Ultrasonic vibration internal cooling lubrication grinding system based on deep reinforcement learning

Abstract

The invention discloses an ultrasonic vibration internal cooling lubrication grinding system based on deep reinforcement learning. The ultrasonic vibration internal cooling lubrication grinding system comprises a PC end deep reinforcement learning system, a power distribution collection control system, an ultrasonic vibration platform system, a sensor measurement system, an internal cooling lubrication grinding wheel system and a cooling lubrication liquid supply system. According to the system, the ultrasonic vibration platform internally provided with the embedded micro heat pipe array and the piezoelectric ceramic gradient polarization array is combined with the inner-cooling and inner-lubricating asymmetric fractal runner grinding wheel for cooperative machining; and meanwhile, multi-physics field perception and AI deep reinforcement learning are assisted to achieve effective cooperation of vibration, cooling and lubrication in the machining process of the ultrasonic vibration platform and the inner cooling and lubricating grinding wheel, and efficient, high-quality and low-consumption machining of materials difficult to machine is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of precision grinding machining, and particularly to an ultrasonic vibration internal cooling and lubrication grinding system based on deep reinforcement learning. Background Technique

[0002] Ultrasonic vibration-assisted grinding can effectively reduce the grinding force and improve the grinding surface quality. At present, it has been widely used in difficult-to-machine materials such as aerospace. The implementation methods of ultrasonic vibration-assisted grinding mainly include workpiece ultrasonic vibration and tool ultrasonic vibration. However, most of the existing ultrasonic grinding processes mainly use small-diameter grinding wheel tools for vibration, which not only limits the application of large-size grinding wheels, but also restricts the improvement of grinding speed and processing efficiency, and cannot meet the basic requirements of ultra-high-speed grinding technology for complex shaped components. Compared with the tool ultrasonic vibration grinding method, the workpiece ultrasonic vibration grinding method has the characteristics of high energy utilization rate and simple structure, and is very suitable for the situation with large grinding force and high material removal rate, so it has broad application space and good development prospects. However, there are also problems in workpiece ultrasonic vibration grinding, such as uneven vibration of the ultrasonic vibration platform, deformation of the platform caused by heat during processing, and deviation of processing accuracy.

[0003] During the grinding process of difficult-to-machine materials, a large amount of frictional heat and grinding force will be generated. Therefore, effective cooling and lubrication must be ensured. For traditional pouring cooling and lubrication, the cooling and lubrication liquid cannot effectively penetrate the grinding air barrier layer, resulting in poor cooling and lubrication effects in the grinding area, and at the same time causing a large amount of waste of cooling and lubrication liquid, increasing the processing cost. The internal cooling and lubrication grinding wheel can effectively cool and lubricate the inside of the grinding area, effectively improving the processing surface quality. However, most of the current internal cooling and lubrication grinding wheels use a single vertical channel or a symmetric vertical channel for the internal flow channel. When the grinding wheel rotates at a high speed, the coolant lubricant is thrown out at a high speed under the centrifugal force, resulting in uneven cooling and lubrication in the internal grinding area, and at the same time, a large amount of cooling and lubrication liquid is wasted.

[0004] The adjustment and coordination of process parameters during the grinding process are crucial. At present, the process parameters of most grinding systems are adjusted based on experience, which is time-consuming and laborious, and the accuracy is not high. There are also some processing systems that use feedback control to automatically adjust the processing parameters, but the feedback algorithm cannot meet the complex multi-objective optimization control, and the accuracy is limited. At present, the existing ultrasonic vibration grinding systems operate independently in terms of vibration, cooling, and lubrication without synergy. At the same time, ordinary grinding wheels are mostly used for ultrasonic vibration grinding, and they do not cooperate with internal cooling and lubrication grinding wheels for processing, so they cannot further effectively improve the processing efficiency and processing quality.

[0005] In summary, a grinding processing system based on deep reinforcement learning, which combines an ultrasonic vibration platform with an internally cooled and lubricated grinding wheel, is proposed. The existing problems of the ultrasonic vibration platform and the internally cooled and lubricated grinding wheel are improved. At the same time, multi-physical field perception and AI deep reinforcement learning are used to effectively coordinate vibration, cooling, and lubrication during the processing of the ultrasonic vibration platform and the internally cooled and lubricated grinding wheel, so as to achieve efficient, high-quality, and low-consumption processing of difficult-to-machine materials. Summary of the Invention

[0006] To solve the above problems, the present invention proposes a grinding processing system based on deep reinforcement learning, which combines an ultrasonic vibration platform with an internally cooled and lubricated grinding wheel. Through the collaborative processing of the ultrasonic vibration platform with embedded micro heat pipe arrays and piezoelectric ceramic gradient polarization arrays and the internally cooled and internally lubricated asymmetric fractal flow channel grinding wheel, and with the assistance of multi-physical field perception and AI deep reinforcement learning, the effective coordination of vibration, cooling, and lubrication during the processing of the ultrasonic vibration platform and the internally cooled and lubricated grinding wheel is achieved, so as to achieve efficient, high-quality, and low-consumption processing of difficult-to-machine materials.

[0007] The technical solution of the present invention is: a deep reinforcement learning-based ultrasonic vibration internal cooling and lubrication grinding system, which is characterized in that it includes a PC-side deep reinforcement learning system, a power distribution acquisition and control system, an ultrasonic vibration platform system, a sensor measurement system, an internally cooled and lubricated grinding wheel system, and a cooling and lubricating fluid supply system.

[0008] The power distribution acquisition and control system is built-in with a power supply unit, a control unit, a data acquisition unit, and an ultrasonic generation unit, and is responsible for simultaneously supplying power and signal transmission to the PC-side deep reinforcement learning system, the ultrasonic vibration platform system, the sensor measurement system, the internally cooled and lubricated grinding wheel system, and the cooling and lubricating fluid supply system.

[0009] The ultrasonic vibration platform system includes an ultrasonic vibration platform and a workpiece. The ultrasonic vibration platform includes a transducer, a horn, a vibration conversion module, a support base, a vibration platform, an embedded micro heat pipe array, and a piezoelectric ceramic gradient polarization array. The support base, the transducer, the horn, the vibration conversion module, and the vibration platform are sequentially connected to each other. The workpiece is fixed on the vibration platform. The embedded micro heat pipe array is parallelly embedded in the vibration platform. A certain amount of special low-melting-point alloy is filled inside the micro heat pipe to prevent the platform from deforming by melting and absorbing heat. The piezoelectric ceramic gradient polarization array is installed at the bottom of the vibration platform and is distributed in a rectangular array. The polarization intensity of the piezoelectric ceramic gradient polarization array in the edge area of the vibration platform is 20%-35% higher than that in the central area, which solves the problem of insufficient vibration at the edge. The transducer is connected to the power distribution acquisition and control system, and the power supply unit and the ultrasonic generation unit of the power distribution acquisition and control system supply power and transmit ultrasonic waves to the ultrasonic vibration platform system respectively.

[0010] The sensor measurement system includes a dynamometer, a CCD camera, and a semiconductor laser. The dynamometer measures the grinding force during the grinding process, and the signal collected by the dynamometer is transmitted to the data acquisition unit of the power distribution and acquisition control system. The CCD camera and the semiconductor laser are used in combination to measure the surface roughness during the grinding process, and the measured signal is also transmitted to the data acquisition unit of the power distribution and acquisition control system.

[0011] The internal cooling and lubrication grinding wheel system includes an internal cooling and lubrication grinding wheel and a machine tool spindle. The internal cooling and lubrication grinding wheel is installed on the machine tool spindle.

[0012] The internal cooling and lubrication grinding wheel includes a grinding wheel shank, an electro-hydraulic slip ring, a grinding wheel base body, and MEMS thin film thermocouples. The electro-hydraulic slip ring is installed on the grinding wheel shank, the grinding wheel shank is connected to the grinding wheel base body, and the MEMS thin film thermocouples are evenly installed inside the grinding wheel base body near the grinding surface. The wiring terminals of the MEMS thin film thermocouples are connected to the electro-hydraulic slip ring. The grinding wheel base body is internally provided with an asymmetric tree-shaped fractal flow channel, which is a biomimetic tree-shaped structure with at least three levels of branches, and the branch angles are asymmetrically designed to utilize centrifugal force to transport fluid. The branch angle range is (θ = 15° - 75°), which can utilize the centrifugal transport of fluid during the high-speed rotation of the grinding wheel to achieve effective active lubrication in the grinding area, and at the same time control and save the cooling and lubricating fluid. The liquid inlet end of the electro-hydraulic slip ring is connected to the cooling and lubricating fluid supply system through a gas-liquid transport pipeline. The cooling and lubricating fluid supply system pumps the cooling and lubricating fluid through the internal flow channel of the electro-hydraulic slip ring and flows through the tree-shaped flow channel of the grinding wheel base body to reach the grinding area. The power distribution and acquisition control system is connected to the incoming line end of the electro-hydraulic slip ring to supply power to the electro-hydraulic slip ring, and the current is transmitted to the MEMS thin film thermocouples. The MEMS thin film thermocouples measure the grinding surface temperature, and the measured temperature signal is then transmitted to the data acquisition unit of the power distribution and acquisition control system.

[0013] The cooling and lubricating fluid supply system includes two parts: coolant supply and lubricant supply. The two parts can be controlled separately and cooperatively, and each is equipped with a flow control valve. The coolant supply and lubricant supply are powered by the built-in power supply unit of the power distribution and acquisition control system to achieve the pumping of the coolant and lubricant. The flow control valve is powered by the built-in power supply unit of the power distribution and acquisition control system and is controlled for flow by the built-in control unit of the power distribution and acquisition control system.

[0014] The PC - side deep reinforcement learning system is built - in with a DRL decision - maker. The real - time grinding force, surface roughness, and temperature measured by each sensor are used as target signals. They are uniformly transmitted to the DRL decision - maker via the data acquisition unit of the power distribution and acquisition control system. The internal reward function is: R = α·(1 - ΔT / T_max)+β·Ra_min / Ra + γ·F_c / F_c0 (where ΔT is the temperature rise, Ra is the surface roughness, F_c is the cutting force, and α + β + γ = 1). The system conducts self - deep reinforcement learning by setting target weights, searches for the optimal input strategy, then makes decision information, and transmits the decision information to the control unit built - in the power distribution and acquisition control system. The control unit then adjusts process parameters such as the spindle speed, coolant flow rate, lubricant flow rate, and ultrasonic vibration frequency of the input, making the processing target results (grinding force, surface roughness of the workpiece, grinding temperature) close to the optimal solution.

[0015] The operation process of the present invention is as follows: Start the power distribution and acquisition control system to supply power and transmit signals to the PC - side deep reinforcement learning system, ultrasonic vibration platform system, sensor measurement system, internal - cooling and lubricating grinding wheel system, and cooling and lubricating fluid supply system simultaneously. The cooling and lubricating fluid supply system pumps cooling and lubricating fluid to the internal - cooling and lubricating grinding wheel system. The internal - cooling and lubricating grinding wheel system and the ultrasonic vibration platform system act simultaneously to start processing. The sensor measurement system and the internal - cooling and lubricating grinding wheel system respectively measure the grinding force, workpiece surface roughness, and grinding temperature in real - time. The measured signals are transmitted to the control unit built - in the power distribution and acquisition control system, and then transmitted to the DRL decision - maker built - in the PC - side deep reinforcement learning system via the control unit. The DRL decision - maker conducts self - deep reinforcement learning through the set reward function and target weights, searches for the optimal input strategy, and then makes decision information. The decision information is transmitted to the control unit built - in the power distribution and acquisition control system. The control unit then adjusts process parameters such as the spindle speed, coolant flow rate, lubricant flow rate, and ultrasonic vibration frequency of the input, making the processing target (grinding force, workpiece surface roughness, grinding temperature) results close to the optimal solution.

[0016] Compared with the prior art, the technical effect of the present invention is: A grinding processing system combining an ultrasonic vibration platform based on deep reinforcement learning and an internal - cooling and lubricating grinding wheel is proposed. Through the collaborative processing of an ultrasonic vibration platform with an embedded micro - heat pipe array and a piezoelectric ceramic gradient polarization array and an internal - cooling and internal - lubricating asymmetric fractal flow channel grinding wheel, and by assisting multi - physical - field perception and AI deep reinforcement learning to achieve the effective coordination of vibration, cooling, and lubrication during the processing of the ultrasonic vibration platform and the internal - cooling and lubricating grinding wheel, high - efficiency, high - quality, and low - consumption processing of difficult - to - machine materials is realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 is a schematic structural diagram according to an embodiment of the present invention;

[0018] Figure 2It is a system hardware composition diagram according to an embodiment of the present invention;

[0019] Figure 3 It is a structural diagram of an ultrasonic vibration platform according to an embodiment of the present invention;

[0020] Figure 4 It is a cross-sectional view of an internally cooled and lubricated grinding wheel according to an embodiment of the present invention;

[0021] Figure 5 It is a working flow chart of a deep reinforcement learning (DRL) decision maker according to an embodiment of the present invention; Detailed implementation manners

[0022] For a clearer understanding of the technical features, objectives, and effects of the present invention, the present invention will be further described below in conjunction with the accompanying drawings and embodiments, but the content of the present invention is not limited to the described scope.

[0023] As Figure 1 and 2 shown, an ultrasonic vibration internal cooling and lubrication grinding system based on deep reinforcement learning is characterized in that it includes a PC-side deep reinforcement learning system (1), a power distribution acquisition control system (2), an ultrasonic vibration platform system (3), a sensor measurement system (4), an internal cooling and lubrication grinding wheel system (5), and a cooling and lubricating fluid supply system (6);

[0024] The power distribution acquisition control system (2) is internally provided with a power supply unit, a control unit, a data acquisition unit, and an ultrasonic generation unit, and is responsible for simultaneously supplying power and signal transmission to the PC-side deep reinforcement learning system (1), the ultrasonic vibration platform system (3), the sensor measurement system (4), the internal cooling and lubrication grinding wheel system (5), and the cooling and lubricating fluid supply system (6).

[0025] As Figure 2 and 3As shown in the figure, the ultrasonic vibration platform system (3) includes an ultrasonic vibration platform (31) and a workpiece (32). The ultrasonic vibration platform (31) includes a support base (310), a transducer (311), a horn (312), a vibration conversion module (313), a vibration platform (314), an embedded micro heat pipe array (315), and a piezoelectric ceramic gradient polarization array (316). The support base (310), transducer (311), horn (312), vibration conversion module (313), and vibration platform (314) are connected to each other in sequence. The workpiece (32) is fixed on the vibration platform (314). The embedded micro heat pipe array (315) is parallelly embedded and installed in the vibration platform (314). A certain amount of special low melting point alloy is filled inside the micro heat pipes to prevent the vibration platform (314) from deforming by melting and absorbing heat. The piezoelectric ceramic gradient polarization array (316) is installed at the bottom of the vibration platform (314) and is distributed in a rectangular array. The polarization intensity of the piezoelectric ceramic gradient polarization array (316) in the edge area of the vibration platform (314) is 20%-35% higher than that in the central area to solve the problem of insufficient vibration at the edge. The transducer (311) is connected to the power distribution and acquisition control system (2), and the power supply unit and the ultrasonic generation unit of the power distribution and acquisition control system (2) supply power to and transmit ultrasonic waves to the ultrasonic vibration platform system (3) respectively.

[0026] As Figure 2 shown, the sensor measurement system (4) includes a dynamometer (41), a CCD camera (42), and a semiconductor laser (43). The grinding force during the grinding process is measured by the dynamometer (41), and the signal collected by the dynamometer (41) is transmitted to the data acquisition unit of the power distribution and acquisition control system (2). The CCD camera (42) and the semiconductor laser (43) are used in combination to measure the surface roughness during the grinding process, and the measured signal is also transmitted to the data acquisition unit of the power distribution and acquisition control system (2).

[0027] As Figure 2 and 4 shown, the internal cooling and lubrication grinding wheel system (5) includes an internal cooling and lubrication grinding wheel (51) and a machine tool spindle (52). The internal cooling and lubrication grinding wheel (51) is installed on the machine tool spindle (52);

[0028] The internally cooled and lubricated grinding wheel (51) includes a grinding wheel shank (510), an electro-hydraulic slip ring (511), a grinding wheel base body (512), and a MEMS thin-film thermocouple (513). The electro-hydraulic slip ring (511) is installed on the grinding wheel shank (510). The grinding wheel shank (510) is connected to the grinding wheel base body (512). The MEMS thin-film thermocouples (513) are evenly installed inside the grinding wheel base body (512) near the grinding surface. The wiring terminals of the MEMS thin-film thermocouples (513) are connected to the electro-hydraulic slip ring (511). The grinding wheel base body (512) is internally provided with an asymmetric tree-shaped fractal flow channel (512-1). This fractal flow channel is a biomimetic tree-like structure with at least three levels of branches, and the branch angles are designed asymmetrically to utilize the centrifugal force to transport the fluid. The branch angle range is (θ = 15° - 75°), which can utilize the centrifugal transport of the fluid during the high-speed rotation of the grinding wheel to achieve effective active lubrication in the grinding area, while controlling and saving the cooling and lubricating fluid. A grinding grain layer (512-2) is plated on the grinding wheel base body (512). The liquid inlet end of the electro-hydraulic slip ring (511) is connected to the cooling and lubricating fluid supply system through a gas-liquid transport pipeline. The cooling and lubricating fluid supply system (6) pumps the cooling and lubricating fluid through the internal flow channel of the electro-hydraulic slip ring (511) and flows through the asymmetric tree-shaped fractal flow channel (512-1) of the grinding wheel base body (512) and then reaches the grinding area. The power distribution and acquisition control system (2) is connected to the incoming line end of the electro-hydraulic slip ring (511) to supply power to the electro-hydraulic slip ring (511). The current is transmitted to the MEMS thin-film thermocouple (513). The MEMS thin-film thermocouple (513) measures the grinding surface temperature, and the measured temperature signal is then transmitted to the data acquisition unit of the power distribution and acquisition control system (2).

[0029] As Figure 2 shown, the cooling and lubricating fluid supply system (6) includes two parts: a coolant supply and a lubricant supply. These two parts can be controlled separately and work together, and each is separately equipped with a flow control valve. The coolant supply and the lubricant supply are powered by the built-in power supply unit of the power distribution and acquisition control system (2) to achieve the pumping of the coolant and the lubricant. The flow control valve is powered by the built-in power supply unit of the power distribution and acquisition control system (2), and at the same time, the flow is controlled by the built-in control unit of the power distribution and acquisition control system (2).

[0030] As Figure 2 and 5As shown, the PC-side deep reinforcement learning system (1) is built-in with a DRL decision maker. The real-time grinding force, surface roughness, and temperature measured by each sensor are used as target signals and are uniformly transmitted to the DRL decision maker via the data acquisition unit of the power distribution and acquisition control system (2). The internal reward function is: R = α·(1 - ΔT / T_max) + β·Ra_min / Ra + γ·F_c / F_c0 (where ΔT is the temperature rise, Ra is the surface roughness, F_c is the cutting force, and α + β + γ = 1); the system performs self-deep reinforcement learning by setting target weights, searches for the optimal input strategy, and then makes decision information. The decision information is transmitted to the control unit built-in the power distribution and acquisition control system (2), and then the control unit adjusts process parameters such as the spindle speed, coolant flow rate, lubricant flow rate, and ultrasonic vibration frequency input, so that the processing target result is close to the optimal solution.

[0031] The operation process of the present invention is as follows: Start the power distribution and acquisition control system (2) to supply power and transmit signals to the PC-side deep reinforcement learning system (1), the ultrasonic vibration platform system (3), the sensor measurement system (4), the internal cooling and lubricating grinding wheel system (5), and the cooling and lubricating fluid supply system (6) simultaneously. The cooling and lubricating fluid supply system (6) pumps the cooling and lubricating fluid to the internal cooling and lubricating grinding wheel system (5). The internal cooling and lubricating grinding wheel system (6) and the ultrasonic vibration platform system (3) act simultaneously to start processing. The sensor measurement system (4) and the internal cooling and lubricating grinding wheel system (5) respectively measure the grinding force, workpiece surface roughness, and grinding temperature in real time. The measured signals are transmitted to the control unit built-in the power distribution and acquisition control system (2), and then transmitted to the DRL decision maker built-in the PC-side deep reinforcement learning system (1) via the control unit. The DRL decision maker performs self-deep reinforcement learning through the set reward function and target weights, searches for the optimal input strategy, and then makes decision information. The decision information is transmitted to the control unit built-in the power distribution and acquisition control system (2), and then the control unit adjusts process parameters such as the spindle speed, coolant flow rate, lubricant flow rate, and ultrasonic vibration frequency input, so that the processing target (grinding force, workpiece surface roughness, grinding temperature) result is close to the optimal solution.

[0032] Although the embodiments of the present invention have been shown and described, it is obvious to those skilled in the art that the present invention is not limited to the details of the above specific embodiments, and without departing from the spirit or basic characteristics of the present invention, the present invention can be implemented in other specific forms. Therefore, in any regard, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be encompassed by the present invention. Any reference signs in the claims should not be regarded as limiting the claims involved.

Claims

1. An ultrasonic vibration internal cooling and lubrication grinding system based on deep reinforcement learning, characterized in that: It includes a PC - side deep reinforcement learning system (1), a power distribution acquisition and control system (2), an ultrasonic vibration platform system (3), a sensor measurement system (4), an internal cooling and lubricating grinding wheel system (5), and a coolant and lubricant supply system (6); The PC - side deep reinforcement learning system (1) is powered by the power distribution acquisition and control system (2), and at the same time, signal transmission is carried out between them. The power distribution acquisition and control system (2) supplies power and conducts signal transmission to the ultrasonic vibration platform system (3), the sensor measurement system (4), the internal cooling and lubricating grinding wheel system (5), and the coolant and lubricant supply system (6) simultaneously. The internal cooling and lubricating grinding wheel system (5) is connected to the coolant and lubricant supply system (6) through a gas - liquid transmission pipeline; The PC - side deep reinforcement learning system (1) is built - in with a DRL decision - maker; The ultrasonic vibration platform system (3) is built - in with an embedded micro - heat pipe array (315) and a piezoelectric ceramic gradient polarization array (316); The internal cooling and lubricating grinding wheel system (5) includes an internal cooling and lubricating grinding wheel (51). The internal cooling and lubricating grinding wheel (51) includes a grinding wheel matrix (512) and a MEMS thin - film thermocouple (513). The grinding wheel matrix (512) is built - in with an asymmetric tree - shaped fractal flow channel (512 - 1).

2. The ultrasonic vibration internal cooling and lubrication grinding system based on deep reinforcement learning according to claim 1, wherein: Built - in with a DRL decision - maker, using a deep reinforcement learning algorithm, its reward function is: R = α·(1 - ΔT / T_max)+β·Ra_min / Ra+γ·F_c / F_c0 (where ΔT is the temperature rise, Ra is the surface roughness, F_c is the cutting force, and α + β + γ = 1).

3. The ultrasonic vibration internal cooling and lubrication grinding system based on deep reinforcement learning according to claim 1, wherein: The embedded micro - heat pipe array (315) is parallelly embedded and installed in the vibration platform (314) of the ultrasonic vibration platform system (3). A quantitative special low - melting - point alloy is filled inside the micro - heat pipes. The piezoelectric ceramic gradient polarization array (316) is installed at the bottom of the vibration platform (314) of the ultrasonic vibration platform system (3) and is distributed in a rectangular array. The polarization intensity of the piezoelectric ceramic gradient polarization array (316) in the edge area of the vibration platform (314) is 20% - 35% higher than that in the central area.

4. The ultrasonic vibration internal cooling and lubrication grinding system based on deep reinforcement learning according to claim 1, characterized in that: The asymmetric tree - shaped fractal flow channel (512 - 1) is at least a three - level - branched bionic tree - like structure, and the branch angles are asymmetrically designed to utilize centrifugal force to transport fluids. The branch angle range is (θ = 15° - 75°). The MEMS thin - film thermocouples (513) are evenly distributed and installed inside the grinding wheel matrix (512) near the grinding surface.

5. The ultrasonic vibration internal cooling and lubrication grinding system based on deep reinforcement learning according to claim 1, characterized in that: The coolant and lubricant supply system (6) is built - in with two parts: coolant supply and lubricant supply. The two parts can be controlled separately and work together, and each is separately equipped with a flow control valve.

Images