasonic near-field region to the ultrasonic far-field region. The plurality of temperature sensors are evenly distributed in the cleaning tank along the ultrasonic propagation direction. The temperature signal outputs of the plurality of temperature sensors are connected to the data acquisition card, and the temperature signal output of the data acquisition card is connected to the main control computer.
4. The intelligent temperature-controlled ultrasonic cleaning machine for automobile parts according to claim 3, characterized in that The sound pressure sensor and the temperature sensor are both arranged in a straight line along the ultrasonic wave propagation direction in the cleaning tank.
5. The intelligent temperature-controlled ultrasonic cleaning machine for automobile parts according to claim 4, characterized in that The number of the sound pressure sensors and the temperature sensors is the same, and the positions along the ultrasonic wave propagation direction are arranged in a one-to-one correspondence.
6. The intelligent temperature-controlled ultrasonic cleaning machine for automobile parts according to claim 5, characterized in that The ultrasonic transducer is arranged at the center of the bottom wall of the cleaning tank, the refrigeration device is arranged on the bottom side wall of the cleaning tank, the heating device is arranged at the upper center of the cleaning tank, and the sound pressure sensor and the temperature sensor are distributed in the cleaning tank along the vertical center line of the cleaning tank.
7. Intelligent temperature-controlled ultrasonic cleaning method for automobile parts, characterized in that The intelligent temperature-controlled ultrasonic cleaning machine for automobile parts according to any one of claims 1 to 6 is adopted. By heating and / or cooling the cleaning liquid in different areas of the cleaning tank, a temperature gradient with gradually increasing temperature is formed in the ultrasonic propagation direction from the ultrasonic near-field area to the ultrasonic far-field area of the cleaning liquid, thereby increasing the cavitation threshold of the cleaning liquid in the ultrasonic near-field area, weakening the cavitation shielding phenomenon in the ultrasonic near-field area, reducing the cavitation threshold of the cleaning liquid in the ultrasonic far-field area, expanding the cavitation occurrence area, and making the cavitation energy in the cleaning tank evenly distributed.
8. The ultrasonic cleaning method for intelligent temperature-controlled automobile parts according to claim 7, characterized in that A PID control algorithm is used to control the heating and / or cooling of the cleaning liquid in different areas of the cleaning tank, and the control parameters of the PID control algorithm are calculated based on multiple sound pressure signals collected in the ultrasonic propagation direction from the ultrasonic near-field area to the ultrasonic far-field area.
9. The ultrasonic cleaning method for intelligent temperature-controlled automobile parts according to claim 8, characterized in that The overall cavitation energy uniformity in the cleaning fluid is calculated based on multiple sound pressure signals collected in the ultrasonic propagation direction from the ultrasonic near-field area to the ultrasonic far-field area; based on the cavitation energy uniformity in the cleaning fluid, the preset cavitation energy uniformity threshold and the difference between the two, a deep neural network algorithm is used to calculate the control parameters of the PID control algorithm.
10. The ultrasonic cleaning method for intelligent temperature-controlled automobile parts according to claim 9, characterized in that The cavitation energy uniformity in the cleaning liquid, the preset cavitation energy uniformity threshold and the difference between the two are input into the deep neural network to obtain the control parameters of the PID control algorithm, and thereby control the heating and / or cooling of the cleaning liquid in different areas of the cleaning tank, and calculate the difference between the new cavitation energy uniformity after the cleaning liquid is heated and / or cooled and the preset cavitation energy uniformity. If the difference is within the set allowable range, the calculation is terminated; if the difference is not within the set allowable range, the deep neural network automatically adjusts the weights of each layer according to the difference, and continues the calculation in a cycle until the difference between the cavitation energy uniformity in the cleaning liquid and the preset cavitation energy uniformity threshold is within the set allowable range.