Production equipment intelligent regulation and control method and system based on real-time feedback

Abstract

The embodiment of the invention provides a production equipment intelligent regulation and control method and system based on real-time feedback, and the method comprises the steps: collecting a real-time monitoring data set in the operation process of production equipment, including equipment operation parameters and environment correlation parameters, carrying out the dynamic feature extraction of the real-time monitoring data set, and carrying out the dynamic feature extraction of the real-time monitoring data set; obtaining an equipment operation feature set and an environment association feature set; and then real-time operation state information of the production equipment is identified according to the incidence relation of the equipment parameter adjustment instruction and the environment adaptation instruction, a dynamic regulation and control strategy set containing the equipment parameter adjustment instruction and the environment adaptation instruction is generated based on the real-time operation state information, and finally, the operation parameters of the production equipment are dynamically adjusted according to the dynamic regulation and control strategy set. And a real-time monitoring data set is updated, so that intelligent and accurate regulation and control of production equipment can be realized, and the production efficiency and the product quality are improved.

Description

Technical Field

[0001] The present invention relates to the field of smart factory technology, and in particular to a method and system for intelligently controlling production equipment based on real-time feedback. Background Art

[0002] In modern industrial production, the stable operation and efficient output of production equipment are crucial to a company's economic benefits and competitiveness. With the continuous improvement of industrial automation, production equipment has become increasingly complex, and its operation is affected by a variety of factors, including the equipment's own operating parameters and those related to the surrounding environment.

[0003] Currently, traditional production equipment control methods mainly rely on manual experience or pre-set fixed programs. Manual experience control methods have strong subjectivity and difficulty in coping with complex and changing production environments, while fixed program control methods lack flexibility and cannot be adjusted in time according to the real-time operating status of the equipment and environmental changes. This leads to problems such as low efficiency, excessive energy consumption, and unstable product quality during the operation of production equipment, making it difficult to meet the requirements of modern industrial production for high efficiency, energy saving, and high quality. Therefore, there is a need for a method that can sense the operating status of production equipment and environmental changes in real time and perform intelligent control accordingly to improve the operating efficiency of production equipment and product quality. Summary of the Invention

[0004] In view of this, an object of an embodiment of the present invention is to provide a method and system for intelligently controlling production equipment based on real-time feedback.

[0005] According to one aspect of an embodiment of the present invention, a method for intelligently controlling production equipment based on real-time feedback is provided, the method comprising: Collecting real-time monitoring data sets of production equipment during operation, wherein the real-time monitoring data sets include equipment operating parameters and environment-related parameters; Performing dynamic feature extraction processing on the real-time monitoring data set to obtain a device operation feature set and an environment-related feature set; identifying real-time operating status information of the production equipment according to an association relationship between the equipment operating feature set and the environment-related feature set; Based on the real-time operating status information, a dynamic control strategy set is generated for the production equipment, wherein the dynamic control strategy set includes equipment parameter adjustment instructions and environment adaptation instructions; The operating parameters of the production equipment are dynamically adjusted according to the dynamic control strategy set, and the real-time monitoring data set is updated based on the adjusted operating parameters.

[0006] According to another aspect of an embodiment of the present invention, a production equipment intelligent control system based on real-time feedback is provided, comprising: a processor, a communication interface, a memory and a communication bus, wherein the processor, communication interface and memory communicate with each other through the communication bus; the memory is used to store computer programs; the processor is used to implement any one of the above steps of the production equipment intelligent control method based on real-time feedback when executing the computer program.

[0007] According to another aspect of an embodiment of the present invention, a readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned intelligent control method of production equipment based on real-time feedback can be executed.

[0008] Through any of the above aspects, the embodiment of the present invention collects a set of real-time monitoring data during the operation of production equipment, performs dynamic feature extraction processing on the real-time monitoring data set, and can accurately extract the equipment operation feature set and the environment-related feature set, and then deeply analyze the correlation between the two, so as to accurately identify the real-time operation status information of the production equipment. The dynamic control strategy set generated based on the real-time operation status information includes both equipment parameter adjustment instructions and environment adaptation instructions, thereby realizing the coordinated control of the production equipment operation parameters and environmental factors. Dynamically adjusting the production equipment operation parameters according to the dynamic control strategy set and updating the real-time monitoring data set based on the adjusted operation parameters can significantly improve the operation efficiency of the production equipment, reduce energy consumption, improve product quality, and enhance the stability and reliability of the production process.

[0009] In order to make the above-mentioned objects, features and advantages of the embodiments of the present invention more obvious and easy to understand, the embodiments will be described in detail below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0011] Figure 1 A schematic diagram of components of a production equipment intelligent control system based on real-time feedback provided by an embodiment of the present invention is shown; Figure 2 A flow chart of a method for intelligently controlling production equipment based on real-time feedback provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0012] To help those skilled in the art better understand the present invention, the following will provide a clear and complete description of the technical solutions in the embodiments of the present invention, with reference to the accompanying drawings. It is apparent that the described embodiments are only a portion of the embodiments of the present invention, not all of them. All other embodiments derived by those skilled in the art based on the embodiments of the present invention without creative effort are intended to fall within the scope of protection of the present invention.

[0013] The terms "first," "second," "third," and the like (if any) in the description and claims of the present invention and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of the invention described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to these processes, methods, products, or apparatus.

[0014] Figure 1 A schematic diagram of exemplary components of a system 100 for intelligently controlling production equipment based on real-time feedback is shown. The system 100 may include one or more processors 104, such as one or more central processing units (CPUs), each of which may implement one or more hardware threads. The system 100 may also include any storage medium 106 for storing any type of information, such as code, settings, data, and the like. For example, and without limitation, the storage medium 106 may include any one or more combinations of the following: any type of RAM, any type of ROM, a flash memory device, a hard drive, an optical disk, and the like. More generally, any storage medium may use any technology to store information. Furthermore, any storage medium may provide volatile or non-volatile retention of information. Furthermore, any storage medium may represent a fixed or removable component of the system 100. In one embodiment, when the processor 104 executes instructions stored in any storage medium or combination of storage media with dependencies, the system 100 may perform any operation associated with the associated instructions. The production equipment intelligent control system 100 based on real-time feedback further includes one or more drive units 108 for interacting with any storage medium, such as a hard disk drive unit, an optical disk drive unit, and the like.

[0015] The intelligent control system for production equipment based on real-time feedback 100 also includes input / output 110 (I / O), which is used to receive various inputs (via input unit 112) and provide various outputs (via output unit 114). A specific output mechanism may include a presentation device 116 and a dependent graphical user interface (GUI) 118. The intelligent control system for production equipment based on real-time feedback 100 may also include one or more network interfaces 120, which are used to exchange data with other devices via one or more communication units 122. One or more communication buses 124 couple the components described above together.

[0016] The communication unit 122 can be implemented in any manner, for example, through a local area network, a wide area network (e.g., the Internet), a point-to-point connection, etc., or any combination thereof. The communication unit 122 may include any combination of hard-wired links, wireless links, routers, gateway functions, and the like, governed by any protocol or combination of protocols.

[0017] Figure 2 The flow chart of the intelligent control method and system of production equipment based on real-time feedback provided by the embodiment of the present invention is shown. The intelligent control method and system of production equipment based on real-time feedback can be Figure 1 The production equipment intelligent control system 100 based on real-time feedback shown in the figure is executed, and the detailed steps of the production equipment intelligent control method based on real-time feedback are introduced as follows.

[0018] Step S110: collecting a real-time monitoring data set of the production equipment during operation, wherein the real-time monitoring data set includes equipment operation parameters and environment-related parameters.

[0019] In waste paper recycling production scenarios, intelligently controlling production equipment requires comprehensive and accurate real-time monitoring data from the equipment's operation. This real-time data, encompassing both equipment operating parameters and environmental parameters, is crucial for understanding the equipment's operational status and environmental impact. For example, in a waste paper recycling production line, production equipment includes pulpers, screens, dewatering machines, and other equipment. These devices generate a variety of parameters during operation, and the environment in which they operate also impacts their operation, necessitating the collection of relevant data.

[0020] Step S111: configuring a plurality of sensor nodes connected to the production equipment, wherein the sensor nodes include equipment parameter sensors and environmental parameter sensors.

[0021] Next, to accurately collect the required data, multiple sensor nodes connected to the production equipment are deployed. These sensor nodes are categorized into two types: equipment parameter sensors and environmental parameter sensors. For equipment in the waste paper recycling line, equipment parameter sensors can be installed on the pulper motor to monitor its operating status; sensors can be installed on the screen of the screening machine to monitor its operating conditions. Environmental parameter sensors are placed in various locations throughout the production workshop, such as near windows and around equipment, to collect environmental data.

[0022] Step S112: collecting the operating parameters of the production equipment through the equipment parameter sensor, wherein the operating parameters include temperature parameters, pressure parameters and energy consumption parameters.

[0023] Next, equipment parameter sensors are used to collect the operating parameters of the production equipment. For example, during the pulping process, the motor generates heat, causing the temperature to rise. The equipment parameter sensor collects the motor's temperature parameter, recorded as T, in real time. When the waste paper is shredded and stirred inside the pulper, a certain pressure is generated. The sensor records this pressure parameter, recorded as P. Furthermore, the motor consumes electricity during operation, and the sensor records the energy consumption parameter, recorded as E. These parameters reflect the pulper's operating status and performance.

[0024] Step S113: collecting environment-related parameters of the environment in which the production equipment is located through the environmental parameter sensor, wherein the environment-related parameters include humidity parameters, vibration parameters, and air quality parameters.

[0025] Environmental parameter sensors then begin collecting environmental parameters related to the production equipment's surroundings. In a waste paper recycling workshop, air humidity affects waste paper processing and equipment lifespan. Environmental parameter sensors collect humidity parameters, denoted as H. The operation of other equipment and the transport of materials within the workshop may generate vibrations, which the sensors collect as V. Furthermore, air quality within the workshop, such as dust levels, also affects equipment and operators. Sensors collect air quality parameters, denoted as A.

[0026] Step S114: performing data standardization processing on the operating parameters and the environment-related parameters to generate a real-time monitoring data set with a unified dimension.

[0027] Since the dimensions of the collected operating parameters and environmental related parameters are different, these parameters need to be normalized to ensure the accuracy and consistency of subsequent processing.

[0028] Step S1141: normalizing the temperature parameter, pressure parameter, and energy consumption parameter in the operating parameters to obtain a standardized temperature parameter, a standardized pressure parameter, and a standardized energy consumption parameter.

[0029] For the temperature parameter T, assuming its original value range is between T_min and T_max, it is normalized to the range [0, 1] to obtain the standardized temperature parameter T_norm. The normalization calculation logic is: first calculate the ratio of the temperature parameter to the value range, that is, (T - T_min) / (T_max - T_min), and the result is the standardized temperature parameter T_norm.

[0030] For the pressure parameter P, it is also assumed that its original value range is between P_min and P_max. According to the same normalization method as the temperature parameter, (P-P_min) / (P_max-P_min) is calculated to obtain the standardized pressure parameter P_norm.

[0031] For the energy consumption parameter E, assuming that its original value range is between E_min and E_max, the standardized energy consumption parameter E_norm is obtained by calculating (E-E_min) / (E_max-E_min).

[0032] Step S1142: normalizing the humidity parameter, vibration parameter, and air quality parameter in the environment-related parameters to obtain a standardized humidity parameter, a standardized vibration parameter, and a standardized air quality parameter.

[0033] For the humidity parameter H, assuming that its original value range is between H_min and H_max, the standardized humidity parameter H_norm is obtained by calculating (H-H_min) / (H_max-H_min).

[0034] For the vibration parameter V, assuming that its original value range is between V_min and V_max, the standardized vibration parameter V_norm is obtained by calculating (V-V_min) / (V_max-V_min).

[0035] For the air quality parameter A, assuming that its original value range is between A_min and A_max, the standardized air quality parameter A_norm is obtained by calculating (A-A_min) / (A_max-A_min).

[0036] Step S1143: Align the standardized temperature parameter, the standardized pressure parameter, the standardized energy consumption parameter, the standardized humidity parameter, the standardized vibration parameter, and the standardized air quality parameter according to timestamps to generate the unified dimension real-time monitoring data set.

[0037] Finally, the obtained standardized temperature parameter T_norm, standardized pressure parameter P_norm, standardized energy consumption parameter E_norm, standardized humidity parameter H_norm, standardized vibration parameter V_norm, and standardized air quality parameter A_norm are aligned according to the acquisition timestamp. For example, these standardized parameters collected at the same time t are combined to form a multidimensional data vector. All these data vectors at different times are combined to generate a unified dimension of real-time monitoring data set.

[0038] Step S120: performing dynamic feature extraction processing on the real-time monitoring data set to obtain a device operation feature set and an environment-related feature set.

[0039] After obtaining a unified dimension of real-time monitoring data, dynamic feature extraction is performed to extract useful information from the data, resulting in a set of equipment operation features and a set of environmental correlation features. In waste paper recycling production scenarios, this helps to gain a deeper understanding of the equipment's operating characteristics and the impact of the environment on the equipment.

[0040] Step S121: performing time series segmentation processing on the equipment operation parameters in the real-time monitoring data set to obtain equipment operation sub-parameter sets corresponding to multiple continuous time windows.

[0041] First, the equipment operating parameters in the real-time monitoring data set are segmented into time series. Taking the pulper operating parameters as an example, the temperature parameter T_norm, pressure parameter P_norm, and energy consumption parameter E_norm collected over a period of time are divided into continuous time windows. Assuming the length of the time window is Δt, starting from the starting time t_0, the first time window is [t_0, t_0+Δt]. The temperature parameters, pressure parameters, and energy consumption parameters within this time window constitute an equipment operating sub-parameter set D_1. The second time window is [t_0+Δt, t_0+2Δt], and the corresponding equipment operating sub-parameter set is D_2. And so on, the equipment operating sub-parameter sets D_1, D_2, ..., D_n corresponding to multiple continuous time windows are obtained.

[0042] Step S122: Perform multi-dimensional feature extraction processing on each of the device operation sub-parameter sets to obtain a device operation feature subset corresponding to each time window, wherein the multi-dimensional feature extraction processing includes parameter fluctuation feature extraction, trend correlation feature extraction and abnormal accumulation feature extraction.

[0043] Next, multi-dimensional feature extraction processing is performed on each device operation sub-parameter set.

[0044] For parameter fluctuation feature extraction, take the temperature parameter as an example. Within a certain time window [t_i, t_i + Δt], the temperature parameter T_norm may fluctuate. The standard deviation σ_T of the temperature parameter within this time window can be calculated to indicate the degree of temperature fluctuation. A larger standard deviation indicates more severe temperature fluctuations. The same method is used for the pressure parameter P_norm and the energy consumption parameter E_norm, calculating their standard deviations σ_P and σ_E, respectively. These standard deviations are combined to form the parameter fluctuation feature vector F_1 = [σ_T, σ_P, σ_E] for this time window.

[0045] To extract trend-related features, analyze the changing trends of temperature, pressure, and energy consumption parameters within a time window and their correlations. For example, we can calculate the difference ΔT between adjacent time points in temperature, ΔP between pressure, and ΔE between energy consumption parameters, and then analyze the correlation between these differences. If ΔT and ΔE show a positive correlation, it indicates that an increase in temperature is likely accompanied by an increase in energy consumption. These trend-related features are combined to form a trend-related feature vector F_2.

[0046] To extract the cumulative anomaly feature, set a normal range for each parameter. If a parameter exceeds the normal range within a time window, it is considered an anomaly. Count the number of anomalies for each parameter within the time window. For example, the number of anomalies for temperature is N_T, the number of anomalies for pressure is N_P, and the number of anomalies for energy consumption is N_E. Combine these cumulative anomaly counts to form the cumulative anomaly feature vector F_3 = [N_T, N_P, N_E].

[0047] Finally, the parameter fluctuation feature vector F_1, the trend correlation feature vector F_2 and the abnormal accumulation feature vector F_3 are spliced ​​together to obtain the equipment operation feature subset C_i corresponding to each time window.

[0048] Step S123: performing spatial division processing on the environment-related parameters in the real-time monitoring data set to obtain an environment-related sub-parameter set corresponding to each time window.

[0049] Then, the environment-related parameters in the real-time monitoring data set are spatially divided. In a waste paper recycling workshop, the environmental parameters at different locations may be different. The workshop is divided into multiple spatial areas, for example, it is divided into m areas according to the layout of the workshop. For each time window, the humidity parameter H_norm, vibration parameter V_norm and air quality parameter A_norm collected in each area within the time window are combined to form an environment-related sub-parameter set. For example, in the time window [t_j, t_j+Δt], the environment-related sub-parameter set of the kth area is E_kj=[H_norm_kj, V_norm_kj, A_norm_kj], where k=1, 2,…, m.

[0050] Step S124: performing environmental response feature extraction processing on each of the environment-related sub-parameter sets to obtain an environment-related feature subset corresponding to each time window.

[0051] Subsequently, environmental response feature extraction is performed on each set of environmentally relevant sub-parameters. Taking humidity as an example, the potential impact of its variations within different regions and time windows on device operation is analyzed. The mean μ_H and variance σ_H of the humidity parameter for each region are calculated. The mean reflects the average humidity level in the region, while the variance reflects humidity fluctuations. The same method is used for vibration and air quality parameters, calculating their mean μ_V and μ_A and variances σ_V and σ_A, respectively. These mean and variance values ​​are combined to form the environmental response feature vector R_kj = [μ_H_kj, σ_H_kj, μ_V_kj, σ_V_kj, μ_A_kj, σ_A_kj] for that region within that time window. The environmental response feature vectors for each region are combined to obtain the corresponding environmentally relevant feature subset E_j for each time window.

[0052] Step S125: Aggregate the device operation feature subsets and the environment association feature subsets of each time window to generate the device operation feature set and the environment association feature set.

[0053] Finally, the device operation feature subsets C_1, C_2, ..., C_n of each time window are aggregated to form the device operation feature set C. A simple concatenation method can be used to sequentially concatenate the device operation feature subsets of each time window to obtain the device operation feature set C. Similarly, the environment-related feature subsets E_1, E_2, ..., E_n of each time window are aggregated to form the environment-related feature set E.

[0054] Step S130: Identify the real-time operation status information of the production equipment according to the association relationship between the equipment operation feature set and the environment association feature set.

[0055] After obtaining the device operation feature set and the environmental correlation feature set, it is necessary to analyze the correlation between them to identify the real-time operating status information of the production equipment. In the waste paper recycling production scenario, the operating status of the equipment will be affected by environmental factors, so analyzing this correlation is crucial for accurately determining the equipment status.

[0056] Step S131: obtaining a device operation feature subset of each time window in the device operation feature set, and an environment-related feature subset of the corresponding time window in the environment-related feature set.

[0057] First, extract the device operation feature subset C_i for each time window from the device operation feature set C, and simultaneously extract the environment-related feature subset E_i for the corresponding time window from the environment-related feature set E. For example, within the time window [t_i, t_i+Δt], obtain the device operation feature subset C_i and the environment-related feature subset E_i.

[0058] Step S132: performing joint encoding processing on the device operation feature subset and the environment association feature subset to generate a joint state feature vector for each time window.

[0059] Next, the subset of equipment operation features, C_i, and the subset of environment-related features, E_i, are jointly encoded. This can be accomplished by using an encoder network from deep learning, taking C_i and E_i as inputs and performing feature transformation and fusion through the encoder's multi-layer neural network. The encoder network's input layer receives the feature vectors of C_i and E_i, which, after undergoing nonlinear transformations in the hidden layer, ultimately output a joint state feature vector, S_i. This joint state feature vector integrates both equipment operation characteristics and environment-related features, providing a more comprehensive picture of the production equipment's state within that time window.

[0060] Step S133: performing state recognition processing on the joint state feature vector according to a preset state classification model to obtain a preliminary state classification result for each time window.

[0061] Then, a preset state classification model is used to perform state recognition processing on the joint state feature vector S_i. The state classification model can be a trained neural network classifier, such as a multi-layer perceptron (MLP). The joint state feature vector S_i is input into the MLP. The input layer of the MLP receives the feature vector of S_i. The intermediate hidden layer performs feature processing and classification decisions, and the final output layer outputs a preliminary state classification result. The state classification result can be categorized as normal operation, mild abnormality, severe abnormality, etc.

[0062] Step S134: performing state transition analysis on the preliminary state classification results of the continuous time window to generate real-time operation state information of the production equipment within the continuous operation cycle, wherein the real-time operation state information includes equipment performance degradation trends and environmental coupling abnormality patterns.

[0063] Finally, the preliminary state classification results for the continuous time window are processed through state transition analysis. A state transition matrix can be constructed to record the transition probabilities from the state of one time window to the state of the next. For example, the probability of transitioning from a normal operating state to a mildly abnormal state, the probability of transitioning from a mildly abnormal state to a severely abnormal state, and so on. By analyzing the state transition matrix and the state changes in the continuous time window, real-time operating status information of the production equipment during the continuous operation cycle can be generated. Equipment performance degradation trends can be determined by observing the gradual change from normal to abnormal state within the continuous time window, and environmental coupling anomaly patterns can be identified by analyzing the relationship between environmental correlation characteristics and equipment state changes.

[0064] Step S140: Based on the real-time operating status information, a dynamic control strategy set for the production equipment is generated, where the dynamic control strategy set includes equipment parameter adjustment instructions and environment adaptation instructions.

[0065] After identifying the real-time operating status of production equipment, a set of dynamic control strategies for the equipment needs to be generated based on this information. In the waste paper recycling production scenario, this dynamic control strategy includes instructions for adjusting equipment parameters and adapting to the environment to ensure stable and efficient operation of the equipment under different operating conditions.

[0066] Step S141: performing degradation stage division processing on the equipment performance degradation trend in the real-time operation status information to obtain performance attenuation characteristics corresponding to different degradation stages.

[0067] First, the degradation trend of equipment performance in the real-time operating status information is divided into degradation stages. According to the change process of equipment performance from normal to abnormal, the degradation process can be divided into multiple stages, such as the early degradation stage, the mid-stage degradation stage, and the late degradation stage. For each degradation stage, the performance degradation characteristics of the equipment are analyzed. Taking the pulper as an example, in the early degradation stage, it may be manifested as a slight increase in energy consumption and a slightly increased temperature fluctuation; in the mid-stage degradation stage, the pressure parameters may become unstable and the processing efficiency of the equipment will decrease; in the late degradation stage, the equipment may frequently experience abnormalities and the processing capacity will be greatly reduced. The performance degradation characteristics of each degradation stage are recorded to form a set of performance degradation characteristics corresponding to different degradation stages.

[0068] Step S142: matching a preset control rule library according to the performance degradation characteristics, and screening to obtain candidate device parameter adjustment instructions corresponding to the current degradation stage.

[0069] Next, the system matches the performance degradation characteristics to a preset control rule library. This control rule library stores device parameter adjustment instructions corresponding to different performance degradation characteristics. For example, if the performance degradation characteristic manifests as increased energy consumption, the control rule library may contain device parameter adjustment instructions such as reducing motor speed or optimizing the operating mode. Based on the performance degradation characteristics of the current degradation stage, candidate device parameter adjustment instructions that match the characteristics are selected from the control rule library.

[0070] Step S143: performing abnormal impact assessment processing on the environmental coupling abnormality pattern in the real-time operation status information to obtain an environmental abnormality impact level.

[0071] Next, the impact of environmental coupling anomaly patterns in real-time operating status information is assessed and processed. The impact of environmental anomalies on equipment operation is analyzed. For example, abnormal humidity may cause equipment to rust, affecting waste paper processing efficiency; abnormal vibration may cause equipment components to loosen, affecting equipment stability. Based on the severity of the environmental anomaly and the extent of its impact on equipment operation, the impact level is classified as mild, moderate, or severe.

[0072] Step S144: Filtering the candidate environment adaptation instructions corresponding to the current environment abnormality mode from the control rule library according to the environment abnormality impact level.

[0073] Subsequently, candidate environmental adaptation instructions corresponding to the current environmental anomaly pattern are screened from the control rule library based on the environmental anomaly impact level. The control rule library stores environmental adaptation instructions corresponding to different environmental anomaly impact levels. For example, if the environmental anomaly impact level is mild, possible environmental adaptation instructions may include increasing ventilation or adjusting humidity control parameters. If the environmental anomaly impact level is severe, more radical measures may be required, such as suspending production or conducting comprehensive environmental remediation.

[0074] Step S145: performing policy priority sorting processing on the candidate device parameter adjustment instructions and the candidate environment adaptation instructions to generate the dynamic control policy set.

[0075] Finally, the candidate device parameter adjustment instructions and candidate environment adaptation instructions are prioritized. Each instruction is assigned a priority level, taking into account its importance and urgency to device operation and environmental improvement. For example, instructions that severely impact device safety are given a higher priority, while instructions that have a limited impact on device performance but are not urgent are given a lower priority. The ranked candidate device parameter adjustment instructions and candidate environment adaptation instructions are combined to generate a dynamic control policy set.

[0076] Step S150: dynamically adjusting the operating parameters of the production equipment according to the dynamic control strategy set, and updating the real-time monitoring data set based on the adjusted operating parameters.

[0077] After generating a dynamic control strategy set, the operating parameters of production equipment need to be dynamically adjusted based on this strategy set, and the real-time monitoring data set needs to be updated based on the adjusted operating parameters. In waste paper recycling production scenarios, this step ensures that equipment is optimized and adjusted based on real-time operating conditions, improving production efficiency and quality.

[0078] Step S151: parsing the device parameter adjustment instructions in the dynamic control strategy set to determine the target device parameters to be adjusted and the adjustment range.

[0079] In the waste paper recycling production scenario, parsing the equipment parameter adjustment instructions in the dynamic control strategy set is the key first step. For example, if the dynamic control strategy set contains equipment parameter adjustment instructions for the pulper, the instructions may clearly indicate that the motor speed, pulping time and other parameters of the pulper need to be adjusted. For the target equipment parameter of the motor speed, the specific adjustment range needs to be determined. Assuming that the current motor speed is a parameter represented by a letter R1, and the adjustment instruction requires it to be adjusted to a new speed R2, then the adjustment range is the difference between R2 and R1, that is, R2-R1. For the parameter of pulping time, it is also assumed that the current pulping time is T1 and the adjusted pulping time is T2, the adjustment range is T2-T1. Through such a parsing process, the target equipment parameters to be adjusted and the corresponding adjustment range can be accurately determined, providing a clear basis for subsequent parameter adjustment operations.

[0080] Step S152: matching a corresponding device control interface according to the type of the target device parameter, and sending a parameter adjustment instruction to the execution mechanism of the production device through the device control interface.

[0081] Next, the corresponding device control interface is matched according to the type of the determined target device parameter. In the waste paper recycling production line, different device parameters correspond to different control interfaces. Taking the pulper as an example, if the target device parameter is the motor speed, then the corresponding device control interface may be the motor speed controller. Through the speed controller, the parameter adjustment instruction for adjusting the motor speed can be sent to the motor actuator of the pulper. Similarly, if the target device parameter is the pulping time, the corresponding device control interface may be the time control module of the pulper, through which the instruction for adjusting the pulping time is sent to the corresponding actuator of the pulper. In the process of sending instructions, it is necessary to ensure the accuracy and completeness of the instructions to ensure that the actuator can correctly perform the adjustment operation.

[0082] Step S153: monitoring the response data of the actuator to the parameter adjustment instruction in real time, and calculating the parameter adjustment error rate according to the response data.

[0083] After the parameter adjustment instruction is sent to the actuator, the actuator's response data to the instruction needs to be monitored in real time. For the motor speed adjustment of the pulper, the actual speed R_actual of the motor is monitored in real time. The actual speed R_actual is compared with the adjusted target speed R2 to calculate the speed adjustment error rate. The calculation logic of the error rate is to first calculate the absolute value of the difference between the actual speed and the target speed, that is, |R_actual-R2|, and then divide the difference by the target speed R2 to obtain the speed adjustment error rate E_R. For the adjustment of the pulping time, the actual pulping time T_actual of the pulper is monitored in real time. Similarly, the absolute value of the difference between the actual pulping time and the target pulping time T2 is calculated, that is, |T_actual-T2|, and then divided by the target pulping time T2 to obtain the pulping time adjustment error rate E_T. By real-time monitoring and calculation of the parameter adjustment error rate, the execution status of the parameter adjustment instruction by the actuator can be timely understood.

[0084] Step S154: If the parameter adjustment error rate exceeds a preset threshold, a regeneration process of the dynamic control strategy set is triggered, and the parameter adjustment instruction is updated based on the regenerated dynamic control strategy set.

[0085] If the calculated parameter adjustment error rate exceeds the preset threshold, it means that the current parameter adjustment has not achieved the expected effect, and it is necessary to trigger the regeneration process of the dynamic control strategy set. Taking the pulper motor speed adjustment as an example, if the speed adjustment error rate E_R exceeds the preset speed error threshold E_R_threshold, it is necessary to re-evaluate the real-time operating status of the production equipment. This includes re-collecting equipment operating parameters and environmental related parameters, performing dynamic feature extraction processing again, identifying the real-time operating status information of the equipment, and then regenerating the dynamic control strategy set based on the new real-time operating status information. The regenerated dynamic control strategy set will include new equipment parameter adjustment instructions, which will update the previous parameter adjustment instructions with these new instructions and send them to the actuator for adjustment again, in the hope of achieving better adjustment results.

[0086] Step S160: According to claim 8, after dynamically adjusting the operating parameters of the production equipment according to the dynamic control strategy set, it also includes obtaining an equipment response data set corresponding to the adjusted operating parameters, and the equipment response data set includes the equipment operating status indicators and environmental status indicators after the parameters are adjusted.

[0087] After completing the dynamic adjustment of the production equipment's operating parameters, it is necessary to obtain the equipment response data set corresponding to the adjusted operating parameters. In the waste paper recycling production scenario, for a pulper, the equipment operating status indicators may include parameters such as the adjusted motor speed, pulping efficiency, and energy consumption, and the environmental status indicators may include parameters such as the humidity and air quality in the workshop. For example, after adjusting the pulper motor speed, the actual speed of the adjusted motor is collected in real time through the corresponding sensor and used as one of the equipment operating status indicators. At the same time, the humidity sensor data and air quality sensor data in the workshop are collected as environmental status indicators. These data constitute the equipment response data set, which can reflect the actual operation of the equipment after the parameter adjustment and the changes in the environment.

[0088] Step S161: performing difference analysis on the equipment operation status indicator and the expected adjustment target to generate a parameter adjustment effect evaluation result.

[0089] Next, a discrepancy analysis is performed between the equipment operating status indicators and the expected adjustment targets. Taking the pulper motor speed as an example, the expected adjustment target speed is R2, while the actual adjusted motor speed collected is R_actual. The difference between the two, |R_actual-R2|, is calculated, and the parameter adjustment effect is evaluated based on the size of this difference and the preset evaluation criteria. If the difference is small, it indicates that the adjustment effect is good, and the parameter adjustment effect evaluation result may be "good"; if the difference is large, it indicates that the adjustment effect is poor, and the evaluation result may be "poor." A similar method is used to perform discrepancy analysis for other equipment operating status indicators, such as pulping efficiency. The evaluation results of each equipment operating status indicator are combined to generate the final parameter adjustment effect evaluation result.

[0090] Step S162: performing difference analysis on the environmental status indicator and the expected environmental adaptation target to generate an environmental adaptation effect evaluation result.

[0091] Environmental status indicators are similarly analyzed for differences from the expected environmental adaptation target. For example, the expected environmental adaptation target humidity is H2, and the actual adjusted humidity collected is H_actual. The difference between the two values, |H_actual - H2|, is calculated. The environmental adaptation effect is evaluated based on the size of this difference and pre-set evaluation criteria. If the difference is within an acceptable range, the environmental adaptation effect is good, and the environmental adaptation effect evaluation result may be "meets the standard." If the difference is outside the range, the environmental adaptation effect is unsatisfactory, and the evaluation result may be "does not meet the standard." Similar analysis is performed for other environmental status indicators, such as air quality. The evaluation results of each environmental status indicator are combined to generate the environmental adaptation effect evaluation result.

[0092] Step S163: If the parameter adjustment effect evaluation result indicates that the equipment operation status indicator has not reached the expected adjustment target, the priority of the control strategy associated with the current equipment parameter adjustment instruction is increased.

[0093] If the parameter adjustment effect evaluation results indicate that the equipment operating status indicators have not reached the expected adjustment targets, it indicates that the current equipment parameter adjustment instructions may need to be executed and optimized with higher priority. For example, if the pulper motor speed adjustment does not reach the expected target, the control strategy associated with the motor speed adjustment instruction will need to be prioritized. This means that in the subsequent control process, further adjustments to the motor speed will be given higher priority, perhaps with larger adjustments or more refined adjustment strategies, to quickly bring the equipment operating status indicators back to the expected targets.

[0094] Step S164: If the environmental adaptation effect evaluation result indicates that the environmental status indicator does not reach the expected environmental adaptation target, the priority of the control strategy associated with the current environmental adaptation instruction is increased.

[0095] Similarly, if the environmental adaptation evaluation results indicate that the environmental status indicators have not reached the expected environmental adaptation target, the control strategy associated with the current environmental adaptation instruction needs to be prioritized. For example, if the humidity in the workshop has not reached the expected adaptation target, the control strategy priority of the environmental adaptation instruction related to humidity adjustment will be increased. This may involve increasing the operating power of the humidity control equipment or adjusting the parameters of the ventilation system to accelerate the process of achieving the expected environmental status indicators.

[0096] Step S165: If both the parameter adjustment effect evaluation result and the environment adaptation effect evaluation result reach the expected goals, the priority of the executed control strategy is lowered, and the next candidate control strategy is activated.

[0097] When both the parameter adjustment effect evaluation results and the environmental adaptation effect evaluation results reach the expected goals, it means that the currently executed control strategy has achieved good results. At this time, lowering the priority of the executed control strategy means that these successfully executed strategies will no longer be given priority in the subsequent control process. At the same time, the next candidate control strategy is activated to continue optimizing the operating parameters and environment of the production equipment to further improve production efficiency and quality. For example, if the motor speed of the pulper and the humidity in the workshop have reached the expected goals, then the priority of the control strategies related to motor speed adjustment and humidity adjustment is lowered, and the next candidate control strategy for equipment energy consumption or waste paper processing quality is activated.

[0098] After updating the real-time monitoring data set based on the adjusted operating parameters, the method further includes: Step S170: performing anomaly detection processing on the updated real-time monitoring data set to identify abnormal data segments generated during the data collection process.

[0099] After the real-time monitoring data set is updated, it needs to be processed for anomaly detection to identify abnormal data fragments that may be generated during the data collection process. In the waste paper recycling production scenario, abnormal data may be generated due to sensor failure, communication interference, and other reasons. For the operating parameters of the pulper, such as temperature, pressure and other data, anomaly detection is performed by setting a normal value range. Assuming that the normal value range of the temperature parameter is between T_min and T_max, if the collected temperature data exceeds the normal value range, it will be marked as abnormal data. The same method is used for anomaly detection for other operating parameters such as pressure parameters, energy consumption parameters, and environmental related parameters such as humidity and vibration. By traversing the updated real-time monitoring data set, all data fragments that exceed the normal value range are found and identified as abnormal data fragments.

[0100] Step S171: performing data repair processing on the abnormal data segment to generate a repaired real-time monitoring data subset.

[0101] After identifying the abnormal data segment, it is necessary to perform data repair processing on it. A common repair method is to use interpolation. Taking the temperature data of the pulper as an example, if the temperature data collected at a certain moment is an abnormal value T_abnormal, and the normal temperature data before and after this moment are T_before and T_after respectively, the abnormal value can be repaired by linear interpolation. The calculation logic of linear interpolation is to calculate the estimated temperature value T_estimate at this moment based on the time interval and numerical relationship between the normal data before and after. Assuming that the time interval between the normal data before and after is Δt, and the time interval between the abnormal value moment and the previous normal data moment is Δt1, then T_estimate=T_before+(T_after-T_before)*(Δt1 / Δt). For abnormal data of other parameters, similar interpolation methods are also used for repair. All the repaired data are combined to generate a repaired real-time monitoring data subset.

[0102] Step S172: merging the repaired real-time monitoring data subset with the normal data segment to generate an updated real-time monitoring data set.

[0103] The repaired subset of real-time monitoring data is merged with normal data segments that were not previously identified as abnormal. This merging process arranges the data in chronological order and inserts the repaired data into the corresponding time position. For example, in a time series, repaired temperature data, pressure data, energy consumption data, and other normal data are sorted according to timestamps to form a complete, updated set of real-time monitoring data. This real-time monitoring data set, which includes both repaired abnormal data and normal data, can more accurately reflect the operating status and environmental conditions of production equipment.

[0104] Step S173: re-execute the dynamic feature extraction process and subsequent steps based on the updated real-time monitoring data set to achieve closed-loop control of the production equipment.

[0105] Finally, the dynamic feature extraction process and subsequent steps are re-executed based on the updated real-time monitoring data set. The updated real-time monitoring data set is again subjected to time-series segmentation and spatial partitioning to extract the equipment operation feature set and the environmental correlation feature set. Based on these feature sets, the real-time operating status information of the production equipment is identified, and a dynamic control strategy set is generated. The operating parameters of the production equipment are dynamically adjusted, and the real-time monitoring data set is updated. By continuously executing these steps, closed-loop control of the production equipment is achieved, ensuring that the production equipment always operates in optimal conditions, thereby improving the efficiency and quality of waste paper recycling production.

[0106] Regarding the construction and training of the state classification model: Step S210: constructing a state classification model, wherein the state classification model adopts a multi-layer perceptron (MLP) including an input layer, a hidden layer, and an output layer.

[0107] When building the state classification model, a multilayer perceptron (MLP) was chosen as the model architecture. The number of neurons in the input layer is determined by the dimension of the joint state feature vector. Assuming the dimension of the joint state feature vector is n, the input layer has n neurons to receive the eigenvalues ​​of the joint state feature vector. Multiple hidden layers can be configured, each containing a certain number of neurons that transform the input data and extract features using a nonlinear activation function. The number of neurons in the output layer is determined by the number of state classification categories. For example, if the state classification results are categorized as normal operation, mild abnormality, and severe abnormality, the output layer will have three neurons.

[0108] Step S220: prepare training data, where the training data includes joint state feature vector samples and corresponding state classification labels.

[0109] When preparing training data, a large number of joint state feature vector samples and corresponding state classification labels are collected. These joint state feature vector samples can be obtained by jointly encoding a set of historical equipment operation features and a set of environmentally relevant features. State classification labels are assigned based on expert experience or actual equipment operation conditions. For example, a normally operating device state is labeled "normal operation state"; a state with minor faults but no impact on normal production is labeled "mild abnormal state"; and a state that severely impacts production and may even cause equipment damage is labeled "severe abnormal state." These joint state feature vector samples and corresponding state classification labels are combined to form the training dataset.

[0110] Step S230: Initialize model parameters, including the weight matrix from the input layer to the hidden layer, the weight matrix from the hidden layer to the output layer, and the bias vectors of each layer.

[0111] Before training begins, the model parameters must be initialized. The dimensions of the weight matrix W1 from the input layer to the hidden layer are the number of neurons in the hidden layer multiplied by the number of neurons in the input layer. The dimensions of the weight matrix W2 from the hidden layer to the output layer are the number of neurons in the output layer multiplied by the number of neurons in the hidden layer. The dimensions of the bias vectors b1 and b2 for each layer are the same as the number of neurons in the hidden and output layers, respectively. The initial values ​​of these weight matrices and bias vectors can be randomly initialized, for example, using a uniform or Gaussian distribution to generate random values.

[0112] Step S240: Perform model training and use the back propagation algorithm and optimizer to update the model parameters.

[0113] During model training, the backpropagation algorithm and optimizer are used to update the model parameters. First, the joint state feature vector samples from the training dataset are input into the MLP. After calculations in the input, hidden, and output layers, the model output is obtained. The model output is compared with the corresponding state classification label to calculate the loss function. The loss function can be a cross-entropy loss function, which measures the degree of difference between the model output and the true label. Then, the backpropagation algorithm is used to calculate the gradient of the loss function with respect to the model parameters. Based on this gradient information, the model parameters are updated using an optimizer (such as stochastic gradient descent or Adam optimizer). This process is repeated until the model's loss function converges or the preset number of training rounds is reached.

[0114] Step S250: Evaluate model performance, use the test data set to test the trained model, and calculate indicators such as accuracy and recall.

[0115] After model training is complete, the model's performance is evaluated using a test dataset. The test dataset is prepared similarly to the training dataset, containing samples of the joint state feature vector and corresponding state classification labels. Samples from the test dataset are fed into the trained model to obtain the model's predictions. The predictions are compared with the true labels, and metrics such as precision and recall are calculated. Precision represents the proportion of samples correctly predicted by the model to the total number of samples, while recall represents the proportion of samples correctly predicted for a particular category to the number of true samples of that category. These metrics are used to evaluate model performance. If performance is unsatisfactory, the model structure or training parameters can be adjusted and training can be repeated.

[0116] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0117] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations that come within the meaning and range of equivalents of the claims be embraced therein.

Claims

1. A method for intelligent control of production equipment based on real-time feedback, characterized in that: The method comprises: Collecting real-time monitoring data sets of production equipment during operation, wherein the real-time monitoring data sets include equipment operating parameters and environment-related parameters; Performing dynamic feature extraction processing on the real-time monitoring data set to obtain a device operation feature set and an environment-related feature set; Identifying real-time operating status information of the production equipment based on an association relationship between the equipment operating feature set and the environment-related feature set; Based on the real-time operating status information, a dynamic control strategy set is generated for the production equipment, wherein the dynamic control strategy set includes equipment parameter adjustment instructions and environment adaptation instructions; The operating parameters of the production equipment are dynamically adjusted according to the dynamic control strategy set, and the real-time monitoring data set is updated based on the adjusted operating parameters.

2. The intelligent control method for production equipment based on real-time feedback according to claim 1 is characterized in that: The dynamic feature extraction processing is performed on the real-time monitoring data set to obtain a device operation feature set and an environment-related feature set, including: Performing time series segmentation processing on the equipment operating parameters in the real-time monitoring data set to obtain equipment operating sub-parameter sets corresponding to multiple continuous time windows; Performing multi-dimensional feature extraction processing on each of the device operation sub-parameter sets to obtain a device operation feature subset corresponding to each time window, wherein the multi-dimensional feature extraction processing includes parameter fluctuation feature extraction, trend correlation feature extraction, and abnormal accumulation feature extraction; Performing spatial division processing on the environment-related parameters in the real-time monitoring data set to obtain an environment-related sub-parameter set corresponding to each time window; Performing environmental response feature extraction processing on each of the environment-related sub-parameter sets to obtain an environment-related feature subset corresponding to each time window; The device operation feature subsets and the environment-related feature subsets of each time window are aggregated respectively to generate the device operation feature set and the environment-related feature set.

3. The intelligent control method for production equipment based on real-time feedback according to claim 1 is characterized in that: The identifying the real-time operation status information of the production equipment according to the association relationship between the equipment operation feature set and the environment-related feature set includes: Obtaining a device operation feature subset for each time window in the device operation feature set, and an environment-related feature subset for the corresponding time window in the environment-related feature set; performing joint encoding processing on the device operation feature subset and the environment association feature subset to generate a joint state feature vector for each time window; Performing state recognition processing on the joint state feature vector according to a preset state classification model to obtain a preliminary state classification result for each time window; The preliminary state classification results of the continuous time window are subjected to state transition analysis processing to generate real-time operation state information of the production equipment within the continuous operation cycle, wherein the real-time operation state information includes equipment performance degradation trends and environmental coupling abnormal patterns.

4. The intelligent control method for production equipment based on real-time feedback according to claim 1 is characterized in that: Generating a dynamic control strategy set for the production equipment based on the real-time operating status information includes: Dividing the performance degradation trend of the equipment in the real-time operating status information into degradation stages to obtain performance attenuation characteristics corresponding to different degradation stages; Matching a preset control rule library according to the performance degradation characteristics to screen out candidate device parameter adjustment instructions corresponding to the current degradation stage; Performing an abnormal impact assessment on the environmental coupling abnormality mode in the real-time operating status information to obtain an environmental abnormality impact level; Filtering the candidate environmental adaptation instructions corresponding to the current environmental abnormality mode from the control rule library according to the environmental abnormality impact level; The candidate device parameter adjustment instructions and the candidate environment adaptation instructions are subjected to policy priority sorting to generate the dynamic control policy set.

5. The intelligent control method for production equipment based on real-time feedback according to claim 1 is characterized in that: The dynamically adjusting the operating parameters of the production equipment according to the dynamic control strategy set includes: Parsing the device parameter adjustment instructions in the dynamic control strategy set to determine the target device parameters to be adjusted and the adjustment range; Matching a corresponding device control interface according to the type of the target device parameter, and sending a parameter adjustment instruction to the execution mechanism of the production device through the device control interface; monitoring the response data of the actuator to the parameter adjustment instruction in real time, and calculating the parameter adjustment error rate based on the response data; If the parameter adjustment error rate exceeds a preset threshold, a regeneration process of the dynamic control strategy set is triggered, and the parameter adjustment instruction is updated based on the regenerated dynamic control strategy set.

6. The intelligent control method for production equipment based on real-time feedback according to claim 1 is characterized in that: The real-time monitoring data set collected during the operation of the production equipment includes: Configuring a plurality of sensor nodes connected to the production equipment, wherein the sensor nodes include equipment parameter sensors and environmental parameter sensors; The operating parameters of the production equipment are collected by the equipment parameter sensor, wherein the operating parameters include temperature parameters, pressure parameters and energy consumption parameters; Collecting environmental parameters of the environment in which the production equipment is located by the environmental parameter sensor, wherein the environmental parameters include humidity parameters, vibration parameters and air quality parameters; Data standardization is performed on the operating parameters and the environment-related parameters to generate a real-time monitoring data set with a unified dimension.

7. The intelligent control method for production equipment based on real-time feedback according to claim 6 is characterized in that: The step of performing data standardization on the operating parameters and the environment-related parameters to generate a real-time monitoring data set with a unified dimension includes: Normalizing the temperature parameter, pressure parameter, and energy consumption parameter in the operating parameters to obtain a standardized temperature parameter, a standardized pressure parameter, and a standardized energy consumption parameter; Normalizing the humidity parameter, vibration parameter, and air quality parameter in the environmental related parameters to obtain a standardized humidity parameter, a standardized vibration parameter, and a standardized air quality parameter; The standardized temperature parameter, the standardized pressure parameter, the standardized energy consumption parameter, the standardized humidity parameter, the standardized vibration parameter and the standardized air quality parameter are aligned according to timestamps to generate the real-time monitoring data set of the unified dimension.

8. The intelligent control method for production equipment based on real-time feedback according to claim 1 is characterized in that: After dynamically adjusting the operating parameters of the production equipment according to the dynamic control strategy set, the method further includes: Acquire a device response data set corresponding to the adjusted operating parameters, the device response data set including the device operating status indicator and the environmental status indicator after the parameter adjustment; Performing difference analysis on the equipment operating status indicators and the expected adjustment targets to generate parameter adjustment effect evaluation results; Performing difference analysis on the environmental status indicators and the expected environmental adaptation targets to generate an environmental adaptation effect evaluation result; If the parameter adjustment effect evaluation result indicates that the equipment operating status indicator has not reached the expected adjustment target, then increasing the priority of the control strategy associated with the current equipment parameter adjustment instruction; If the environmental adaptation effect evaluation result indicates that the environmental status indicator has not reached the expected environmental adaptation target, then increasing the priority of the control strategy associated with the current environmental adaptation instruction; If both the parameter adjustment effect evaluation result and the environment adaptation effect evaluation result reach the expected goals, the priority of the executed control strategy is lowered, and the next candidate control strategy is activated.

9. The intelligent control method for production equipment based on real-time feedback according to claim 1 is characterized in that: After updating the real-time monitoring data set based on the adjusted operating parameters, the method further includes: Perform anomaly detection on the updated real-time monitoring data set to identify abnormal data fragments generated during the data collection process; Performing data repair processing on the abnormal data segment to generate a repaired real-time monitoring data subset; Merging the repaired real-time monitoring data subset with the normal data segment to generate an updated real-time monitoring data set; The dynamic feature extraction process and subsequent steps are re-executed based on the updated real-time monitoring data set to achieve closed-loop control of the production equipment.

10. An intelligent control system for production equipment based on real-time feedback, characterized in that: include: A processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus; The memory is used to store computer programs; the processor is used to implement the steps of the intelligent control method of production equipment based on real-time feedback as described in any one of claims 1 to 9 when executing the computer program.

Images