A spinning machine cluster pressure collaborative control method based on multi-source evidence fusion

Abstract

The application discloses a spinning machine cluster pressure collaborative control method based on multi-source evidence fusion, comprising the following steps: acquiring real-time state parameters of each spinning machine, calculating derivative state parameters of each spinning machine and a connecting area thereof through the real-time state parameters; forming abnormal evidence according to abnormal state parameters of each spinning machine, fusing the abnormal evidence of all spinning machines through DS evidence theory to obtain global abnormal confidence; dynamically generating a cluster safety envelope line and a dynamic safety threshold of real-time state parameters according to the global abnormal confidence and state parameters of each section; and judging the real-time state through the dynamic safety threshold and the global abnormal confidence and controlling the spinning machine cluster to work.

Description

Technical Field

[0001] The present invention relates to the field of spinning machine control technology, and in particular to a spinning machine cluster pressure collaborative control method based on multi-source evidence fusion. Background Art

[0002] A spinning machine is a type of plastic processing equipment, a type of metal forming machine tool, primarily used for high-precision machining of thin-walled pipes, cylindrical, or conical parts. Through the combined action of a die and a spinning wheel, the metal sheet is gradually deformed plastically along the die shape by applying localized pressure while rotating. The key to its control lies in the precise regulation of the force (pressure) applied by the spinning wheel.

[0003] With the growing demand for large, extra-long metal cylinders and extra-large curved parts in industries such as aerospace, energy, and pressure vessels, distributed collaborative parallel forming of multiple spinning machines has become the mainstream process for solving the overall manufacturing problem of super-large components. This model uses multiple spinning machines to be responsible for different segments or forming areas of the parts, realizing spatial parallel processing. Then, in actual processing, each segment workstation needs to be coordinated in real time. Working conditions such as pressure, strain, and temperature are highly correlated. Local anomalies can easily affect global performance, and the connection parts of the forming area are particularly sensitive. When a single device has weak anomalies such as material inclusions, local overloads, and mold wear, the signal of a single machine is often difficult to accurately determine, but the anomaly can be transmitted through the workpiece, causing risks to the forming quality of the entire large part. Traditional spinning machine control is limited to local parameter feedback, without a multi-machine information fusion mechanism, and the abnormal response delay is large, which cannot meet the high consistency and high safety production requirements of large structures.

[0004] In view of the above problems in the prior art, the purpose of this invention is to design a cluster pressure collaborative control method of spinning machines based on multi-source evidence fusion. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to propose a cluster pressure collaborative control method of spinning machines based on multi-source evidence fusion, which can solve the above problems.

[0006] The present invention provides a method for coordinated pressure control of a spinning machine cluster based on multi-source evidence fusion, comprising:

[0007] Obtain the real-time status parameters of each spinning machine and calculate the derived status parameters of each spinning machine and its connecting area based on the real-time status parameters;

[0008] Abnormal evidence is generated based on the abnormal state parameters of each spinning machine. The abnormal evidence of all spinning machines is fused through DS evidence theory to obtain the global abnormal confidence level.

[0009] Dynamically generate cluster security envelopes and dynamic security thresholds of real-time status parameters based on global anomaly confidence and status parameters of each segment;

[0010] Real-time status judgment and control of the spinning machine are carried out through dynamic safety thresholds and global abnormality confidence.

[0011] Beneficial effects of the present invention:

[0012] First, through the parameter coupling analysis of local and connecting areas (such as local pressure change rate, material flow coupling coefficient, etc.), a digital description of the dynamic coordination of processes between equipment is achieved, and adjacent anomalies and uneven fluctuations in material areas are effectively perceived; compared with traditional single-point monitoring or absolute value comparison, the collection and calculation of derivative quantities such as change rate and coordinated difference can perceive the signs of mutations and asynchronous responses in advance, with higher sensitivity and predictability, which is conducive to rapid early warning and regulation.

[0013] Secondly, by introducing the DS theory, it is possible to automatically integrate the acquisition qualities of different devices and different signals by assigning weights. Even under conditions of information missing, local inconsistency or even data conflict, it can still output a reasonable global anomaly confidence level, greatly improving the robustness and fault tolerance of the system. The formed global confidence level provides a quantitative and traceable criterion for subsequent dynamic safety threshold adjustment and hierarchical regulation, enhancing the scientific nature of anomaly diagnosis and its practical operational guidance.

[0014] Third, dynamic safety thresholds are used to achieve safety control boundaries that are linked to the global process status and adaptive to time and place; dynamic adjustment of envelopes and critical values ​​can be adaptively adjusted in time with changes in production status, material batches and equipment health, significantly reducing the frequency of false alarms, missed alarms and manual adjustments; multi-parameter and multi-segment thresholds can be refined separately, improving the local sensitivity and overall coordination of complex systems, and improving the safety and automation level of the overall production process.

[0015] Fourth, through a graded response mechanism based on dual criteria of multi-dimensional state quantities and global risks, the operating mode (normal / adjustment / isolation) can be automatically switched according to the severity of abnormal risks, eliminating blind shutdowns or allowing abnormalities to spread. Each control result is fed back to the system to form a closed-loop optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0017] Figure 1 It is a flow chart of the method of this embodiment. DETAILED DESCRIPTION

[0018] To facilitate understanding by those skilled in the art, the structure of the present invention will now be further described in detail with reference to the embodiments and accompanying drawings. It should be understood that the steps mentioned in this embodiment, unless otherwise specified, can be adjusted in sequence according to actual needs, and can even be executed simultaneously or partially simultaneously.

[0019] Example 1

[0020] like Figure 1 As shown, this embodiment provides a method for coordinated pressure control of a spinning machine cluster based on multi-source evidence fusion, including:

[0021] S1 obtains the real-time state parameters of each spinning machine and calculates the derived state parameters of each spinning machine and its connection area based on the real-time state parameters;

[0022] S101 obtains the radial pressure of the spinning wheel of each spinning machine , spindle torque , segmental strain , strain rate ,temperature , feed speed , workstation space coordinates ;

[0023] S102 through the radial pressure of the spinning wheel of each spinning machine Calculate its local transient pressure change rate , the calculation formula is as follows:

[0024] ,

[0025] in, Indicates the radial pressure of the wheel at the current moment, Indicates the pressure value at the previous sampling moment, Indicates the sampling period;

[0026] S103 performs collaborative dynamic matching of the real-time state parameters of adjacent workstations based on the adjacent distances and material flow relationships in the workstation connection areas of each spinning machine, and calculates the parameter differences and coupling pressure change rates of each workstation and its connection area.

[0027] S1031 calculates the material flow coupling coefficient based on the adjacent distance and the feed rate between materials. The calculation formula is as follows:

[0028] ，

[0029] in, represents the material flow coupling coefficient, Indicates the maximum allowable spacing, represents the average feed speed, Represent the spatial coordinates of the i-th and j-th stations respectively, Respectively represent the feed speed of the i-th and j-th spinning machines;

[0030] In this step, the material flow coupling coefficient represents the degree of mutual influence between the material flow processes of spinning machines (stations) number i and j due to differences in spatial distance and feed speed. A larger value indicates a stronger material flow connection between the two stations.

[0031] S1032 local transient pressure change rate and material flow coupling coefficient Calculate the dynamic coupling pressure change rate using the following formula:

[0032] ,

[0033] in, Indicates the Dynamic coupling pressure change rate between stations, represents the calibrated strain-pressure coupling coefficient, represents the strain coordination deviation, The table shows the transient pressure change rate of the i-th and j-th stations. ∈[0,1].

[0034] In this step, the dynamic coupling pressure change rate comprehensively reflects the amplification or suppression effect of spatial proximity, velocity coordination, pressure dynamics and strain coordination on the local pressure mutation of the material. The larger the rate, the higher the probability of abnormal or impact events in the area. It indicates the difference in local deformation distribution of the material between the two workstations. The larger the strain difference, the more uneven the deformation in the section, which can easily lead to abnormal fluctuations (such as local hardening, cracking and other failure modes).

[0035] Traditionally, only looking at absolute pressure values ​​fails to reveal early signs of sudden pressure changes. The rate of change of pressure (i.e., pressure slope) can sensitively reflect short-term process disturbances such as material hardening, inclusions, and roller anomalies. The rate of change of local pressure on each spinning machine can reveal instances of pressure asynchrony and out-of-phase response at adjacent stations. If a spinning machine at an adjacent station is suddenly subjected to abnormal force, its pressure will surge, potentially causing a localized "strip break" in the material within that section and transmitting the "impact" to other stations. This provides dynamic, coordinated characteristic data of pressure and strain at each station / section for subsequent DS fusion and distributed global judgment.

[0036] S2 generates abnormal evidence based on the abnormal state parameters of each spinning machine, and fuses the abnormal evidence of all spinning machines through DS evidence theory to obtain the global abnormal confidence;

[0037] S201 compares the real-time state parameters and derived state parameters of each spinning machine in the current cycle with the corresponding process allowable range, determines whether each parameter is abnormal, and forms an abnormal mark;

[0038] In this step, if the abnormality is marked as 1 and the normality is marked as 0, multiple parameters form a vector, and a healthy (0) / abnormal (1) distinction is established for each device and each parameter, which is the basic condition for evidence construction.

[0039] S202 constructs a basic probability distribution function under the DS evidence theory by marking the anomalies, and forms an intra-group evidence set for all DS anomaly evidence of the spinning machine;

[0040] S2021 defines the weights of various state parameters through expert experience and historical data;

[0041] In this step, the individual anomaly marks do not reflect the importance and sensitivity of the parameters. Therefore, weights can be defined based on expert experience and historical data to reflect the contribution of different physical quantities to the anomaly criterion. For example, pressure may be more sensitive than torque.

[0042] S2022 calculates the abnormal support by using the abnormal mark of each state parameter and the corresponding state parameter weight. The calculation formula is as follows:

[0043] ,

[0044] in, represents the abnormal state support of the i-th spinning machine, k Represents various state parameters, Indicates the k The importance weight of each parameter in anomaly detection, Indicates the k Exception indication of parameters;

[0045] In this step, weighting is used to assign higher weights to the most sensitive and prioritized state parameters for more accurate anomaly detection. The weighted summation of the anomaly results for multiple parameters yields an anomaly support score between 0 and 1, which intuitively reflects the overall confidence level that the device is experiencing an anomaly. DS theory requires input of the basic probability allocation (BPA), which measures the support for the "anomaly" proposition for each device / source. Anomaly support is essential for integrating multidimensional monitoring data into usable "evidence," ensuring that subsequent evidence from multiple devices can be effectively combined.

[0046] S2023 adjusts the uncertainty weight for stations with missing data. The calculation formula is as follows:

[0047] ,

[0048] in, represents the basic probability distribution of the data fusion set of the i-th spinning machine, represents the abnormal state support of the i-th spinning machine, represents the health status support of the i-th spinning machine, Represents the complete set of all possible states.

[0049] In this step, according to Dempster-Shafer (DS) theory, the confidence scores for each device / evidence source should sum to 1. If device data is missing, some signals are corrupted, or parameters are incomplete, an accurate judgment of abnormality or health cannot be made. In this case, the "missing judgment" portion cannot be forcibly classified as abnormal or healthy. Instead, the weight of this "unable to judge" portion is assigned to the entire set Θ, reflecting the "unclear tendency" of the belief about the status of this device (period).

[0050] S203 recursively integrates all local evidence according to the Dempster synthesis rule in DS theory to obtain the global anomaly confidence, classifies the evidence according to the size of the global anomaly confidence, and outputs the current anomaly level.

[0051] In this step, DS theory (evidence theory) is a classic uncertainty management method. It can unify and recursively integrate information from different sources, with varying degrees of reliability and granularity—in the form of "support," "skepticism," and "uncertainty"—to provide a unified model and ultimately a holistic credibility and risk assessment. Global anomaly confidence transforms each spinning machine's anomaly into credible evidence. Through DS theory integration, the total support for the anomaly event across the entire production cluster is derived, serving as a basis for decision-making regarding safety, risk, and intelligent control.

[0052] Different equipment has varying anomaly types, abnormal signal distributions, and data reliability. DS can weight and fuse these confidence levels to enhance the robustness of the overall assessment. When data from some workstations or equipment is severely abnormal, while data from other equipment is normal or missing, DS theory can still generate adaptive weights, handle uncertainty, and ultimately generate a reliable output, unlike traditional probability statistics, which require "deterministic data." The abnormal (or healthy) information of each piece of equipment in a spinning cluster is locally independent yet mutually influential. This requires a comprehensive and dynamic integration of abnormal information to produce a reliable, comprehensive conclusion.

[0053] S3 dynamically generates the cluster security envelope and dynamic security thresholds of real-time status parameters based on the global anomaly confidence and status parameters of each segment;

[0054] S301 performs statistical analysis on the real-time status parameters of each section, extracts the interval boundaries of each section under normal operating conditions, and obtains the basic safety envelope;

[0055] S302 dynamically modifies the basic safety of each segment using the global anomaly confidence level to obtain the cluster safety envelope and dynamic safety threshold.

[0056] S3021 defines the adjustment factor, and the calculation formula is as follows:

[0057] ,

[0058] in, Indicates sensitivity, represents the anomaly confidence threshold, represents the global anomaly confidence;

[0059] S3022 If the global anomaly confidence >Confidence threshold , then tighten the envelope range, the calculation formula is as follows:

[0060] New boundary = original boundary × (1-λ);

[0061] S3023 If the global anomaly confidence ≤Confidence Threshold , then relax the envelope range, and the calculation formula is as follows:

[0062] New boundary = original boundary × (1 + λ / 2);

[0063] S3024 obtains the dynamic safety threshold of the real-time status parameters of each segment based on the latest cluster safety envelope.

[0064] In this step, factors such as process, material batch, equipment status, and external disturbances often render static and empirical thresholds ineffective, leading to missed or misjudgment. Using global anomaly confidence, the system dynamically adjusts safety thresholds globally. Based on the latest cluster safety envelope, dynamic safety thresholds for key state parameters such as pressure, strain, and torque are determined for each section. Using the pressure threshold as the primary control indicator, combined with strain and torque, this system performs multi-dimensional identification of abnormal operating conditions, improving system safety and robustness.

[0065] S4 uses dynamic safety thresholds and global anomaly confidence levels to make real-time status judgments and control the operation of the spinning machine cluster.

[0066] S401 uses the ratio of real-time status parameters and their corresponding safety thresholds, and the global anomaly confidence Construct state diagnosis vector D;

[0067] S402 Based on the global anomaly confidence and state diagnosis vectors for hierarchical control of spinning machine cluster operation;

[0068] S4021 ≤0.3 and max(D)<1, normal automatic operation is carried out;

[0069] S4022 > 0.3 and max(D)∈[1,1.2], prioritize the real-time status parameters that exceed the dynamic safety threshold and adjust the workstation accordingly;

[0070] Furthermore, if the material flow coupling coefficient between this station and the adjacent station is If the maximum value is greater than 0.6, the corresponding real-time status parameters of adjacent workstations are adjusted synchronously.

[0071] S4023 >0.6 or max(D)>1.5, start isolation warning and switch to manual review and fault self-diagnosis mode.

[0072] In this step, it is assumed that the following data are collected at a certain moment in a certain process: =95MPa, =100MPa, =0.09, =0.10, =0.25, then D=[0.95,0.90,…,0.25], max(D)=0.95<1, indicating that the global anomaly confidence is extremely low, there is no sign of anomaly, and each parameter has not reached the threshold. The system continues to operate automatically according to the predetermined process without intervention.

[0073] Assumptions =0.38, D=[1.05,0.98,…,0.38], max(D)=1.05∈[1,1.2], indicating that some parameters are close to or just exceed the safety threshold, but have not exceeded it far. The global anomaly confidence If cluster anomalies are showing an increasing trend, gentle adjustments, such as reducing the feed rate, can be implemented to return the system to a safe zone. Feed rate is the most influential, directly controllable variable in plastic forming processes like spinning. Reducing the feed rate typically reduces stress and temperature rise in the deformation zone (slowing the expansion of anomalies), improving process stability and safety while allowing the system more time to recover. By determining the material flow coupling coefficient, the system can be brought back to a safe zone more quickly and stably, preventing the deterioration of localized anomaly adjustments or the amplification of new fluctuations to other workstations.

[0074] Assumptions =0.68>0.6, D=[1.52,0.97,…,0.68], max(D)=1.52>1.5, indicating a very high global abnormality risk. A parameter has seriously exceeded its limit, threatening safe operation. Immediately implement mandatory control measures, such as isolating the faulty workstation, issuing an alarm, suspending feed, or performing manual intervention, to prioritize equipment and product safety.

[0075] These grading values ​​are obtained through historical operating condition statistics, risk assessment experience and actual equipment debugging optimization. They can not only sensitively capture signs, but also effectively avoid frequent false alarms due to occasional fluctuations.

[0076] After S403 control is completed, the control execution result is fed back to achieve closed-loop optimization.

[0077] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0078] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0079] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0080] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0081] It should be noted that in the claims, any reference signs placed between parentheses shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claim. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention can be implemented by means of hardware comprising several distinct components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The use of the words first, second, third etc. does not indicate any order. These words may be interpreted as names.

[0082] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0083] Obviously, those skilled in the art may make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if such modifications and variations fall within the scope of the claims and their equivalents, the present invention is intended to include such modifications and variations.

[0084] In the present invention, unless otherwise expressly specified or limited, the terms "mounted," "connected," "connect," "fixed," etc. should be understood broadly. For example, they may refer to fixed connection, detachable connection, or integration; mechanical connection or electrical connection; direct connection or indirect connection through an intermediate medium; internal communication between two components or interaction between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0085] In the description of this specification, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms should not be understood as necessarily referring to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

Claims

1. A method for cluster pressure collaborative control of spinning machines based on multi-source evidence fusion, characterized in that: include: Obtain the real-time status parameters of each spinning machine and calculate the derived status parameters of each spinning machine and its connecting area based on the real-time status parameters; Abnormal evidence is generated based on the abnormal state parameters of each spinning machine. The abnormal evidence of all spinning machines is fused through DS evidence theory to obtain the global abnormal confidence level. Dynamically generate cluster security envelopes and dynamic security thresholds of real-time status parameters based on global anomaly confidence and status parameters of each segment; Real-time status judgment and control of the spinning machine cluster are carried out through dynamic safety thresholds and global abnormality confidence.

2. The method for pressure collaborative control of a spinning machine cluster based on multi-source evidence fusion according to claim 1 is characterized in that: The step of obtaining the real-time status parameters of each spinning machine and calculating the derived status parameters of each spinning machine and its connection area based on the real-time status parameters includes: Get the radial pressure of the spinning wheel of each spinning machine , spindle torque , segmental strain , strain rate ,temperature , feed speed , workstation space coordinates ; Through the radial pressure of the spinning wheel of each spinning machine Calculate its local transient pressure change rate , the calculation formula is as follows: ， in, Indicates the radial pressure of the wheel at the current moment, Indicates the pressure value at the previous sampling moment, Indicates the sampling period; According to the adjacent distance and material flow relationship of the workstation connection area of ​​each spinning machine, the real-time state parameters of adjacent workstations are coordinated and dynamically matched, and the parameter differences and coupling pressure change rates of each workstation and its connection area are calculated.

3. The method for pressure collaborative control of a spinning machine cluster based on multi-source evidence fusion according to claim 2 is characterized in that: The method of dynamically matching the real-time state parameters of adjacent workstations based on the adjacent distances and material flow relationships of the workstation connection areas of each spinning machine and calculating the parameter differences and coupling pressure change rates of the respective workstations and their connection areas includes: The material flow coupling coefficient is calculated by the adjacent distance and the feed speed between the materials. The calculation formula is as follows: ， in, represents the material flow coupling coefficient, Indicates the maximum allowable spacing, represents the average feed speed, Represent the spatial coordinates of the i-th and j-th stations respectively, Respectively represent the feed speed of the i-th and j-th spinning machines; By local transient pressure change rate and material flow coupling coefficient Calculate the dynamic coupling pressure change rate using the following formula: ， in, Indicates the Dynamic coupling pressure change rate between stations, represents the calibrated strain-pressure coupling coefficient, represents the strain coordination deviation, The table shows the transient pressure change rate of the i-th and j-th stations. ∈[0,1].

4. The method for pressure collaborative control of a spinning machine cluster based on multi-source evidence fusion according to claim 1 is characterized in that: The abnormal evidence is formed based on the abnormal state parameters of each spinning machine, and the abnormal evidence of all spinning machines is integrated through the DS evidence theory to obtain the global abnormality confidence, which includes: Compare the real-time status parameters and derived status parameters of each spinning machine in the current cycle with the corresponding process allowable range to determine whether each parameter is abnormal and form an abnormal mark; The abnormality marks are used to construct the basic probability distribution function under the DS evidence theory, and all the DS abnormality evidence of the spinning machine are formed into an intra-group evidence set; According to the Dempster synthesis rule in DS theory, all local evidences are recursively integrated to obtain the global anomaly confidence. The global anomaly confidence is graded and the current anomaly level is output.

5. The method for pressure collaborative control of a spinning machine cluster based on multi-source evidence fusion according to claim 4 is characterized in that: The abnormality mark is used to construct the basic probability distribution function under the DS evidence theory, and the DS abnormality evidence of all spinning machines is formed into an intra-group evidence set including: Define the weights of each state parameter through expert experience and historical data; The abnormal support is calculated by the abnormal mark of each state parameter and the corresponding state parameter weight. The calculation formula is as follows: ， in, represents the abnormal state support of the i-th spinning machine, k Represents various state parameters, Indicates the k The importance weight of each parameter in anomaly detection, Indicates the k Exception indication of parameters; The uncertainty weight is adjusted for the stations with missing data. The calculation formula is as follows: ， in, represents the basic probability distribution of the data fusion set of the i-th spinning machine, represents the abnormal state support of the i-th spinning machine, represents the health status support of the i-th spinning machine, Represents the complete set of all possible states.

6. The method for pressure collaborative control of a spinning machine cluster based on multi-source evidence fusion according to claim 1 is characterized in that: The method of dynamically generating the cluster security envelope and the dynamic security threshold of the real-time status parameters based on the global anomaly confidence and the status parameters of each segment includes: Conduct statistical analysis on the real-time status parameters of each section, extract the interval boundaries of each section under normal operating conditions, and obtain the basic safety envelope; The basic safety of each segment is first dynamically corrected through the global anomaly confidence to obtain the cluster safety envelope and dynamic safety threshold.

7. The method for pressure collaborative control of a spinning machine cluster based on multi-source evidence fusion according to claim 6 is characterized in that: The basic safety of each segment is first dynamically modified by the global anomaly confidence level to obtain the cluster safety envelope and dynamic safety threshold, including: Define the adjustment factor, and the calculation formula is as follows: ， in, Indicates sensitivity, represents the anomaly confidence threshold, represents the global anomaly confidence; If the global anomaly confidence >Confidence threshold , then tighten the envelope range, the calculation formula is as follows: New boundary = original boundary × (1-λ); If the global anomaly confidence ≤Confidence Threshold , then relax the envelope range, and the calculation formula is as follows: New boundary = original boundary × (1 + λ / 2); Based on the latest cluster safety envelope, the dynamic safety threshold of the real-time status parameters of each segment is obtained.

8. The method for pressure collaborative control of a spinning machine cluster based on multi-source evidence fusion according to claim 1 is characterized in that: The real-time status judgment and control of the spinning machine cluster by using the dynamic safety threshold and global abnormality confidence level include: Through the ratio of real-time status parameters and their corresponding safety thresholds, global anomaly confidence Construct state diagnosis vector D; According to the global anomaly confidence and state diagnosis vectors for hierarchical control of spinning machine cluster operation; After the control is completed, the control execution results will be fed back to achieve closed-loop optimization.

9. The method for pressure collaborative control of a spinning machine cluster based on multi-source evidence fusion according to claim 8 is characterized in that: According to the global anomaly confidence The hierarchical control of spinning machine cluster operation based on status diagnostic vectors includes: when ≤0.3 and max(D)<1, normal automatic operation is carried out; when > 0.3 and max(D)∈[1,1.2], prioritize the real-time status parameters that exceed the dynamic safety threshold and adjust the workstation accordingly; when >0.6 or max(D)>1.5, start isolation warning and switch to manual review and fault self-diagnosis mode.

10. The method for pressure collaborative control of a spinning machine cluster based on multi-source evidence fusion according to claim 9 is characterized in that: Prioritize the real-time status parameters that exceed the dynamic safety threshold and further adjust the workstation: If the material flow coupling coefficient between this station and the adjacent station If the maximum value is greater than 0.6, the corresponding real-time status parameters of adjacent workstations are adjusted synchronously.

Images