Method and system for dynamic scheduling of veneer production tasks based on equipment state awareness

Abstract

The application provides a plate rolling machine production task dynamic scheduling method and system based on equipment state perception, relates to the technical field of production scheduling, and the method comprises the following steps: obtaining an equipment real-time database; performing feature extraction on the equipment real-time database to obtain an equipment real-time feature set; applying preset threshold rules to analyze the equipment real-time feature set, outputting a preliminary health level state, updating the equipment real-time feature set based on the preliminary health level state; realizing deep evaluation of the equipment health state based on the updated equipment real-time feature set, and obtaining an equipment comprehensive state diagnosis result; obtaining a production task, and constructing a dynamic scheduling scheme in combination with the equipment comprehensive state diagnosis result. The technical problems that the existing technology is difficult to realize efficient and reliable scheduling of the plate rolling machine production task, and is prone to sudden equipment failure, resulting in interruption of the production plan and overall efficiency reduction are solved.

Description

Technical Field

[0001] This application relates to the field of production scheduling technology, and in particular to a method and system for dynamic scheduling of production tasks of plate rolling machines based on equipment status awareness. Background Technology

[0002] Plate rolling machines, as core equipment for metal sheet forming, are widely used in shipbuilding, pressure vessels, energy equipment, rail transportation, and other fields to produce key structural components such as cylinders, cones, and curved plates. With the rapid development of high-end equipment manufacturing, the requirements for forming accuracy, efficiency, and equipment reliability in plate rolling processes are increasing. The stable operation of plate rolling machines directly affects production progress and product quality.

[0003] Currently, factories typically use fixed scheduling or experience-based scheduling methods to arrange plate rolling machine production tasks. However, these methods have significant problems: First, scheduling relies on static equipment status assessments, failing to detect real-time changes in equipment health, leading to a mismatch between task allocation and actual equipment capacity. Second, the lack of real-time monitoring of key components of the plate rolling machine results in delayed fault warnings and frequent sudden shutdowns disrupting production plans. Furthermore, scheduling optimization only considers task priority and delivery dates, without dynamically adjusting to real-time equipment status, easily leading to the accumulation of high-load tasks and accelerating equipment deterioration. The most prominent problem is the inaccurate assessment of the plate rolling machine's health status by existing scheduling methods, leading to sudden equipment failures after scheduling. This not only prevents timely task completion but also further delays subsequent production due to emergency repairs, severely impacting overall efficiency.

[0004] In summary, existing technologies are insufficient for efficient and reliable scheduling of plate rolling machine production tasks. They are prone to production plan interruptions and overall efficiency decline due to sudden equipment failures. Therefore, there is an urgent need for a solution to these problems. Through precise health assessment and intelligent task optimization, production efficiency can be maximized while ensuring stable equipment operation. Summary of the Invention

[0005] This disclosure provides a method and system for dynamic scheduling of plate rolling machine production tasks based on equipment status awareness, in order to solve the technical problems in the prior art that make it difficult to achieve efficient and reliable scheduling of plate rolling machine production tasks, and that production plans are easily interrupted and overall efficiency is reduced due to sudden equipment failures.

[0006] According to a first aspect of this disclosure, a method for dynamic scheduling of production tasks on a plate rolling machine based on equipment status awareness is provided, including:

[0007] A sensor network is deployed at key locations of the plate rolling machine to continuously collect raw physical data of the plate rolling machine's operation. The raw physical data is preprocessed to obtain a real-time equipment database. The real-time equipment database includes equipment number, timestamp, vibration information, temperature information, and current information. The preprocessing includes data cleaning and preliminary filtering.

[0008] Feature extraction is performed on the real-time database of the device to obtain a real-time feature set of the device. The feature extraction includes vibration information feature extraction, temperature information feature extraction and current information feature extraction.

[0009] The device's real-time feature set is analyzed using preset threshold rules to output a preliminary health level status, and the device's real-time feature set is updated based on the preliminary health level status.

[0010] Based on the updated real-time feature set of the equipment, a deep assessment of the equipment health status is achieved, and a comprehensive equipment status diagnosis result is obtained.

[0011] Obtain production tasks and construct a dynamic scheduling scheme based on the comprehensive equipment status diagnosis results.

[0012] According to a second aspect of this disclosure, a dynamic scheduling system for plate rolling machine production tasks based on equipment status awareness is provided, comprising:

[0013] The data acquisition module is used to deploy a sensor network at key locations of the plate rolling machine to continuously collect raw physical data of the plate rolling machine's operation. The raw physical data is preprocessed to obtain a real-time equipment database. The real-time equipment database includes equipment number, timestamp, vibration information, temperature information, and current information. The preprocessing includes data cleaning and preliminary filtering.

[0014] A multi-source data feature extraction module is used to extract features from the real-time database of the device to obtain a real-time feature set of the device. The feature extraction includes vibration information feature extraction, temperature information feature extraction and current information feature extraction.

[0015] The device health status preliminary assessment module is used to analyze the device real-time feature set by applying preset threshold rules, output a preliminary health level status, and update the device real-time feature set based on the preliminary health level status.

[0016] The equipment health deep diagnosis module is used to perform a deep assessment of the equipment health status based on the updated real-time feature set of the equipment, and obtain a comprehensive equipment status diagnosis result.

[0017] The intelligent dynamic scheduling decision module is used to acquire production tasks and construct a dynamic scheduling scheme by combining the comprehensive equipment status diagnosis results.

[0018] One or more technical solutions provided in this disclosure have at least the following technical effects or advantages: A sensor network is deployed at key locations on the plate rolling machine to continuously collect raw physical data of the machine's operation; the raw physical data is preprocessed to obtain a real-time equipment database, which includes equipment number, timestamp, vibration information, temperature information, and current information; the preprocessing includes data cleaning and preliminary filtering; feature extraction is performed on the real-time equipment database to obtain a real-time equipment feature set, which includes vibration information feature extraction, temperature information feature extraction, and current information feature extraction; a preset threshold rule is applied to analyze the real-time equipment feature set to output a preliminary health level status; the real-time equipment feature set is updated based on the preliminary health level status; a deep assessment of the equipment health status is achieved based on the updated real-time equipment feature set to obtain a comprehensive equipment status diagnosis result; production tasks are obtained, and a dynamic scheduling scheme is constructed based on the comprehensive equipment status diagnosis result. This solves the technical problems in the prior art, such as the difficulty in achieving efficient and reliable scheduling of plate rolling machine production tasks, and the susceptibility to production plan interruptions and overall efficiency decline due to sudden equipment failures. It achieves the technical effects of real-time and accurate assessment of equipment health status, dynamic optimization of production task allocation, prevention of sudden equipment failures, ensuring continuous and stable execution of production plans, and improving overall production efficiency.

[0019] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0021] Figure 1 A flowchart illustrating the dynamic scheduling method for plate rolling machine production tasks based on equipment status awareness provided in this application embodiment;

[0022] Figure 2 This is a schematic diagram of the structure of a dynamic scheduling system for plate rolling machine production tasks based on equipment status awareness, provided in an embodiment of this application.

[0023] Figure labeling: 11 Data acquisition module, 12 Multi-source data feature extraction module, 13 Preliminary equipment health status assessment module, 14 In-depth equipment health diagnosis module, 15 Intelligent dynamic scheduling decision module. Detailed Implementation

[0024] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0025] Example 1: The dynamic scheduling method for plate rolling machine production tasks based on equipment status awareness provided in this embodiment of the present disclosure is as follows: Figure 1 The methods include:

[0026] S1: Deploy a sensor network at key locations of the plate rolling machine to continuously collect raw physical data of the plate rolling machine's operation. Preprocess the raw physical data to obtain a real-time equipment database. The real-time equipment database includes equipment number, timestamp, vibration information, temperature information, and current information. The preprocessing includes data cleaning and preliminary filtering.

[0027] Furthermore, step S1 also includes:

[0028] A sensor network is constructed by integrating M sets of multi-source sensors to acquire the raw physical data of M plate rolling machines during operation. Each set of multi-source sensors includes a vibration sensor, a temperature sensor, and a current sensor.

[0029] The acquired raw physical data is cleaned, and obviously erroneous sampling points in the raw physical data are removed based on rigid rules and dynamic statistical rules;

[0030] The raw physical data after cleaning is initially filtered, and band-pass and low-pass filters are used to filter out environmental interference.

[0031] Using device number and timestamp as spatiotemporal labels, the pre-processed raw physical data is integrated to construct a real-time device database.

[0032] Specifically, a sensor network is constructed by integrating M sets of multi-source sensors to acquire vibration, temperature, and current information from M plate rolling machines during operation. Each set of multi-source sensors includes a vibration sensor, a temperature sensor, and a current sensor. The vibration sensor uses a triaxial accelerometer, fixed to the middle section of the main drive gearbox housing using a magnetic base. The axis is aligned with the input and output shafts, and the installation torque is strictly controlled at 12 N·m to ensure a secure installation and prevent deformation of the base due to excessive tightness. The temperature sensor uses a pre-embedded PT100 probe embedded in the temperature measurement hole of the main motor bearing housing and the hydraulic cylinder seal ring, directly contacting the surface of the measured component. Thermal grease is used to fill the gaps to ensure good heat conduction. The current sensor uses a Rogowski coil current transformer installed on the main motor power input line. All sensors are connected to the edge computing gateway via industrial-grade shielded cables, and clock synchronization is performed using the IEEE 1588 precision time protocol to ensure a unified time base for data from different sources, with timestamp errors controlled within 1 millisecond. Raw physical data collected directly through the sensor network is packaged and transmitted to the central server in 1-minute intervals. Each data packet contains the device number, timestamp, vibration information, temperature information, and current information.

[0033] The acquired raw physical data undergoes preprocessing, including data cleaning and preliminary filtering. Data cleaning is performed twice. The first cleaning is based on a rigid rule base established using the physical characteristics of the sensors and equipment safety thresholds. Specifically, vibration sensors have independent ±50g range thresholds set for each axis; sampling points exceeding this range are directly identified as abnormal and immediately discarded. Temperature sensors have an effective range set to 0-150℃; data exceeding this threshold is considered invalid and is discarded after secondary verification using equipment ambient temperature logs. Current information acquired from current sensors has an effective value range of 20%~130% of the rated current; zero values ​​and extreme points caused by short circuits or open circuits are discarded. After the first data cleaning, for sampling points still exhibiting abnormalities within the range, a sliding time window statistical analysis method is used to perform a second data cleaning based on dynamic statistical rules. Using a 30-second time window, the mean μ and standard deviation σ of the effective values ​​of vibration information within the window are calculated in real time. If the Z-score of the vibration amplitude at a single point exceeds ±3σ for three consecutive times, it is judged as instantaneous electromagnetic interference or mechanical collision and is discarded. The rate of change between adjacent temperature information is calculated. When the slope exceeds 5℃ / second, the start / stop status log of the linked equipment is checked. If the equipment is in operation, it is retained; if it is in standby mode, it is discarded. The moving average of the effective values ​​of current information is calculated within a 1-minute window. If the deviation at a single point exceeds the mean ±15% and continues for more than 5 cycles, it is judged as abnormal sampling and the data is discarded.

[0034] The raw physical data after cleaning undergoes preliminary filtering to remove environmental interference. For vibration information, a bandpass filter is applied to preserve the characteristic frequencies of bearings and gears while filtering out environmental interference. A fourth-order Butterworth bandpass filter is used, with the passband frequency strictly limited to 5Hz-5kHz. The low-frequency cutoff is set to 5Hz to eliminate low-frequency interference such as foundation vibration and overall equipment shaking. The high-frequency cutoff is set to 5kHz to suppress high-frequency electromagnetic noise such as motor brush noise and inverter switching harmonics. A forward and backward bidirectional filter is used to eliminate signal phase shift and ensure the accuracy of impact event timing. Mirror extension is used at the beginning and end of the data segment to avoid endpoint distortion during the filtering process. For temperature and current information, a low-pass filter is applied to preserve the thermal inertia and load fluctuation characteristics of the equipment while filtering out high-frequency noise. A Chebyshev Type I low-pass filter is used, with the cutoff frequency set to 1Hz and the passband ripple controlled within 0.5dB. Specifically, the temperature signal needs to filter out sensor circuit thermal noise and electromagnetic coupling interference while retaining the slow oil temperature change trend. The current signal needs to eliminate inverter carrier interference while retaining the motor load fluctuation characteristics. Applying a moving average to the filtered data further smooths out random fluctuations and improves trend visibility.

[0035] The raw physical data after cleaning and filtering is reorganized using equipment number and timestamp as spatiotemporal tags to form a real-time equipment database.

[0036] S2: Perform feature extraction on the real-time database of the device to obtain a real-time feature set of the device. The feature extraction includes vibration information feature extraction, temperature information feature extraction and current information feature extraction.

[0037] Furthermore, step S2 also includes:

[0038] Obtain vibration information, temperature information, and current information from the device's real-time database;

[0039] Vibration information features were extracted using time-domain and frequency-domain multidimensional analysis methods. The extracted features included the effective value of vibration, peak value of vibration, vibration kurtosis, and the proportion of energy in the high-frequency band.

[0040] Temperature information features were extracted using trend analysis methods. The extracted features included absolute temperature values ​​and temperature rise rates.

[0041] Current information features are extracted based on the electrical and mechanical coupling mechanism. The extracted features include the effective value of the current and the three-phase current imbalance.

[0042] All feature information is integrated into a structured real-time feature set for each device, organized by device number.

[0043] Specifically, the system acquires a real-time database of the equipment, extracts features from three types of information—vibration, temperature, and current—from the database, and transforms the raw data into indicators reflecting the operating status of key components of the equipment, providing a basis for subsequent health status assessments.

[0044] The main function of vibration information feature extraction is to identify mechanical faults such as imbalance, wear, looseness, asymmetry, and impact damage in bearings, gears, and shaft systems. The specific operation process includes:

[0045] Time-domain feature extraction. The vibration waveform is directly analyzed to obtain the effective vibration value, peak value, and kurtosis. The formulas are as follows:

[0046] RMS vibration value: ;

[0047] Where RMS represents the effective value of the vibration signal, used to characterize the average intensity of the vibration energy, N represents the number of sampling points, and y i This represents the amplitude of the i-th vibration signal sampling point, where i is the index value. The RMS value reflects the overall level of vibration energy and is the most basic and stable indicator for assessing the overall deterioration of equipment. A continuously rising RMS value usually indicates that a fault is developing.

[0048] Peak vibration: ;

[0049] Where Peak represents the peak value of the vibration signal, used to capture the maximum intensity of the instantaneous impact, max is a function for selecting the maximum value, and |y i | Represents the absolute value of the signal sampling point. Vibration peak values ​​can capture transient impact intensity. They are highly sensitive to localized damage. A single increase in vibration peak value may indicate an impact event.

[0050] Vibration kurtosis: ;

[0051] Where, Kurtosis is the kurtosis of the vibration signal, used to quantify the sharpness of the impact signal, N is the total number of sampling points, and y i η represents the amplitude of the i-th vibration signal sampling point, where i is the index value, η represents the signal average, and σ represents the signal standard deviation. Vibration kurtosis is extremely sensitive to impact signals and serves as an early warning indicator of faults. In the early stages of a fault, the impact signal is mixed with background vibration, and the RMS change may be small, but the vibration kurtosis value will increase significantly. As the fault progresses and impacts increase, the vibration kurtosis may actually decrease.

[0052] Frequency domain feature extraction. Vibration information is transformed from the time domain to the frequency domain using a Fast Fourier Transform (FFT) to obtain a spectrum. Frequency domain feature extraction is then performed to obtain the proportion of high-frequency energy in the vibration information, as shown in the formula:

[0053] High-frequency energy percentage: ;

[0054] Where HFR represents the high-frequency energy percentage, used to indicate the proportion of high-frequency friction or impact noise. The numerator in the formula represents high-frequency energy, the denominator represents full-frequency energy, and k represents the frequency band number. max k represents the maximum index. min Let X(f) represent the minimum index, f represent the frequency band boundary, and X(f) represent the minimum index. k ) represents the signal amplitude, and M is the window length for FFT analysis.

[0055] Specifically, to obtain the energy proportion of the high-frequency band, the frequency resolution Δf needs to be obtained first through FFT spectrum, as shown in the formula: , where f s The sampling rate is represented by Δf, and M is the window length for the FFT analysis. Based on the obtained frequency resolution Δf, combined with the determined frequency band boundary f... min with f max Calculate the frequency band number k min With k max The formula is Where f is the frequency band boundary. For example, assuming a sampling rate of 20 kHz and a window length M of 2048 for FFT analysis, Δf is approximately 9.77 Hz. f is determined based on commonly used engineering values. min For 4000Hz and f max Given a frequency of 20000Hz, k is calculated. min For 410, k max The value is 1024, therefore the formula for calculating HFR becomes... The resulting high-frequency energy percentage can be used to monitor the degree of wear and degradation. As component wear intensifies, the friction and impact noise energy in the high-frequency band tends to increase, leading to a rise in this percentage.

[0056] The main function of temperature information feature extraction is to identify problems such as bearing overheating, motor overheating, lubrication failure, poor cooling, and uneven load. Specifically, the main features extracted include absolute temperature value and temperature rise rate, with the following formulas:

[0057] Absolute temperature value: ;

[0058] Among them, T current Represents the absolute temperature value, P represents the moving average window size, and T represents the absolute temperature value. j Represents the original sampled value of the sensor, j is the index value, and C cal This represents the correction factor, which is set to 0.385 in this case. offset This represents environmental compensation. Absolute temperature value is the most direct basis for overheating alarms. The operating status of the plate rolling machine can be determined by setting different alarm threshold levels.

[0059] Temperature rise rate: ;

[0060] Among them, R rise T represents the rate of temperature rise. t T represents the current temperature, Δt is the time window, and T t -Δt represents the historical temperature value. The rate of temperature rise can detect anomalies at an earlier stage. During normal equipment startup or load increase, the temperature rise is gradual. If the rate of temperature rise suddenly accelerates, it often indicates that a serious problem is about to occur, such as sudden lubrication failure, cooling system failure, or severe friction.

[0061] The main function of current information feature extraction is to identify electrical faults in motors, power quality problems, and abnormal mechanical loads reflected in motor current. Specifically, the main features extracted include the effective value of the current and the three-phase current imbalance, with the following formulas:

[0062] RMS current value: ;

[0063] Among them, I RMS The effective value of the current is a core indicator of the load size. N represents the number of sampling points within the cycle, Ir represents the instantaneous sampled current value, and r is the index value. The instantaneous sampled current value reflects the motor load. If the effective current value continuously exceeds the rated value, it indicates that the motor is overloaded. Combined with vibration and temperature, the cause of the overload can be determined.

[0064] Three-phase current imbalance: ;

[0065] Among them, I A I B I C Representing the effective value of the three-phase current, I is calculated separately according to the formula for the effective value of current. avg The three-phase current unbalance represents the average three-phase current, while unbalance represents the degree of imbalance. The three-phase current unbalance can be used to diagnose electrical faults such as power supply voltage imbalance, motor winding asymmetry, and rotor bar breakage.

[0066] All feature information extracted above is integrated into an information table, and index values ​​are assigned according to the plate rolling machine number. Simultaneously, timestamps are added to the original data to obtain the equipment's real-time feature set. The feature set includes vibration RMS values, vibration peak values, vibration kurtosis, and high-frequency energy percentage data for vibration information; absolute temperature values ​​and temperature rise rate data for temperature information; and RMS current values ​​and three-phase current imbalance data for current information, all arranged in order. Furthermore, it is categorized using the equipment number and timestamp as spatiotemporal labels, storing all feature information for M plate rolling machines per minute.

[0067] S3: Analyze the real-time feature set of the device using preset threshold rules, output a preliminary health level status, and update the real-time feature set of the device based on the preliminary health level status;

[0068] Furthermore, step S3 also includes:

[0069] Set threshold limits for each feature in the real-time feature set of the device, including normal limits and fault limits;

[0070] The comprehensive health index is calculated based on the device's real-time feature set and threshold limits. The specific formula is as follows:

[0071] ;

[0072] Where HI(t) represents the comprehensive health index at the current moment, i is the index value, and U i (t) represents the current actual value of the i-th feature, L i,normal L represents the normal limit of the i-th feature. i,fault The fault limit representing the i-th feature, ω i The weight represents the i-th feature, and α represents the non-linear adjustment coefficient.

[0073] The preliminary health status of the equipment at the current moment is determined based on the comprehensive health index. When HI(t)≥0.85, the preliminary health status is determined to be Level 1; when 0.6≦HI(t)<0.85, the preliminary health status is determined to be Level 2 warning; and when HI(t)<0.6, the preliminary health status is determined to be Level 3 fault.

[0074] Equipment whose initial health status is determined to be a level three fault is shut down. At the same time, the real-time feature set of the equipment is updated, and all data under the shut-down equipment number is deleted.

[0075] Specifically, the system uses preset threshold rules to analyze the feature data in the real-time feature set of the equipment to determine the health status of the plate rolling machine in terms of vibration, temperature, and current.

[0076] Regarding vibration characteristic information, the normal limit for vibration effective value (RMS) is set to 2.5g, the fault limit to 4.0g, the normal limit for vibration peak value (Peak) is set to 5.0g, the fault limit to 8.0g, the normal limit for vibration kurtosis is set to 3.5, the fault limit to 4.5, and the normal limit for high-frequency energy ratio (HFR) is set to 8%, the fault limit to 15%.

[0077] Regarding temperature information, the absolute temperature value T is set. current The normal temperature limit is 70°C, the fault temperature limit is 85°C, and the set temperature rise rate R is... riseThe normal limit is 0.2°C / min, and the fault limit is 0.5°C / min.

[0078] Regarding current information, the effective value of the current I is set. RMS The normal limit is 1.1I, where I is the rated current. The fault limit is 1.2I. The normal limit for three-phase current unbalance is set to 5%, and the fault limit is set to 10%.

[0079] The overall health index of each plate rolling machine is calculated based on the threshold values ​​set for the above parameters. The specific formula is as follows:

[0080] ;

[0081] Where HI(t) represents the comprehensive health index at the current moment, i is the index value, and U i (t) represents the current actual value of the i-th feature, L i,normal L represents the normal limit of the i-th feature. i,fault The fault limit representing the i-th feature, ω i The weights representing the i-th feature satisfy all ω i The sum of these values ​​is 1, where α represents the nonlinear adjustment coefficient, used to control the sensitivity after exceeding the threshold, and is typically set to 1.5.

[0082] The specific weighting values ​​in the formula are ω1=0.15, ω2=0.10, ω3=0.20, ω4=0.10, ω5=0.15, ω6=0.15, ω7=0.10, and ω8=0.05. Here, ω1 represents the effective vibration value weight, reflecting the overall mechanical load; ω2 represents the peak vibration value weight, reflecting sensitivity to instantaneous impacts; ω3 represents the vibration kurtosis, a core indicator of early failures; ω4 represents the proportion of high-frequency energy, characterizing the long-term trend of wear and deterioration; ω5 represents the absolute temperature value, directly related to equipment safety; ω6 represents the temperature rise rate, crucial for predicting sudden failures; ω7 represents the effective current value, reflecting the electrical load; and ω8 represents the three-phase current imbalance, reflecting sensitivity to motor symmetry.

[0083] Finally, the preliminary health level of the plate rolling machine is determined based on the calculated comprehensive health index. When HI(t)≥0.85, the preliminary health level is determined to be Level 1; when 0.6≦HI(t)<0.85, the preliminary health level is determined to be Level 2 warning; and when HI(t)<0.6, the preliminary health level is determined to be Level 3 fault.

[0084] When the initial health level is determined to be a level three fault, the operation of the plate rolling machine is stopped immediately, a maintenance alarm is activated, the plate rolling machine is no longer included in the subsequent scheduling process, and all data under the stopped equipment number in the equipment real-time feature set is deleted.

[0085] S4: Based on the updated real-time feature set of the equipment, a deep assessment of the equipment health status is achieved to obtain a comprehensive equipment status diagnosis result;

[0086] Furthermore, step S4 also includes:

[0087] Read the updated real-time feature set of the device, calculate the moving average and trend slope of each feature information, and construct a feature enhancement vector set;

[0088] The threshold limits of each feature are dynamically adjusted based on the moving average, and the weight allocation of each feature is dynamically adjusted based on the trend slope.

[0089] Based on the adjusted threshold limits, membership functions for the three states are defined, and the membership degree of the current feature is calculated. The membership functions include a trapezoidal function for the healthy state, a triangular function for the warning state, and a Z-shaped function for the fault state. The specific expressions are as follows:

[0090] healthy: ;

[0091] Warning: ;

[0092] Fault: ;

[0093] Where, μ normal (x) represents the membership degree of feature x to health status, μ warning (x) represents the membership degree of feature x to the warning status, μ fault (x) represents the membership degree of feature x to the fault state, where x represents the original value of the feature, L normal L represents the normal boundary. warning L represents the warning threshold. mid L represents the decay limit. fault Represents the fault limit, where L warning =L normal +0.3(L fault- L normal ), L mid =(L normal +L fault ) / 2;

[0094] The final confidence level of the device in the three states is calculated using a weighted average method. The specific calculation formula is as follows:

[0095] ;

[0096] Where Belief(A) represents the overall confidence level of the device in state A, ω i μ represents the feature weight of the i-th feature after dynamic adjustment based on the trend slope. i (A) represents the membership degree of the i-th feature;

[0097] Based on the confidence level of the final equipment in the three areas of health, early warning, and fault, the overall equipment status diagnosis result of the plate rolling machine at this time is determined.

[0098] Specifically, the updated real-time feature set of the devices is read, and all feature data for each device over the previous 60 minutes based on the current time are obtained. For each type of feature information, its moving average and trend slope are calculated, and finally, the feature enhancement vector is obtained. The formulas are as follows:

[0099] Moving average: ;

[0100] Among them, Avg i The moving average of the i-th feature represents the average feature information over the past 60 minutes at the current time t, where i is the index, t is the time, m represents the time sequence number, and u i (m) represents the original value of the i-th feature at time m.

[0101] Trend slope: ;

[0102] Among them, Slope i The trend slope represents the i-th feature, where i is the index value and t is the slope. m Represents a timestamp, u i (m) represents the original value of the i-th feature at time m, and the trend slope can be used to determine whether the feature continues to deteriorate.

[0103] Based on the original feature value, moving average, and trend slope, a feature enhancement vector is constructed for each feature. Taking the effective vibration value (RMS) as an example, if the current RMS value is 3.2g, the calculated moving average is 2.8g, and the trend slope is 0.01g / min, then the final feature enhancement vector generated for the effective vibration value is [original RMS=3.2, mean=2.8, slope=0.01]. Therefore, each plate rolling machine will ultimately obtain 8 feature enhancement vectors for the current time, corresponding to the eight original feature data, thus constructing a feature enhancement vector set.

[0104] The threshold limits of each feature are dynamically adjusted based on the moving average contained in the feature enhancement vector set, and the weight allocation of each feature data is also dynamically adjusted based on the trend slope. Based on the adjusted threshold limits and weight allocation, fuzzy membership degree and equipment confidence degree are calculated for the feature data to determine the comprehensive equipment status diagnosis result.

[0105] Based on the feature information, three membership functions are defined for each state: a trapezoidal function for a healthy state, a triangular function for a warning state, and a Z-shaped function for a fault state. The threshold range for each function is defined according to dynamically adjusted threshold limits. For example, taking the effective vibration value as an example, if the moving average is calculated to be 2.8g, the threshold limits are dynamically adjusted, with the new normal limit at 2.8g and the new fault limit at 4.0 + (2.8 - 2.5) = 4.3g. Based on these two threshold limits, a warning limit and an attenuation limit are further set, ultimately defining the specific thresholds for the three types of functions. The membership degree of the current feature is calculated according to the defined membership functions, with the specific expression as follows:

[0106] healthy: ;

[0107] Warning: ;

[0108] Fault: ;

[0109] Where, μ normal (x) represents the membership degree of feature x to health status, μ warning (x) represents the membership degree of feature x to the warning status, μ fault (x) represents the membership degree of feature x to the fault state, where x represents the original value of the feature, L normal L represents the normal boundary. warning L represents the warning threshold. mid L represents the decay limit. fault Represents the fault limit, where L warning =L normal +0.3(L fault- L normal ), L mid =(L normal +L fault ) / 2.

[0110] The fuzzy membership degrees of multiple features are transformed into a comprehensive equipment status diagnosis result. First, the membership degrees are normalized to facilitate subsequent calculations. Second, a weighted average method is used to calculate the final confidence level of the equipment in three states. Feature weights are introduced during the calculation process; these weights are no longer those used in the previous comprehensive health index calculation, but rather updated weights based on the trend slope calculation results. The specific calculation formula is as follows:

[0111] ;

[0112] Where Belief(A) represents the overall confidence level of the device in state A, ω i μ represents the feature weight of the i-th feature. i (A) represents the membership degree of the i-th feature.

[0113] Based on the confidence level of the final equipment in the three areas of health, early warning, and fault, the overall equipment status diagnosis result of the plate rolling machine at this time is determined.

[0114] S5: Obtain production tasks and construct a dynamic scheduling scheme based on the comprehensive equipment status diagnosis results.

[0115] Furthermore, step S5 also includes:

[0116] Obtain the production task list to be scheduled, which includes task number, processing accuracy requirement, estimated time, delivery date and priority attribute;

[0117] Based on the comprehensive equipment status diagnosis results, perform preliminary task filtering to eliminate tasks that no equipment can complete at this stage, retain other tasks, and generate a candidate task set;

[0118] Obtain the candidate task set, and for the tasks that pass the initial screening, calculate their overall matching score with the device. The specific formula is as follows:

[0119] ;

[0120] Wherein, Score represents the overall matching degree between production tasks and equipment, S Precision The degree of fit between quantitative task requirements and equipment capabilities is specifically measured by the following values: Where TaskAccuracy represents task accuracy, and EquipmentPrecision represents device accuracy; S Health This reflects the workload that the equipment can safely handle under its current condition; the specific value is... Where E represents health status; S DeliveryTime This indicates that tasks with close deadlines will be processed first, with the specific value being... Where CurrentTime represents the current time, and DeliveryDeadline represents the delivery date; S Cost The representative considered economic constraints such as energy and mold changing, and the value was 0.8 during peak processing hours and 1.0 during off-peak hours;

[0121] A genetic algorithm is used to optimize task allocation and time scheduling. Chromosome encoding is performed based on candidate task sets and device information to determine the gene structure. Each gene represents a task allocation unit, which includes device number, task number, start time, and load percentage.

[0122] Initialize the population, randomly generate 100 scheduling schemes, and introduce the previously calculated comprehensive matching score;

[0123] All schemes after introducing comprehensive matching degree are selected, crossover and mutation operations are performed. The selection operation retains the top 30% of individuals with fitness to directly enter the next generation, and the remaining 70% are selected by roulette. The crossover operation randomly selects two parent individuals and exchanges some gene sequences. The mutation operation randomly adjusts the device number, start time or load of a certain gene.

[0124] New populations are generated through selection, crossover, and mutation operations. Iterative calculations are performed to gradually improve the population until convergence or 50 iterations are reached, ultimately outputting a dynamic scheduling scheme.

[0125] Specifically, a list of production tasks to be scheduled is obtained based on the factory's production orders. This list includes task number, processing accuracy requirements, estimated time, delivery date, and priority attributes. The completeness and timeliness of all production tasks are verified, and expired or unexecutable tasks are removed to ensure the reliability of the input data.

[0126] Based on the comprehensive equipment status diagnosis results, preliminary task filtering is performed to identify tasks that the equipment can currently perform, while tasks that no equipment can currently complete are eliminated. Specifically: healthy equipment can handle all tasks, but the load needs to be monitored; equipment with warnings needs to have its task type and load restricted, prohibiting tasks with processing precision exceeding the equipment's current capabilities; faulty equipment is only allowed to perform low-priority tasks or be shut down for maintenance. This stage is implemented through a rule engine. Finally, a candidate task set is generated based on the preliminary screening results.

[0127] Obtain the candidate task set, and for the tasks that pass the initial screening, calculate their overall matching score with the device. The specific formula is as follows:

[0128] ;

[0129] Wherein, Score represents the overall matching degree between production tasks and equipment, S Precision The degree of fit between quantitative task requirements and equipment capabilities is specifically measured by the following values: Where TaskAccuracy represents task accuracy, and EquipmentPrecision represents device accuracy; S Health This reflects the workload that the equipment can safely handle under its current condition; the specific value is... Where E represents health status; S DeliveryTime This indicates that tasks with close deadlines will be processed first, with the specific value being... Where CurrentTime represents the current time, and DeliveryDeadline represents the delivery date; S Cost The value represents 0.8 during peak processing hours and 1.0 during off-peak hours, taking into account economic constraints such as energy and mold changing.

[0130] A genetic algorithm is used to optimize task allocation and scheduling. The specific steps are as follows: Chromosome encoding is performed based on production tasks and equipment information to determine the gene structure. Each gene represents a task allocation unit, including equipment number, task number, start time, and load percentage. The population is initialized, and 100 scheduling schemes are randomly generated. Each scheme must meet the following requirements: equipment load does not exceed the upper limit allowed by the health status, task times do not overlap, and materials are ready. For each scheme involving equipment and task numbers, the previously calculated comprehensive matching score is introduced. All schemes after introducing the comprehensive matching score are subjected to selection, crossover, and mutation operations. The selection operation retains individuals with the top 30% fitness to directly enter the next generation, and the remaining 70% are selected by roulette wheel selection. The crossover operation randomly selects two parent individuals and exchanges part of their gene sequences. The mutation operation randomly adjusts the equipment number, start time, or load of a certain gene. A new population is generated through selection, crossover, and mutation operations, and iterative calculations are performed to gradually improve the scheme until convergence or 50 generations are reached. Finally, the Pareto optimal solution set is output, containing multiple non-dominated scheduling schemes for decision-making. Staff members choose the scheme that best suits the actual production situation.

[0131] Example 2: Based on the same inventive concept as the dynamic scheduling method for plate rolling machine production tasks based on equipment status awareness in the aforementioned examples, this application also provides a dynamic scheduling system for plate rolling machine production tasks based on equipment status awareness. Please refer to the appendix. Figure 2 The system includes:

[0132] Data acquisition module 11 is used to deploy a sensor network at key locations of the plate rolling machine to continuously collect raw physical data of the plate rolling machine operation, preprocess the raw physical data to obtain a real-time equipment database, the real-time equipment database includes equipment number, timestamp, vibration information, temperature information and current information, and the preprocessing includes data cleaning and preliminary filtering.

[0133] Multi-source data feature extraction module 12 is used to extract features from the real-time database of the device to obtain a real-time feature set of the device. The feature extraction includes vibration information feature extraction, temperature information feature extraction, and current information feature extraction. Obtaining the real-time feature set of the device further includes:

[0134] Obtain vibration information, temperature information, and current information from the device's real-time database;

[0135] Vibration information features were extracted using time-domain and frequency-domain multidimensional analysis methods. The extracted features included the effective value of vibration, peak value of vibration, vibration kurtosis, and the proportion of energy in the high-frequency band.

[0136] Temperature information features were extracted using trend analysis methods. The extracted features included absolute temperature values ​​and temperature rise rates.

[0137] Current information features are extracted based on the electrical and mechanical coupling mechanism. The extracted features include the effective value of the current and the three-phase current imbalance.

[0138] All feature information is integrated into a structured real-time feature set for the device, categorized by device number.

[0139] The device health status preliminary assessment module 13 is used to analyze the device's real-time feature set by applying preset threshold rules, output a preliminary health level status, and update the device's real-time feature set based on the preliminary health level status. The outputting of the preliminary health level status and the updating of the device's real-time feature set based on the preliminary health level status include:

[0140] Set threshold limits for each feature in the real-time feature set of the device, including normal limits and fault limits;

[0141] The comprehensive health index is calculated based on the device's real-time feature set and threshold limits. The specific formula is as follows:

[0142] ;

[0143] Where HI(t) represents the comprehensive health index at the current moment, i is the index value, and U i (t) represents the current actual value of the i-th feature, L i,normal L represents the normal limit of the i-th feature. i,fault The fault limit representing the i-th feature, ω i The weight represents the i-th feature, and α represents the non-linear adjustment coefficient.

[0144] The preliminary health status of the equipment at the current moment is determined based on the comprehensive health index. When HI(t)≥0.85, the preliminary health status is determined to be Level 1; when 0.6≦HI(t)<0.85, the preliminary health status is determined to be Level 2 warning; and when HI(t)<0.6, the preliminary health status is determined to be Level 3 fault.

[0145] Equipment whose initial health status is determined to be a level three fault will be shut down. At the same time, the real-time feature set of the equipment will be updated and all data under the shut-down equipment number will be deleted.

[0146] The equipment health deep diagnosis module 14 is used to perform a deep assessment of the equipment health status based on the updated real-time feature set of the equipment, and obtain a comprehensive equipment status diagnosis result.

[0147] The intelligent dynamic scheduling decision module 15 is used to acquire production tasks and construct a dynamic scheduling scheme based on the comprehensive equipment status diagnosis results. Acquiring production tasks and constructing a dynamic scheduling scheme based on the comprehensive equipment status diagnosis results includes:

[0148] Obtain the production task list to be scheduled, which includes task number, processing accuracy requirement, estimated time, delivery date and priority attribute;

[0149] Based on the comprehensive equipment status diagnosis results, perform preliminary task filtering to eliminate tasks that no equipment can complete at this stage, retain other tasks, and generate a candidate task set;

[0150] Obtain the candidate task set, and for the tasks that pass the initial screening, calculate their overall matching score with the device. The specific formula is as follows:

[0151] ;

[0152] Wherein, Score represents the overall matching degree between production tasks and equipment, S Precision The degree of fit between quantitative task requirements and equipment capabilities is specifically measured by the following values: Where TaskAccuracy represents task accuracy, and EquipmentPrecision represents device accuracy; S Health This reflects the workload that the equipment can safely handle under its current condition; the specific value is... Where E represents health status; S DeliveryTime This indicates that tasks with close deadlines will be processed first, with the specific value being... Where CurrentTime represents the current time, and DeliveryDeadline represents the delivery date; S Cost The representative considered economic constraints such as energy and mold changing, and the value was 0.8 during peak processing hours and 1.0 during off-peak hours;

[0153] A genetic algorithm is used to optimize task allocation and time scheduling. Chromosome encoding is performed based on candidate task sets and device information to determine the gene structure. Each gene represents a task allocation unit, which includes device number, task number, start time, and load percentage.

[0154] Initialize the population, randomly generate 100 scheduling schemes, and introduce the previously calculated comprehensive matching score;

[0155] All schemes after introducing comprehensive matching degree are selected, crossover and mutation operations are performed. The selection operation retains the top 30% of individuals with fitness to directly enter the next generation, and the remaining 70% are selected by roulette. The crossover operation randomly selects two parent individuals and exchanges some gene sequences. The mutation operation randomly adjusts the device number, start time or load of a certain gene.

[0156] New populations are generated through selection, crossover, and mutation operations. Iterative calculations are performed to gradually improve the population until convergence or 50 iterations are reached, ultimately outputting a dynamic scheduling scheme.

[0157] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0158] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for dynamic scheduling of production tasks of a veneer lathe based on awareness of the state of the equipment, characterized by, The method comprises: Lay out a sensor network at key positions of a plate rolling machine to continuously collect original physical data of the operation of the plate rolling machine, pre-process the original physical data to obtain a device real-time database, the device real-time database comprising a device number, a time stamp, vibration information, temperature information and current information, the pre-processing comprising data cleaning and preliminary filtering; Extract features from the device real-time database to obtain a device real-time feature set, the feature extraction comprising vibration information feature extraction, temperature information feature extraction and current information feature extraction, wherein obtaining the device real-time feature set further comprises: Obtaining the vibration information, the temperature information and the current information in the device real-time database; Extracting vibration information features by time domain and frequency domain multi-dimensional analysis methods, the extracted feature information comprising a vibration effective value, a vibration peak value, a vibration kurtosis and a high frequency band energy proportion; Extracting temperature information features by a trend analysis method, the extracted feature information comprising an absolute temperature value and a temperature rise rate; Extracting current information features based on an electrical and mechanical coupling mechanism, the extracted feature information comprising a current effective value and a three-phase current unbalance degree; Integrating all the feature information into a structured device real-time feature set according to the device number; Analyzing the device real-time feature set by applying preset threshold rules to output a preliminary health level state, updating the device real-time feature set based on the preliminary health level state, wherein updating the device real-time feature set based on the preliminary health level state comprises: Setting threshold limits for each feature in the device real-time feature set, including normal limits and fault limits; Calculating a comprehensive health index according to the device real-time feature set and the threshold limits, the specific formula being: ； wherein, HI(t) represents the comprehensive health index at the current time, i is an index value, U i (t) represents the current actual value of the i-th feature, L i,normal represents the normal limit of the i-th feature, L i,fault represents the failure limit of the i-th feature, ω i represents the weight of the i-th feature, and α represents a nonlinear adjustment coefficient. Determining the preliminary health level state of the device at the current time according to the comprehensive health index, determining the preliminary health level state as a first health level when HI(t)≥0.85, determining the preliminary health level state as a second early warning level when 0.6≦HI(t)<0.85, and determining the preliminary health level state as a third fault level when HI(t)<0.6; Stopping the device whose preliminary health level state is determined as the third fault level, and updating the device real-time feature set by deleting all the data under the device number of the stopped device; Realizing deep evaluation of the health state of the device based on the updated device real-time feature set to obtain a device comprehensive state diagnosis result; Obtaining a production task and constructing a dynamic scheduling scheme in combination with the device comprehensive state diagnosis result, wherein obtaining the production task and constructing the dynamic scheduling scheme in combination with the device comprehensive state diagnosis result comprises: Obtaining a production task list to be scheduled, the production task list comprising a task number, a processing precision requirement, a predicted time consumption, a delivery period and a priority attribute; Performing preliminary task filtering according to the device comprehensive state diagnosis result to eliminate tasks that cannot be completed by any device at the present stage and retain other tasks to generate a candidate task set; Obtaining the candidate task set, calculating a comprehensive matching degree Score of a task that passes the preliminary screening with a device, the specific formula being as follows: ； wherein Score represents the comprehensive matching degree between production tasks and equipment, S Precision quantifies the matching degree between task requirements and equipment capabilities, and the specific value is wherein TaskAccuracy represents task accuracy, and EquipmentPrecision represents equipment accuracy; S Health reflects the task intensity that the equipment can safely bear in the current state, and the specific value is wherein E is the health state, and Belief(E) represents the confidence of the equipment as a whole in the health state E; S DeliveryTime represents the task that is preferentially processed near the deadline, and the specific value is wherein CurrentTime represents the current time, and DeliveryDeadline represents the delivery deadline; S Cost represents the economic constraint of energy and mold change, and the value is 0.8 during peak processing period and 1.0 during non-peak processing period; The genetic algorithm is used to optimize task allocation and time scheduling, chromosome coding is performed based on a candidate task set and equipment information, a gene structure is determined, each gene represents a task allocation unit, and contains equipment number, task number, start time, and load percentage; A population is initialized, 100 scheduling schemes are randomly generated, and a previously calculated comprehensive matching degree Score is introduced; Selection, crossover, and mutation operations are performed on all schemes after introducing the comprehensive matching degree. The selection operation retains the Top 30% of individuals with the highest fitness to directly enter the next generation, and the remaining 70% are selected by roulette. The crossover operation randomly selects two parent individuals and exchanges part of the gene sequence. The mutation operation randomly adjusts the equipment number, start time, or load of a gene; A new population is generated through selection, crossover, and mutation operations, and iterative calculations are performed to gradually improve the scheduling scheme until convergence or 50 iterations are reached. Finally, the dynamic scheduling scheme is output.

2. The method for dynamic scheduling of veneer peeling machine production tasks based on awareness of the state of the equipment as claimed in claim 1, characterized in that, The real-time database of the equipment is obtained, including: M sets of multi-source sensors are integrated to construct a sensor network, and M raw physical data of the plate rolling machine during operation are obtained. Each set of multi-source sensors includes a vibration sensor, a temperature sensor, and a current sensor. The obtained raw physical data is cleaned based on rigid rules and dynamic statistical rules to remove obviously erroneous sampling points in the raw physical data. The cleaned raw physical data is preliminarily filtered by applying band-pass filtering and low-pass filtering to filter environmental interference. The real-time database of the equipment is constructed by integrating the preprocessed raw physical data based on the equipment number and timestamp as spatiotemporal labels.

3. The method for dynamic scheduling of veneer peeling machine production tasks based on awareness of the state of the equipment as claimed in claim 1, characterized in that, The comprehensive state diagnosis result of the equipment is obtained, including: The updated real-time feature set of the equipment is read, the moving average and trend slope of each feature information are calculated, and a feature enhancement vector set is constructed. The threshold limits of each feature are dynamically adjusted based on the moving average, and the weight distribution of each feature is dynamically adjusted based on the trend slope. Three state membership functions are set based on the adjusted threshold limits to calculate the membership degree of the current feature. The membership functions include a trapezoidal function for the healthy state, a triangular function for the warning state, and a Z-shaped function for the fault state. The specific expressions are: health: ; Warning: ; Fault: ; where μ normal (x) represents the membership of feature x to the healthy state, μ warning (x) represents the membership of feature x to the early warning state, μ fault (x) represents the membership of feature x to the failure state, x represents the original value of the feature, L normal represents the adjusted normal limit, L warning represents the adjusted early warning limit, L mid represents the decay limit, L fault represents the failure limit, where L warning = L normal + 0.3(L fault- L normal ), L mid = (L normal + L fault ) / 2; The final confidence of the equipment in the three states is calculated by the weighted average method, and the specific calculation formula is: ； where Belief(A) represents the overall confidence of the device in state A, ω i represents the dynamically adjusted feature weight of the i-th feature according to the trend slope, μ A (i) represents the membership degree of the i-th feature; The comprehensive state diagnosis result of the plate rolling machine is determined based on the confidence of the equipment in the healthy, warning, and fault states.

4. A log banding machine production task dynamic scheduling system based on equipment state awareness, characterized in that, The system is used to implement the plate rolling machine production task dynamic scheduling method based on equipment state perception according to any one of claims 1 to 3, and the system comprises: A data acquisition module is used to lay out a sensor network at key positions of the plate rolling machine, continuously acquire raw physical data of the plate rolling machine during operation, and preprocess the raw physical data to obtain a real-time database of the equipment. The real-time database of the equipment includes equipment number, timestamp, vibration information, temperature information, and current information. The preprocessing includes data cleaning and preliminary filtering. A multi-source data feature extraction module is configured to perform feature extraction on the equipment real-time database to obtain an equipment real-time feature set, wherein the feature extraction includes vibration information feature extraction, temperature information feature extraction, and current information feature extraction, and wherein obtaining the equipment real-time feature set further includes: obtaining vibration information, temperature information, and current information in the equipment real-time database; extracting vibration information features by a time-domain and frequency-domain multi-dimensional analysis method, and the extracted feature information includes a vibration effective value, a vibration peak value, a vibration kurtosis, and a high-frequency energy proportion; extracting temperature information features by a trend analysis method, and the extracted feature information includes an absolute temperature value and a temperature rise rate; extracting current information features based on an electrical and mechanical coupling mechanism, and the extracted feature information includes a current effective value and a three-phase current unbalance degree; integrating all the feature information into a structured equipment real-time feature set according to equipment numbers; An equipment health state preliminary evaluation module is configured to analyze the equipment real-time feature set by applying preset threshold rules to output a preliminary health level state, and update the equipment real-time feature set based on the preliminary health level state, wherein outputting the preliminary health level state and updating the equipment real-time feature set based on the preliminary health level state includes: setting threshold limits for each feature in the equipment real-time feature set, including normal limits and fault limits; calculating a comprehensive health index according to the equipment real-time feature set and the threshold limits, and the specific formula is: ； wherein, HI(t) represents the comprehensive health index at the current time, i is an index value, U i (t) represents the current actual value of the i-th feature, L i,normal represents the normal limit of the i-th feature, L i,fault represents the failure limit of the i-th feature, ω i represents the weight of the i-th feature, and α represents a non-linear adjustment coefficient. judging the preliminary health level state of the equipment at the current time according to the comprehensive health index, and when HI(t)≥0.85, the preliminary health level state is judged to be a first health level, when 0.6≦HI(t)<0.85, the preliminary health level state is judged to be a second early warning level, and when HI(t)<0.6, the preliminary health level state is judged to be a third fault level; stopping the equipment whose preliminary health level state is judged to be the third fault level, and updating the equipment real-time feature set and deleting all data under the equipment number of the stopped equipment; An equipment health deep diagnosis module is configured to realize deep evaluation of the equipment health state based on the updated equipment real-time feature set to obtain an equipment comprehensive state diagnosis result; An intelligent dynamic scheduling decision module is configured to obtain production tasks, and construct a dynamic scheduling scheme in combination with the equipment comprehensive state diagnosis result, wherein obtaining the production tasks and constructing the dynamic scheduling scheme in combination with the equipment comprehensive state diagnosis result includes: obtaining a production task list to be scheduled, and the production task list includes task numbers, processing precision requirements, estimated time consumptions, delivery periods, and priority attributes; performing preliminary task filtering according to the equipment comprehensive state diagnosis result, and removing tasks that cannot be completed by any equipment at the present stage, retaining other tasks, and generating a candidate task set; obtaining the candidate task set, and calculating a comprehensive matching degree Score of a task that passes the preliminary screening with an equipment, and the specific formula is as follows: ； wherein Score represents the comprehensive matching degree between the production task and the equipment, S Precision quantifies the matching degree between the task requirement and the equipment capability, and the specific value is wherein TaskAccuracy represents the task accuracy, and EquipmentPrecision represents the equipment accuracy; S Health reflects the task intensity that the equipment can safely bear in the current state, and the specific value is wherein E is the health state, and Belief(E) represents the confidence of the equipment as a whole in the health state E; S DeliveryTime represents the task that is preferentially processed near the deadline, and the specific value is wherein CurrentTime represents the current time, and DeliveryDeadline represents the delivery deadline; S Cost represents the economic constraint of energy and mold change, and the value is 0.8 in the peak period of processing and 1.0 in the non-peak period; Genetic algorithm is used to optimize task allocation and time scheduling. Chromosome coding is based on candidate task set and equipment information to determine gene structure. Each gene represents a task allocation unit, including equipment number, task number, start time and load percentage. Initialize population, randomly generate 100 scheduling schemes, and introduce the previously calculated comprehensive matching degree Score. Select, cross and mutate all schemes after introducing the comprehensive matching degree. Selection operation retains the top 30% of individuals with fitness to directly enter the next generation. The remaining 70% is selected by roulette. Cross operation randomly selects two parent individuals and exchanges part of the gene sequence. Mutation operation randomly adjusts the equipment number, start time or load of a gene. New population is generated through selection, cross and mutation operations. Iterative calculation is performed to gradually improve until convergence or 50 iterations are reached. Finally, the dynamic scheduling scheme is output.

Images