A spinning intelligent manufacturing management device and method based on multi-source data fusion

Abstract

The application relates to a spinning intelligent manufacturing management device and method based on multi-source data fusion, which comprises a central processing unit, a data acquisition unit, a broken yarn identification unit, a roving stop feeding execution unit, an energy consumption monitoring unit and a man-machine interaction interface. The application solves the technical problems of long response time between broken yarn detection and roving stop feeding, causing raw material waste, and conventional multi-source data separately guiding corresponding process control, leading to complex operation and inability to further improve spinning efficiency through the following steps: data acquisition, receiving signals of the data acquisition unit by the broken yarn identification unit and executing broken yarn determination, executing judgment results, energy consumption monitoring, multi-source data fusion of the central processing unit and broken end rate analysis and data output management.

Description

Technical Field

[0001] This invention belongs to the field of intelligent manufacturing technology for spinning, specifically relating to an intelligent manufacturing management device and method for spinning based on multi-source data fusion. Background Technology

[0002] Several technical bottlenecks exist in the spinning production process that affect production efficiency and management level, mainly in the following aspects: First, there is a disconnect between yarn breakage detection and handling. The spinning process is a crucial step in spinning production, and yarn breakage has long plagued spinning companies. Traditional methods rely on manual inspection, requiring workers to frequently move around in high-temperature and high-humidity environments, which is not only labor-intensive but also has a high rate of missed detections. While single-spindle detection systems exist, these systems are typically independent of the roving stop-feed device. There is a delay of tens of seconds between the signal being issued after a yarn breakage is detected and the roving stop-feed is executed. During this delay, roving continues to be fed in, resulting in material waste. Statistics show that in spinning workshops without automatic stop-feed devices, continuous feeding of roving after a single spindle breakage can cause tens of grams of waste yarn per spindle per day. For workshops with tens of thousands of spindles, the annual material waste is considerable.

[0003] Second, the perception of abnormal spinning conditions is limited to a single dimension. Existing single-spindle detection devices mostly use photoelectric or electromagnetic sensors to detect the presence or absence of yarn. The accuracy of such detection methods in judging yarn breakage is affected by factors such as ambient light and accumulated fly waste, resulting in high false alarm and false negative rates. Some publicly available technologies use eddy current sensors to detect the movement of the traveler to indirectly determine yarn breakage. Although this improves the reliability of detection to some extent, it still relies on a threshold judgment of a single physical quantity and lacks the ability to predict yarn breakage risks in advance.

[0004] Third, energy consumption monitoring and equipment status monitoring are isolated from each other. In the spinning process, equipment energy consumption accounts for a significant proportion of spinning costs. In a 100,000-spindle green smart factory, 432 sets of digital equipment and over 500,000 data collection points have been deployed, and a 5G industrial internet platform has been built, forming a relatively good data collection foundation. However, data such as the real-time power of the main equipment, the distribution of broken ends, and the status of individual spindles are currently collected through different channels. There is a lack of a systematic solution to integrate and process this multi-source data within the same device and to implement closed-loop control of production based on the fusion results.

[0005] Fourth, workshop managers cannot grasp the full picture of production anomalies in a timely manner. While the single-spindle detection system can report the location of broken ends, managers need to switch between multiple system interfaces to view information such as the distribution of broken ends across the entire machine, the trend of broken end rates in a certain area, and abnormal equipment energy consumption, resulting in low information integration efficiency.

[0006] Most of the published patents focus on a single aspect. For example, CN222139401U discloses a dual-head eddy current yarn breakage detection device that uses eddy current sensors to monitor the movement of the steel wire traveler to determine yarn breakage. However, it only provides detection functionality and does not address stop-feed control or multi-source data fusion analysis after yarn breakage. CN110952180A discloses an online detection system for a single spindle of a spinning frame, which uses a CAN bus network to collect single spindle status data and control roving stop-feed. However, its device functionality is relatively limited, and the fusion and utilization of energy consumption data, equipment health data, and yarn breakage data is insufficient. CN118547409A discloses an intelligent fault diagnosis method for textile equipment that diagnoses faults in spinning frame spindles through vibration data analysis. However, it focuses on the algorithm level and does not form a multi-functional integrated device solution that includes yarn breakage detection, energy consumption monitoring, and roving stop-feed execution. Summary of the Invention

[0007] The primary technical problem to be solved by this invention is to provide a spinning intelligent manufacturing management device based on multi-source data fusion, which addresses the issue of long response time between yarn breakage detection and roving feed stoppage, resulting in raw material waste.

[0008] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a spinning intelligent manufacturing management device based on multi-source data fusion, including a central processing unit, a data acquisition unit, a yarn breakage identification unit, a roving stop feeding execution unit, an energy consumption monitoring unit, and a human-machine interface.

[0009] The data acquisition unit includes a first type of sensor and a second type of sensor that correspond one-to-one with each spindle of the spinning machine. The first type of sensor and the second type of sensor are different types and are used to collect any two states of the corresponding spindle, such as the presence or absence of yarn, spindle speed, and the movement of the traveler.

[0010] The yarn breakage identification unit includes a three-channel comparison circuit and a timing logic circuit, which are used to receive signals from the data acquisition unit, perform logical judgments, and output judgment results.

[0011] The roving stop-feeding actuator is installed at the roving feed channel of the spinning machine and is electrically connected to the yarn breakage detection unit and the central processing unit, respectively. It executes the action according to the judgment result output by the yarn breakage detection unit. The roving stop-feeding actuator includes an electromagnet, a push rod, a return spring, a blade, and a position feedback switch. The electromagnet is fixedly connected to the rear surface of the spinning machine via a support. The push rod is slidably connected to the support and is parallel to the iron core of the electromagnet. An iron plate is vertically connected to the push rod, extending to one end of the electromagnet and facing the iron core of the electromagnet. One end of the push rod is directly opposite the roving, and the blade is fixedly installed at the end of the push rod directly opposite the roving. When the electromagnet is energized, it attracts the iron plate and drives the push rod to slide towards the roving. The push rod drives the blade to cut the roving. The position feedback switch is set on one side of the push rod, and the push rod is provided with a boss that cooperates with the position feedback switch. When the push rod moves to the position of cutting the roving, the boss on the push rod abuts against the position feedback switch, changing the open and closed state of the position feedback switch. The reset spring is sleeved on the push rod and is used to drive the push rod to reset away from the roving. The electromagnet is electrically connected to and controlled by the yarn breakage identification unit, and the position feedback switch is electrically connected to the central processing unit. The energy consumption monitoring unit includes a current transformer and a voltage transformer installed on the main motor power supply line of the spinning machine, and a branch energy metering module installed on the power supply lines of the workshop air conditioner and dust filter auxiliary machine; the energy consumption monitoring unit is electrically connected to the central processing unit.

[0012] The human-machine interface adopts a B / S architecture and supports access from desktop computers, tablets and smartphones via the workshop's 5G private network. The interface content includes a single spindle display area, a machine summary display area, and a workshop cockpit display area.

[0013] As a preferred embodiment, the first type of sensor is a photoelectric reflective sensor, which is installed in alignment with the yarn running path to detect the presence of yarn; the second type of sensor is a miniature Hall element, which is installed in conjunction with the magnet at the end of the spindle shaft to detect the spindle rotation speed.

[0014] As a preferred option, the individual spindle display area displays the spindle distribution of each spinning frame in the form of a layout diagram.

[0015] The machine summary display area shows the breakage rate Rbreak(i,t), cumulative power consumption Ecompute, energy efficiency judgment results, and overall integrity rate of the roving stop-feed execution unit for each spinning frame.

[0016] The workshop cockpit display area shows the trend curve of the overall breakage rate Rshop(t) of the workshop, the correlation chart between the real-time power of each machine and the breakage rate, as well as the top five machines with the most breakages and the top ten work groups with the most breakages.

[0017] The technical problem to be further solved by the present invention is to provide a spinning intelligent manufacturing management method based on multi-source data fusion, which solves the technical problem that conventional multi-source data guides the control of corresponding processes separately, resulting in complicated operation and inability to further improve spinning efficiency.

[0018] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is: a spinning intelligent manufacturing management method based on multi-source data fusion, which uses the above-mentioned spinning intelligent manufacturing management device based on multi-source data fusion and performs the following specific steps: S1. Data Acquisition: The data acquisition unit collects two states of each spindle: presence or absence of yarn and spindle rotation speed. The sampling frequency f1 of the photoelectric reflective sensor is set to 500Hz, and the sampled data is recorded as S1(k). The sampling frequency f2 of the miniature Hall element is set to 100Hz, and the sampled data is recorded as S2(k), where k is the sampling sequence number.

[0019] S2. Receive signals from the data acquisition unit using the yarn breakage identification unit and perform yarn breakage determination according to the following steps: Step A: Determine the amplitude of the photoelectric reflective sensor signal S1(k). When S1(k) is lower than the threshold V... th If the time exceeds T1 (T1=150ms) continuously, it is determined that there is a suspected interruption in the yarn path, and the first judgment flag F1=1 is output.

[0020] Step B: Calculate the spindle rotation speed signal S2(k). Let Np be the number of pulses per revolution, then the formula for calculating the rotation speed n is: ; Where f s The sampling frequency of the Hall element. The average time interval between two adjacent pulses is defined as follows: when n is lower than 90% of the rated speed nrated and the duration exceeds T2 (T2=200ms), the spindle is judged to be in an abnormal operating state, and the second judgment flag F2=1 is output; the rated speed nrated is sent from the central processing unit to the yarn breakage identification unit according to the yarn type.

[0021] Step C: Logically combine the first and second judgment flags; the judgment rules are as follows: when F1=1 and F2=1, confirm that a yarn breakage has occurred and output a yarn breakage signal Fbreak=1; when F1=1 but F2=0, determine that it is a sensor false alarm and do not output a yarn breakage signal; when F1=0 but F2=1, determine that it is a spindle drive fault and output an alarm signal Falarm=1.

[0022] The decision logic can be expressed by the following formula: ; S3. Execution Judgment Result: When the yarn breakage identification unit outputs Fbreak=1, the central processing unit sends a stop-feed command to the roving stop-feed execution unit. The miniature electromagnet of the roving stop-feed execution unit is energized, causing the push rod to compress the reset spring. The blade at the front end of the push rod extends and cuts the roving strip. The entire time from the occurrence of yarn breakage to the completion of the roving cutting action, Tresponse satisfies: ; Where Tdetect is the recognition time from sensor signal acquisition to Fbreak output by the yarn breakage detection unit, Tcomm is the signal transmission time, and Tactuate is the electromagnet action time; the overall response time Tresponse ≤ 75ms.

[0023] When the position feedback switch of the roving stop-feed execution unit is triggered, a position feedback signal is issued and recorded as F_feedback=1. When the central processing unit receives the signal F_feedback=1, it indicates that the stop-feeding action has been completed. If the signal F_feedback=1 is not received within 50ms after the stop-feeding command is sent, the central processing unit records the fault of the spindle position execution unit and reports it to the human-machine interface.

[0024] S4. Energy Consumption Monitoring: The energy consumption monitoring unit samples the operating power of each spinning machine in real time at a sampling frequency of 1Hz. Let the real-time power of the main motor of the spinning machine be P(t), then the cumulative power consumption E of the machine during the time period [t1, t2] is... compute for: ; The theoretical baseline power consumption E of this machine theoretical Based on the actual output Q and the benchmark energy consumption per unit output during this period. ref calculate: ; Where E ref The historical average power consumption per ton of yarn for this product under stable operating conditions is taken and dynamically updated by the central processing unit based on data from the MES system. When E compute > 1.15 · E theoretical When the machine is in operation, the energy consumption monitoring unit determines that the machine has an energy efficiency abnormality, outputs an energy consumption alarm signal Fenergy=1, and pushes relevant data such as the yarn breakage rate and spindle speed distribution of the machine to the human-machine interface to prompt the management personnel to check the equipment.

[0025] S5. Multi-source data fusion and yarn breakage rate analysis of the central processing unit: The central processing unit receives the yarn breakage signal reported by the yarn breakage identification unit and counts the number of yarn breaks according to three dimensions: machine, shift, and time period; the yarn breakage rate Rbreak(i, j, t) of the i-th spinning machine in the j-th shift and the t-th hour is defined as: ; Where Nbreak(i,j,t) is the total number of spindle breaks of the machine in a specified shift hour, Nspindle(i) is the total number of spindles of the machine, and 60 in the denominator is the time normalization coefficient, so that the unit of Rbreak is times / spindle·minute.

[0026] The central processing unit aligns Rbreak(i, j, t) with the real-time power P(t) of the machine reported by the energy consumption monitoring unit along the time axis, and generates a dual-axis trend curve with time on the horizontal axis, breakage rate on the left vertical axis, and main motor power on the right vertical axis, which is then displayed on the human-machine interface.

[0027] The central processing unit is also equipped with a workshop-level comprehensive evaluation index for yarn breakage rate, Rshop. If there are M spinning machines in the workshop, and the yarn breakage rate of the i-th spinning machine in hour t is Rbreak(i,t), then the comprehensive evaluation index for the workshop's yarn breakage rate in that hour is: ; Where wi is the weighting coefficient of the i-th spinning machine, which is set by the system parameters according to the process difficulty of the yarn variety spun by the machine.

[0028] S6. Data Output: The individual spindle display area displays the spindle distribution of each spinning machine in the form of a layout diagram. Spindles with Fbreak=1 are marked with dense red diagonal lines, Ffeedback=1 is marked with green dots, and spindles with energy consumption alarms are marked with flashing yellow. The interface supports touch screen operation. Clicking on any spindle icon can expand the yarn breakage history and sensor signal waveform playback of that spindle.

[0029] The machine summary display area shows the breakage rate Rbreak(i,t), cumulative power consumption Ecompute, energy efficiency judgment results, and overall integrity rate of the roving stop-feed execution unit for each spinning frame.

[0030] The workshop cockpit display area shows the trend curve of the overall breakage rate Rshop(t) of the workshop, the correlation chart between the real-time power of each machine and the breakage rate, as well as the top five machines with the most breakages and the top ten work groups with the most breakages.

[0031] Compared with the prior art, the present invention has the following beneficial effects: First, it achieves integrated yarn breakage detection and roving stop feeding. By integrating the dual-sensor complementary judgment circuit, the roving stop feeding execution unit, and the energy consumption monitoring module into the same device architecture, the device has richer functional dimensions compared with the single-spindle detection system with CAN bus networking, reduces the types of workshop equipment and the amount of on-site wiring, and facilitates maintenance and management.

[0032] Secondly, the response time for yarn breakage has been shortened. The overall response time of the device is ≤75ms, which is more than 50% shorter than the 150-300ms response time of the existing split system. The position feedback function of the roving stop feeding execution unit ensures the reliability of the operation and reduces the waste of raw materials caused by the continued feeding of roving after yarn breakage.

[0033] Third, it improves the reliability of yarn breakage detection. Employing a dual-channel complementary judgment logic of photoelectric detection and spindle speed detection effectively eliminates false alarms caused by changes in ambient light, yarn fly debris obstruction, or electromagnetic interference from a single sensor. Taking a 100,000-spindle factory as an example, the number of workers per 10,000 spindles has been reduced from 55 to 15, significantly improving the monitoring capabilities of each operator. The high reliability of the device provides technical support for reducing manpower and increasing efficiency.

[0034] Fourth, it achieves integrated display of multi-source data. Broken yarn rate, equipment energy consumption, and feed stoppage status are integrated and displayed on the same interface, allowing managers to fully grasp the workshop's operational status without switching between different systems. By aligning the broken yarn rate and equipment energy consumption on the same timeline using dual-axis trend curves, the correlation between abnormal equipment energy efficiency and changes in operating status can be intuitively identified, providing data support for preventative equipment maintenance and enabling early prediction of broken yarn risks. Attached Figure Description

[0035] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings, wherein: Figure 1 This is a schematic diagram of the overall structure of the intelligent manufacturing management device for spinning based on multi-source data fusion, as described in this invention. Figure 2 This is a schematic diagram of a specific structure of the data acquisition unit described in this invention; Figure 3 This is a schematic diagram of the structure of the roving stop-feeding execution unit described in this invention; Figure 4 This is a layout diagram of the single-spindle status monitoring area in the human-computer interaction interface described in this invention; Figure 5 This is a schematic diagram of the machine summary display interface in the human-computer interaction interface described in this invention; Figure 6 This is a schematic diagram of the workshop cockpit display interface in the human-computer interaction interface described in this invention; Figure 7 This is a timing waveform diagram of the complementary decision logic of the data acquisition unit described in this invention; Figure 8 This is a schematic diagram of the decapitation rate-power dual-axis trend curve generated by the central processing unit described in this invention; Figure 2 and Figure 3In Chinese: 1. Type I sensor; 2. Type II sensor; 3. Electromagnet; 4. Push rod; 5. Return spring; 6. Blade; 7. Position feedback switch; 8. Support; 9. Rear face of spinning machine; 10. Iron sheet; 11. Roving; 12. Boss; 13. Magnet; 14. Spindle; 15. Yarn. Detailed Implementation

[0036] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0037] like Figure 1 The device shown is a spinning intelligent manufacturing management device based on multi-source data fusion, including a central processing unit, a data acquisition unit, a yarn breakage identification unit, a roving stop-feed execution unit, an energy consumption monitoring unit, and a human-machine interface.

[0038] like Figure 2 As shown, the data acquisition unit includes a first type of sensor 1 and a second type of sensor 2 that correspond one-to-one with each spindle of the spinning machine. The first type of sensor 1 and the second type of sensor 2 are of different types and are used to collect the presence or absence of yarn and the spindle speed of the corresponding spindle, respectively.

[0039] In this embodiment, the first type of sensor 1 is preferably an Omron EE-SX series photoelectric reflective sensor, installed in alignment with the running path of the yarn 15, used to detect the presence of the yarn. The second type of sensor 2 is preferably an Allegro A3144 miniature Hall element, installed in conjunction with the magnet 13 at the end of the spindle shaft, used to detect the rotational speed of the spindle 14.

[0040] In this embodiment, the yarn breakage identification unit preferably uses a chip, such as an MCU, that includes a comparison circuit and a timing logic circuit. This chip receives signals from the data acquisition unit, performs logical judgments, and outputs the judgment results.

[0041] like Figure 3As shown, the roving stop-feed execution unit is installed at the roving feed channel of the spinning machine and is electrically connected to the yarn breakage identification unit and the central processing unit, respectively. It executes actions based on the judgment result output by the yarn breakage identification unit. The roving stop-feed execution unit includes an electromagnet 3, a push rod 4, a return spring 5, a blade 6, and a position feedback switch 7. The electromagnet 3 is fixedly connected to the rear surface 9 of the spinning machine via a support 8. The push rod 4 is slidably connected to the support 8 and parallel to the iron core of the electromagnet 3. A piece of iron 10 is vertically connected to the push rod 4, extending to one end of the electromagnet 3 and facing the iron core of the electromagnet 3. The top rod 4 is directly opposite the roving 11. The blade 6 is fixedly installed at the end of the top rod 4 directly opposite the roving 11. When the electromagnet 3 is energized, it attracts the iron plate 10 and drives the top rod 4 to slide towards the roving 11. The top rod 4 drives the blade 6 to cut the roving 11. The position feedback switch 7 is set on one side of the top rod 4. The top rod 4 is provided with a boss 12 that cooperates with the position feedback switch 7. When the top rod 4 moves to the position of cutting the roving, the boss 12 on the top rod 4 abuts against the position feedback switch 7, changing the opening and closing state of the position feedback switch 7. The reset spring 5 is sleeved on the top rod 4 and is used to drive the top rod 4 to reset in the direction away from the roving.

[0042] The electromagnet 3 is electrically connected to and controlled by the yarn breakage identification unit. The position feedback switch 7 is electrically connected to the central processing unit. When the boss 12 abuts against the position feedback switch 7, the central processing unit receives a high-level signal.

[0043] In this embodiment, the electromagnet 3 is selected as a model with a stroke of 5mm and an action time of ≤40ms.

[0044] The energy consumption monitoring unit includes current transformers and voltage transformers installed on the power supply line of the main motor of the spinning machine, and branch energy metering modules installed on the power supply lines of the workshop air conditioner and dust filter auxiliary equipment. The current transformers are installed on the U, V, and W three-phase input lines of each spinning machine's main motor, with an accuracy class of 0.5. The energy consumption monitoring unit is electrically connected to the central processing unit.

[0045] Since current transformers, voltage transformers, and energy metering modules are all conventional electrical components and their usage is well known to electrical technicians, the specific assembly structure diagrams of current transformers, voltage transformers, and energy metering modules are omitted in this embodiment.

[0046] The human-machine interface adopts a B / S architecture and supports access from desktop computers, tablets and smartphones via the workshop's 5G private network. The interface content includes a single spindle display area, a machine summary display area, and a workshop cockpit display area.

[0047] like Figure 4As shown, the individual spindle display area displays the spindle distribution of each spinning machine in a layout diagram. Spindles with Fbreak=1 are marked with dense red diagonal lines, Ffeedback=1 is marked with green dots, energy alarm spindles are marked with flashing yellow, and normal operation is marked with blank or very light gray. The interface supports touch screen operation; clicking any spindle icon will expand the yarn breakage history and sensor signal waveform playback for that spindle.

[0048] like Figure 5 As shown, the machine summary display area displays the breakage rate Rbreak(i,t), cumulative power consumption Ecompute, energy efficiency judgment results, and overall integrity rate of the roving stop-feed execution unit for each spinning frame.

[0049] like Figure 6 As shown, the workshop cockpit display area displays the trend curve of the overall decapitation rate Rshop(t), the correlation chart between the real-time power of each machine and the decapitation rate, as well as the top five machines with the most decapitations and the top ten work groups with the most decapitations.

[0050] The following example illustrates the intelligent manufacturing management method for spinning using the aforementioned intelligent manufacturing management device based on multi-source data fusion, with 48 spinning machines in a production team. Each spinning machine has 1216 spindles.

[0051] The specific steps of the intelligent manufacturing management method for spinning based on the above-mentioned multi-source data fusion-based intelligent manufacturing management device, according to the process requirements of spinning variety C14.8 tex, are as follows: S1. Data Acquisition: The data acquisition unit collects two states of each spindle: presence or absence of yarn and spindle rotation speed. The sampling frequency f1 of the photoelectric reflective sensor is set to 500Hz, and the sampled data is recorded as S1(k). The sampling frequency f2 of the miniature Hall element is set to 100Hz, and the sampled data is recorded as S2(k), where k is the sampling sequence number.

[0052] S2, such as Figure 7 As shown, the yarn breakage detection unit receives signals from the data acquisition unit and performs yarn breakage determination according to the following steps: Step A: Determine the amplitude of the photoelectric reflective sensor signal S1(k). When S1(k) is lower than the threshold V... th If the time exceeds T1 (T1=150ms) continuously, it is determined that there is a suspected interruption in the yarn path, and the first judgment flag F1=1 is output.

[0053] Step B: Calculate the spindle rotation speed signal S2(k). Let Np be the number of pulses per revolution, then the formula for calculating the rotation speed n is: ; Where fs The sampling frequency of the Hall element. The average time interval between two adjacent pulses is defined as follows: when n is lower than 90% of the rated speed nrated and the duration exceeds T2 (T2=200ms), the spindle is judged to be in abnormal operation, and the second judgment flag F2=1 is output; the rated speed nrated is sent from the central processing unit to the yarn breakage identification unit according to the yarn type. Step C: Logically combine the first and second judgment flags; the judgment rules are as follows: when F1=1 and F2=1, confirm that a yarn breakage has occurred and output a yarn breakage signal Fbreak=1; when F1=1 but F2=0, determine that it is a sensor false alarm and do not output a yarn breakage signal; when F1=0 but F2=1, determine that it is a spindle drive fault and output an alarm signal Falarm=1. The decision logic can be expressed by the following formula: ; The aforementioned dual-sensor complementary judgment structure can effectively eliminate false alarms caused by single-sensor systems due to changes in ambient light, fly debris obstruction, or electromagnetic interference. The accuracy of yarn breakage identification is improved by approximately 8 to 10 percentage points compared to the single-sensor solution.

[0054] Based on the process requirements of the C14.8 tex yarn variety, the rated spindle speed n_rated = 16300 r / min, and the amplitude threshold V of the photoelectric sensor... th The setting was 40% of the rated bright field reflection amplitude. After 8 hours of operation, the actual data for the yarn breakage detection unit were as follows: a total of 327 yarn breakage events occurred, with the dual-sensor complementary judgment outputting F_break=1 314 times. Of these, 307 were confirmed as genuine yarn breaks by manual verification, and 7 were false alarms. In addition, 13 yarn breakage events did not trigger F_break=1, all due to a malfunction in one of the dual sensors. The calculated yarn breakage detection accuracy was 307 / (307+7) = 97.8%, an improvement of 8.8 percentage points compared to the approximately 89% accuracy achieved when using a single photoelectric sensor solution in the same workshop.

[0055] A review of the seven false alarms revealed that five of them were caused by fly waste obstructing the photoelectric sensor window. However, the spindle speed sensor signals were normal (F1=1, F2=0). In the decision logic, F1=1 and F2=0 did not output a yarn breakage signal, thus preventing roving malfunctions. This demonstrates that the dual-channel decision mechanism effectively avoids invalid feed stop actions caused by sensor false alarms.

[0056] Figure 7 The waveform shown illustrates the timing relationship between sensor 1 (type 1) and sensor 2 (type 2) in a single yarn breakage event, used to explain F. breakThe judgment logic is as follows. In this embodiment, the yarn breakage recognition unit ran continuously for 8 hours, resulting in 327 yarn breakage events and a recognition accuracy rate of 97.8%. All judgments followed the attached... Figure 7 The timing logic shown, the waveforms of each yarn breakage event and the attached... Figure 7 Consistent.

[0057] S3. Execution Judgment Result: When the yarn breakage identification unit outputs Fbreak=1, the central processing unit sends a stop-feed command to the roving stop-feed execution unit. The miniature electromagnet of the roving stop-feed execution unit is energized, causing the push rod to compress the reset spring. The blade at the front end of the push rod extends and cuts the roving strip. The entire time from the occurrence of yarn breakage to the completion of the roving cutting action, Tresponse satisfies: ; Where Tdetect is the recognition time from sensor signal acquisition to Fbreak output by the yarn breakage detection unit, which is ≤30ms in actual measurement; Tcomm is the signal transmission time, which is ≤5ms in actual measurement using 5G edge computing node transmission; and Tactuate is the electromagnet action time, which is ≤40ms in actual measurement. The overall response time Tresponse is ≤75ms, which is more than 50% shorter than the 150-300ms response time of the existing split yarn breakage detection + roving stop feeding system.

[0058] Under continuous production conditions, five spindles were randomly selected, and high-speed cameras were used to record the entire process from yarn breakage to the top rod cutting the roving. The response times for the five samples were 68ms, 72ms, 71ms, 69ms, and 74ms, with an average of 70.8ms. The average yarn breakage recognition time was 28ms, the average 5G communication transmission time was 4ms, and the average electromagnet action time was 38.8ms. Compared to existing split-type systems, the response time is reduced by approximately 70%. Split-type systems, due to the need for multiple intermediate links, have actual measured values ​​between 220ms and 280ms. Assuming that failure to stop feeding promptly after a yarn breakage event results in an additional 2.5g of roving wasted, and considering 1 million yarn breakage events per year in a 10,000-spindle workshop, the reduced response time can reduce roving waste by approximately 2.5 tons.

[0059] When the position feedback switch of the roving stop-feed execution unit is triggered, a position feedback signal is sent and recorded as F_feedback=1. When the central processing unit receives the signal F_feedback=1, it indicates that the stop-feeding action has been completed. If the signal F_feedback=1 is not received within 50ms after the stop-feeding command is sent, the central processing unit records the fault of the spindle position execution unit and reports it to the human-machine interface. S4. Energy Consumption Monitoring: The energy consumption monitoring unit samples the operating power of each spinning machine in real time at a sampling frequency of 1Hz. Let the real-time power of the main motor of the spinning machine be P(t), then the cumulative power consumption E of the machine during the time period [t1, t2] is... compute for: ; The theoretical baseline power consumption of this machine is based on the actual output Q and the baseline energy consumption per unit output E during this period. ref calculate: ; Where E ref The historical average power consumption per ton of yarn for this product under stable operating conditions is taken and dynamically updated by the central processing unit based on data from the MES system. When E compute > 1.15 · E theoretical When the machine is in operation, the energy consumption monitoring unit determines that the machine has an energy efficiency abnormality, outputs an energy consumption alarm signal Fenergy=1, and pushes relevant data such as the yarn breakage rate and spindle speed distribution of the machine to the human-machine interface, prompting the management personnel to check the equipment. S5. Multi-source data fusion and yarn breakage rate analysis of the central processing unit: The central processing unit receives the yarn breakage signal reported by the yarn breakage identification unit and counts the number of yarn breaks according to three dimensions: machine, shift, and time period; the yarn breakage rate Rbreak(i, j, t) of the i-th spinning machine in the j-th shift and the t-th hour is defined as: ; Where Nbreak(i,j,t) is the total number of head breaks of the machine in a specified shift hour, Nspindle(i) is the total number of spindles of the machine, and 60 in the denominator is the time normalization coefficient, so that the unit of Rbreak is times / spindle·minute. The central processing unit aligns Rbreak(i, j, t) with the real-time power P(t) of the machine reported by the energy consumption monitoring unit along the time axis, and generates a dual-axis trend curve with time on the horizontal axis, breakage rate on the left vertical axis, and main motor power on the right vertical axis, which is then displayed on the human-machine interface.

[0060] The central processing unit is also equipped with a workshop-level comprehensive evaluation index for yarn breakage rate, Rshop. If there are M spinning machines in the workshop, and the yarn breakage rate of the i-th spinning machine in hour t is Rbreak(i,t), then the comprehensive evaluation index for the workshop's yarn breakage rate in that hour is: ; Where wi is the weight coefficient of the i-th spinning machine, which is set by the system parameters according to the process difficulty of the yarn variety spun by the machine; the weight coefficient of machines with higher process difficulty is taken as a larger value; this indicator is used to quickly judge the changing trend of the overall operating status of the workshop.

[0061] The energy consumption monitoring unit operated continuously for 72 hours, recording the cumulative power consumption and number of disconnections for each machine every hour. Data from machine number 6 at hour 48 was selected for detailed analysis: (e.g.,...) Figure 8 As shown, the hourly breakage rate Rbreak was 0.38 times / spindle·minute, and the real-time power of the main motor remained stable between 38.2 kW and 39.5 kW, indicating normal energy efficiency (Ecompute / Etheoretical = 1.02). In the 56th hour, the Rbreak of this machine increased to 0.52 times / spindle·minute, while the real-time power of the main motor decreased to the range of 36.1 kW to 37.3 kW, and Ecompute / Etheoretical decreased to 0.94, significantly lower than the normal range. The energy consumption monitoring unit can output an energy consumption alarm signal Fenergy=1 to the human-machine interface. This signal can be displayed as a flashing energy consumption indicator light or as text, depending on the user's needs. Figure 8 The above is a textual representation.

[0062] The central processing unit automatically marked the abnormal energy efficiency range on the dual-axis trend curve for that hour and pushed the relevant data to the management personnel's mobile devices. On-site inspection revealed that insufficient tension in the drive belt of machine No. 6 caused a decrease in the rotational speed of some spindles, resulting in both an increased breakage rate and a reduced load on the main motor. After adjusting the belt tension, the machine's Rbreak recovered to 0.36 times / spindle·minute in the 64th hour, and Ecompute / Etheoretical rose to 0.99. This example demonstrates that the unit's multi-source data fusion capability helps management personnel quickly locate hidden equipment performance degradation issues.

[0063] S6. Data Output: (e.g., ...) Figure 4 As shown, the individual spindle display area displays the spindle distribution of each spinning machine in the form of a layout diagram. Spindles with Fbreak=1 are marked with dense red diagonal lines, Ffeedback=1 is marked with green dots, and spindles with energy consumption alarms are marked with flashing yellow. The interface supports touch screen operation. Clicking on any spindle icon can expand the yarn breakage history and sensor signal waveform playback for that spindle.

[0064] like Figure 5 As shown, the machine summary display area displays the breakage rate Rbreak(i,t), cumulative power consumption Ecompute, energy efficiency judgment results, and overall integrity rate of the roving stop-feed execution unit for each spinning frame.

[0065] like Figure 6As shown, the workshop cockpit display area displays the trend curve of the overall decapitation rate Rshop(t), the correlation chart between the real-time power of each machine and the decapitation rate, as well as the top five machines with the most decapitations and the top ten work groups with the most decapitations.

[0066] The above embodiments are merely illustrative of the principles and effects of the present invention, as well as some examples of its application, and are not intended to limit the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the inventive concept of the present invention, and these modifications and improvements are all within the scope of protection of the present invention.

Claims

1. A spinning intelligent manufacturing management device based on multi-source data fusion, characterized in that, It includes a central processing unit, a data acquisition unit, a yarn breakage identification unit, a roving stop-feed execution unit, an energy consumption monitoring unit, and a human-machine interface; The data acquisition unit includes a first type of sensor (1) and a second type of sensor (2) corresponding to each spindle of the spinning machine. The first type of sensor (1) and the second type of sensor (2) are different types and are used to collect any two states of the corresponding spindle, such as the presence or absence of yarn, spindle speed, and movement of the wire traveler. The yarn breakage identification unit is used to receive signals from the data acquisition unit, perform logical judgments, and output judgment results. The roving stop feeding execution unit is installed at the roving feeding channel of the spinning machine and is electrically connected to the yarn breakage identification unit and the central processing unit respectively. It executes the action according to the judgment result output by the yarn breakage identification unit. The roving stop feeding execution unit includes an electromagnet (3), a push rod (4), a reset spring (5), a blade (6) and a position feedback switch (7). The electromagnet (3) is fixedly connected to the rear surface (9) of the spinning machine through a support (8). The push rod (4) is slidably connected to the support (8) and parallel to the iron core of the battery iron (3). An iron plate (10) is vertically connected to the push rod (4). The iron plate (10) extends to one end of the electromagnet (3) and is directly opposite the iron core of the electromagnet (3). One end of the push rod (4) Opposite to the roving, the blade (6) is fixedly installed on the end of the top rod (4) facing the roving. The electromagnet (3) is energized to attract the iron plate (10) and drive the top rod (4) to slide towards the roving. The top rod (4) drives the blade (6) to cut the roving. The position feedback switch (7) is set on one side of the top rod (4). The top rod (4) is provided with a boss (12) that cooperates with the position feedback switch (7). When the top rod (4) moves to the position of cutting the roving, the boss (12) on the top rod (4) abuts against the position feedback switch (7) to change the opening and closing state of the position feedback switch (7). The reset spring (5) is sleeved on the top rod (4) and is used to drive the top rod (4) to reset away from the roving. The electromagnet (3) is electrically connected to and controlled by the yarn breakage identification unit, and the position feedback switch (7) is electrically connected to the central processing unit. The energy consumption monitoring unit includes a current transformer and a voltage transformer installed on the main motor power supply line of the spinning machine, and a branch energy metering module installed on the power supply lines of the workshop air conditioner and dust filter auxiliary machine; the energy consumption monitoring unit is electrically connected to the central processing unit. The human-machine interface adopts a B / S architecture and supports access from desktop computers, tablets and smartphones via the workshop's 5G private network. The interface content includes a single spindle display area, a machine summary display area, and a workshop cockpit display area.

2. The intelligent manufacturing management device for spinning based on multi-source data fusion according to claim 1, characterized in that, The first type of sensor (1) is a photoelectric reflective sensor, which is installed in alignment with the yarn running path and is used to detect the presence of yarn; the second type of sensor (2) is a miniature Hall element, which is installed in conjunction with the magnet at the end of the spindle shaft and is used to detect the spindle speed.

3. The intelligent manufacturing management device for spinning based on multi-source data fusion according to claim 1, characterized in that, The individual spindle display area shows the spindle distribution of each spinning machine in the form of a layout diagram; The machine summary display area shows the breakage rate Rbreak(i,t), cumulative power consumption Ecompute, energy efficiency judgment results, and overall integrity rate of the roving stop-feed execution unit for each spinning frame; The workshop cockpit display area shows the trend curve of the overall breakage rate Rshop(t) of the workshop, the correlation chart between the real-time power of each machine and the breakage rate, as well as the top five machines with the most breakages and the top ten work groups with the most breakages.

4. A method for intelligent manufacturing management of spinning based on multi-source data fusion, characterized in that, Using the intelligent manufacturing management device for spinning based on multi-source data fusion as described in claim 2 or 3, the following specific steps are performed: S1. Data Acquisition: The data acquisition unit collects two states of each spindle: the presence or absence of yarn and the spindle rotation speed. The sampling frequency f1 of the photoelectric reflective sensor is set to 500Hz, and the sampled data is recorded as S1(k). The sampling frequency f2 of the miniature Hall element is set to 100Hz, and the sampled data is recorded as S2(k), where k is the sampling sequence number. S2. Receive signals from the data acquisition unit using the yarn breakage identification unit and perform yarn breakage determination according to the following steps: Step A: Determine the amplitude of the photoelectric reflective sensor signal S1(k). When S1(k) is lower than the threshold V... th If the time exceeds T1 (T1=150ms) continuously, it is determined that there is a suspected interruption in the yarn path, and the first judgment flag F1=1 is output; Step B: Calculate the spindle rotation speed signal S2(k). Let Np be the number of pulses per revolution, then the formula for calculating the rotation speed n is: ； Where f s The sampling frequency of the Hall element. The average time interval between two adjacent pulses is defined as follows: when n is lower than 90% of the rated speed nrated and the duration exceeds T2 (T2=200ms), the spindle is judged to be in abnormal operation, and the second judgment flag F2=1 is output; the rated speed nrated is sent from the central processing unit to the yarn breakage identification unit according to the yarn type. Step C: Logically combine the first and second judgment flags; the judgment rules are as follows: when F1=1 and F2=1, confirm that a yarn breakage has occurred and output a yarn breakage signal Fbreak=1; when F1=1 but F2=0, determine that it is a sensor false alarm and do not output a yarn breakage signal; when F1=0 but F2=1, determine that it is a spindle drive fault and output an alarm signal Falarm=1. The decision logic can be expressed by the following formula: ； S3. Execution Judgment Result: When the yarn breakage identification unit outputs Fbreak=1, the central processing unit sends a stop-feed command to the roving stop-feed execution unit. The miniature electromagnet of the roving stop-feed execution unit is energized, causing the push rod to compress the reset spring. The blade at the front end of the push rod extends and cuts the roving strip. The entire time from the occurrence of yarn breakage to the completion of the roving cutting action, Tresponse satisfies: ； Where Tdetect is the recognition time from sensor signal acquisition to Fbreak output by the yarn breakage detection unit (measured ≤30ms), Tcomm is the signal transmission time (using 5G edge computing node transmission, measured ≤5ms), Tactuate is the electromagnet action time (measured ≤40ms); the overall response time Tresponse ≤75ms; When the position feedback switch of the roving stop-feed execution unit is triggered, a position feedback signal is sent and recorded as F_feedback=1. When the central processing unit receives the signal F_feedback=1, it indicates that the stop-feeding action has been completed. If the signal F_feedback=1 is not received within 50ms after the stop-feeding command is sent, the central processing unit records the fault of the spindle position execution unit and reports it to the human-machine interface. S4. Energy Consumption Monitoring: The energy consumption monitoring unit samples the operating power of each spinning machine in real time at a sampling frequency of 1Hz. Let the real-time power of the main motor of the spinning machine be P(t), then the cumulative power consumption E of the machine during the time period [t1, t2] is... compute for: ； The theoretical baseline power consumption of this machine is based on the actual output Q and the baseline energy consumption per unit output E during this period. ref calculate: ； Where E ref The historical average power consumption per ton of yarn for this product under stable operating conditions is taken and dynamically updated by the central processing unit based on data from the MES system. When E compute > 1.15 · E theoretical When the machine is in operation, the energy consumption monitoring unit determines that the machine has an energy efficiency abnormality, outputs an energy consumption alarm signal Fenergy=1, and pushes relevant data such as the yarn breakage rate and spindle speed distribution of the machine to the human-machine interface, prompting the management personnel to check the equipment. S5. Multi-source data fusion and yarn breakage rate analysis of the central processing unit: The central processing unit receives the yarn breakage signal reported by the yarn breakage identification unit and counts the number of yarn breaks according to three dimensions: machine, shift, and time period; the yarn breakage rate Rbreak(i, j, t) of the i-th spinning machine in the j-th shift and the t-th hour is defined as: ； Where Nbreak(i,j,t) is the total number of head breaks of the machine in a specified shift hour, Nspindle(i) is the total number of spindles of the machine, and 60 in the denominator is the time normalization coefficient, so that the unit of Rbreak is times / spindle·minute. The central processing unit aligns Rbreak(i,j,t) with the real-time power P(t) of the machine reported by the energy consumption monitoring unit according to the time axis, and generates a dual-axis trend curve with time on the horizontal axis, breakage rate on the left vertical axis and main motor power on the right vertical axis, which is then displayed on the human-machine interface. The central processing unit is also equipped with a workshop-level comprehensive evaluation index for yarn breakage rate, Rshop. If there are M spinning machines in the workshop, and the yarn breakage rate of the i-th spinning machine in hour t is Rbreak(i,t), then the comprehensive evaluation index for the workshop's yarn breakage rate in that hour is: ； Where wi is the weight coefficient of the i-th spinning machine, which is set by the system parameters according to the process difficulty of the yarn variety spun by the machine; S6. Data Output: The individual spindle display area displays the spindle distribution of each spinning machine in the form of a layout diagram. Spindles with Fbreak=1 are marked with dense red diagonal lines, Ffeedback=1 is marked with green dots, and spindles with energy consumption alarms are marked with flashing yellow. The interface supports touch screen operation. Clicking on any spindle icon can expand the yarn breakage history and sensor signal waveform playback of that spindle. The machine summary display area shows the breakage rate Rbreak(i,t), cumulative power consumption Ecompute, energy efficiency judgment results, and overall integrity rate of the roving stop-feed execution unit for each spinning frame; The workshop cockpit display area shows the trend curve of the overall breakage rate Rshop(t) of the workshop, the correlation chart between the real-time power of each machine and the breakage rate, as well as the top five machines with the most breakages and the top ten work groups with the most breakages.

Images