Cooperative control method and system for communication between sugarcane leaf recycling devices

Abstract

The invention relates to the technical field of information, and discloses a communication cooperative control method and system between sugarcane leaf recycling devices. The method comprises the following steps: acquiring high-frequency data from each module through a restart mechanism to obtain an original data set, and processing the original data set by using a data compression algorithm to obtain a compressed data set; if the compressed data set is smaller than the bandwidth threshold value, the compressed data set is transmitted through the communication channel, the transmission completion state is judged, and the transmitted compressed data is decompressed to obtain a reduced data set; standardizing the reduced data set to obtain a standardized data set, and grouping abnormal data to obtain fault mode groups; positioning a high-frequency fault point by fusing the parameter deviation to obtain an optimized restart triggering condition, and adjusting a restart mechanism according to the triggering condition to obtain an adjusted restart mechanism; and acquiring new high-frequency data by using the adjusted mechanism to obtain an updated original data set, and generating a running state log. Through data compression and clustering analysis, the problems of limited high-frequency data transmission bandwidth and inaccurate fault identification are solved, and the system stability is improved.

Description

Technical Field

[0001] This application relates to the field of information technology, and in particular to a communication and collaborative control method and system for sugarcane leaf recycling equipment. Background Technology

[0002] In the industrial production of sugarcane leaf recycling, the equipment consists of multiple modules and needs to collect high-frequency data such as environmental parameter values ​​and parameter deviations in real time. By monitoring the data, a restart mechanism is triggered to ensure the continuous and stable operation of the equipment, thereby maintaining the efficiency of production links such as sugarcane leaf crushing, fermentation, and resource conversion. This process has extremely high requirements for the real-time performance, completeness, and fault response speed of data processing.

[0003] Existing technologies have significant drawbacks in this scenario: First, the large volume and high frequency of high-frequency data collection mean that traditional processing methods are limited by bandwidth bottlenecks and transmission delays, easily leading to data loss or untimely transmission, and failing to provide complete data support for equipment status monitoring. Second, the environmental parameter formats of different modules vary significantly, lacking unified and standardized processing, increasing the complexity of data analysis, making it difficult to effectively group abnormal data through cluster analysis, resulting in inaccurate fault mode identification. Third, existing technologies are mostly designed for static data analysis, such as the safety production cloud service platform based on support vector machines, which lacks the ability to adapt to the entire process of compression, transmission, and standardization of high-frequency dynamic data. This makes it difficult to accurately locate high-frequency fault points, making it difficult to achieve optimized restart trigger conditions based on fault modes. The system cannot quickly adapt to the dynamic operating status of the equipment, ultimately reducing the production efficiency and operational reliability of the sugarcane leaf recycling equipment.

[0004] To address the aforementioned issues, this application proposes a technical approach that integrates data compression, standardized processing, and cluster analysis. By optimizing the entire process of data acquisition, transmission, analysis, and restart mechanism adjustments, it achieves efficient processing of high-frequency data and accurate fault location, thereby enhancing the communication and collaborative control capabilities of equipment. Summary of the Invention

[0005] This application provides a communication and collaborative control method and system for sugarcane leaf recycling equipment. By optimizing the entire process of data acquisition, transmission, analysis and restart mechanism adjustment, it achieves efficient processing of high-frequency data and accurate fault location, thereby improving the communication and collaborative control capabilities of the equipment.

[0006] In a first aspect, this application provides a method for communication and collaborative control between sugarcane leaf recycling equipment, the method comprising: Step S101: Collect environmental parameter values ​​and parameter deviation data from each module of the sugarcane leaf recycling equipment through a restart mechanism to generate an original dataset; process the environmental parameter values ​​in the original dataset using a data compression algorithm to obtain a compressed dataset; Step S102: If the size of the compressed dataset is less than a preset bandwidth threshold, the compressed dataset is transmitted through the communication channel, and the transmission completion status of the compressed dataset is determined; the compressed dataset that has been transmitted is obtained, and the compressed dataset is restored using a decompression algorithm to obtain the restored dataset; Step S103: For the restored dataset, the parameter formats corresponding to different modules in the restored dataset are converted using a standardized interface to obtain a standardized dataset; clustering analysis algorithm is used to group the data with abnormal environmental parameter values ​​in the standardized dataset to obtain fault mode groups; Step S104: By grouping the fault modes and fusing the parameter deviations in the restored dataset, determine the high-frequency fault point locations of the sugarcane leaf recycling equipment; based on the high-frequency fault point locations, obtain optimized restart trigger conditions; and adjust the threshold parameters in the restart mechanism according to the optimized restart trigger conditions to obtain the adjusted restart mechanism. Step S105: Using the adjusted restart mechanism, collect the updated environmental parameter values ​​and updated parameter deviations from each module to obtain the updated original dataset and generate the running status log.

[0007] Secondly, this application provides a communication and collaborative control system for sugarcane leaf recycling equipment, the system comprising: The acquisition and compression module is used to acquire high-frequency data of environmental parameter values ​​and parameter deviations from the modules through a restart mechanism to obtain the original dataset, and to process the high-frequency part of the environmental parameter values ​​according to the original dataset using a data compression algorithm to obtain a compressed dataset. The transmission restoration module is used to transmit compressed datasets smaller than a preset bandwidth threshold through the communication channel and determine the transmission completion status. At the same time, it uses a decompression algorithm to restore environmental parameter values ​​and parameter deviations to determine the restored dataset. The standardization grouping module is used to convert the environmental parameter values ​​and other parameter formats of different modules into a standardized dataset using a standardized interface, and then uses a clustering analysis algorithm to group the environmental parameter values ​​that are abnormal to obtain fault mode groups based on the standardized dataset. The optimization and adjustment module is used to determine the location of high-frequency fault points by the fault mode grouping fusion parameter deviation, obtain the optimized restart trigger conditions, and adjust the threshold parameters in the restart mechanism according to the optimized restart trigger conditions to obtain the adjusted restart mechanism; The update acquisition module is used to collect and update high-frequency data from each module using an adjustment and restart mechanism to obtain the updated original dataset.

[0008] This application provides a communication and collaborative control method and system for sugarcane leaf recycling equipment, applicable to the communication and collaborative control of multi-module equipment in sugarcane leaf recycling scenarios. It solves the problems of limited high-frequency data transmission bandwidth, inaccurate fault identification due to inconsistent parameter formats across multiple modules, and the system's inability to quickly adapt to dynamic operating states in existing technologies. Compared with existing technologies, the beneficial effects of this application's technical solution are at least as follows: First, by processing the collected high-frequency data with a data compression algorithm, a compressed dataset is generated and transmitted when its size is less than a preset bandwidth threshold. This can overcome bandwidth bottlenecks, reduce data transmission delays and losses, ensure the real-time performance and integrity of high-frequency data transmission, and provide reliable data support for equipment status monitoring. Second, a standardized interface is used to convert the parameter formats of different modules in the restored dataset to generate a standardized dataset. This can eliminate the differences in parameter formats among multiple modules, reduce the complexity of data analysis, and lay the foundation for accurately grouping abnormal environmental parameter values ​​through clustering analysis algorithms, thereby improving the accuracy of fault mode identification. Third, by locating high-frequency fault points through fault mode grouping and fusion parameter deviation, and then optimizing the restart trigger conditions and adjusting the restart mechanism, it is possible to quickly adapt to the dynamic operating status of the equipment, solve the problem that existing technologies cannot optimize restart trigger conditions based on fault modes, and improve the operational reliability and production efficiency of sugarcane leaf recycling equipment. Attached Figure Description

[0009] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a flowchart of a communication and collaborative control method between sugarcane leaf recycling equipment according to this application; Figure 2 This is a flowchart illustrating the compression and restoration of data during the communication and collaborative control process between sugarcane leaf recycling devices in this application; Figure 3 This is a flowchart illustrating the process of updating the original dataset during the communication and collaborative control between sugarcane leaf recycling devices in this application; Figure 4 This is a structural diagram of a communication and collaborative control system between sugarcane leaf recycling equipment according to this application. Detailed Implementation

[0011] This application provides a method and system for communication and collaborative control between sugarcane leaf recycling devices. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or devices.

[0012] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of a communication and collaborative control method between sugarcane leaf recycling devices in this application includes: Step S101: Collect environmental parameter values ​​and parameter deviation data from each module of the sugarcane leaf recycling equipment through a restart mechanism to generate an original dataset; process the environmental parameter values ​​in the original dataset using a data compression algorithm to obtain a compressed dataset.

[0013] In one specific embodiment, step S101, generating the original dataset, may specifically include the following steps: By using a restart mechanism, environmental parameters and parameter deviations are obtained from each module to generate a high-frequency raw dataset. Determine whether the acquisition frequency of the high-frequency raw dataset is lower than a preset frequency threshold. If so, adjust the restart mechanism to increase the acquisition frequency to obtain a stable high-frequency raw dataset. Based on the stable high-frequency raw dataset, time series analysis is used to extract the changing trends of environmental parameters and parameter deviations in the stable high-frequency raw dataset, and to determine the parameter fluctuation pattern. Based on the parameter fluctuation pattern, the K-means clustering algorithm is used to classify the operating status of different modules of the sugarcane leaf recycling equipment and obtain module status groups. Determine whether the module status group displays an anomaly. If so, extract environmental parameters and parameter deviation data from the abnormal module status group to determine the source of the anomaly. Based on the sources of the anomalies, the support vector machine algorithm is used to predict potential fault points of the sugarcane leaf recycling equipment and generate fault prediction results. Based on the fault prediction results, the parameter configuration of the restart mechanism is adjusted, the data acquisition process is optimized, and an improved dataset is obtained, which is the original dataset.

[0014] Specifically, in the operation scenario of the sugarcane leaf recycling equipment, after the restart mechanism is activated, environmental parameters and parameter deviations are obtained from various functional modules of the equipment, such as the temperature acquisition module, humidity monitoring module, power speed module, and crushing module. The environmental parameters include temperature values ​​(e.g., 28℃~35℃), humidity values ​​(e.g., 55%~70%), speed values ​​(e.g., 1200r / min~1500r / min), and crushing particle size parameters (e.g., 2mm~5mm). The parameter deviation is calculated by "parameter deviation = target parameter value - measured parameter value". For example, when the target temperature value is set to 30℃, the deviation corresponding to the measured temperature of 32℃ is -2℃. When the target speed value is set to 1350r / min, the deviation corresponding to the measured speed of 1400r / min is -50r / min. These environmental parameters and parameter deviations are integrated to form a high-frequency raw dataset. Determine whether the sampling frequency of the high-frequency raw dataset is lower than the preset frequency threshold (e.g., preset 2 times / second). If the current sampling frequency is 1 time / second, which is lower than the preset threshold, adjust the parameters related to the sampling interval in the restart mechanism to shorten the sampling interval and increase the sampling frequency until a stable high-frequency raw dataset with a sampling frequency that is stably maintained at 2 times / second is obtained.

[0015] Based on a stable, high-frequency raw dataset, time series analysis is employed to chronologically analyze the changes in environmental parameters and parameter deviations. For example, temperature values ​​and corresponding deviations are recorded every 2 seconds for one hour to extract trends over time, such as a temperature increase of 0.5℃ every 10 minutes and a rotational speed deviation increase of 10 r / min every 15 minutes. This helps determine parameter fluctuation patterns. Using these patterns as input, K-means clustering is applied, with the numerical characteristics of environmental parameters and parameter deviations as the clustering criteria. For instance, module operation data with temperatures between 28℃ and 32℃ and absolute deviations ≤ 2℃, and rotational speeds between 1300 r / min and 1400 r / min and absolute deviations ≤ 30 r / min, are grouped together. This categorizes the operational states of different modules, obtaining module state groups. Clustering provides a clear division of module operational states, offering a basis for anomaly identification.

[0016] The system determines whether module status groups display abnormalities. If a group's temperature value consistently exceeds 35℃ with an absolute deviation ≥ 5℃, or its rotational speed consistently falls below 1200 r / min with an absolute deviation ≥ 50 r / min, then that group is identified as an abnormal module status group. Corresponding environmental parameters and parameter deviation data are extracted from this abnormal group. By comparing the numerical range and fluctuation patterns of normal group data (e.g., normal group temperature deviations are mostly within 2℃, while abnormal group deviations exceed 5℃), the source of the anomaly is determined. For example, abnormal temperatures may originate from a heat dissipation module malfunction, or abnormal rotational speeds may be due to unstable power supply to the power module. Based on the source of the anomaly, the environmental parameters, parameter deviations, and corresponding module identifiers in the abnormal group are used as sample data and input into a support vector machine algorithm. The algorithm constructs a classification model to predict potential fault points, such as predicting a shutdown failure of the heat dissipation module within one hour or a further decrease in the power module's power supply voltage, generating a fault prediction result.

[0017] Based on the fault prediction results, the parameter configuration of the restart mechanism is adjusted. For example, for the heat dissipation module fault warning, the temperature parameter collection frequency is increased to 3 times / second; for the power module warning, the monitoring range of speed deviation is expanded. The data acquisition process is optimized to ensure that the collected data can cover parameter changes related to potential faults, resulting in an improved dataset, which becomes the original dataset. By adjusting the acquisition parameters in conjunction with the fault prediction results, the collected data better matches the fault monitoring needs, improving the effectiveness of data acquisition.

[0018] In one specific embodiment, step S101, obtaining the compressed dataset, may specifically include the following steps: The environmental parameter values ​​in the original dataset are processed according to the data compression algorithm to generate an initial compressed dataset; If the compression ratio of the initial compressed dataset is lower than the preset compression ratio threshold, the sampling interval of the data compression algorithm is adjusted to optimize the data compression process and obtain an optimized compressed dataset. Based on the optimized compressed dataset, principal component analysis algorithm is used to extract the main features of the environmental parameter values ​​to obtain a feature dataset; If the feature dimension of the feature dataset is higher than a preset dimension threshold, then the feature dataset is subjected to dimensionality reduction processing to obtain a dimensionality-reduced dataset. Based on the reduced-dimensionality dataset, time series analysis is used to extract the time variation trend of the environmental parameter values ​​to obtain the variation trend. Based on the aforementioned trend, the K-means clustering algorithm is used to classify the operating states in different time periods to obtain state groups. If the state grouping shows an anomaly, the corresponding feature data is extracted from the anomaly group and the source of the anomaly is determined. An anomaly analysis result is generated and then merged with the dimensionality reduction dataset to obtain a compressed dataset.

[0019] Specifically, for the acquired raw dataset, which contains environmental parameter values ​​and deviations collected from various modules of the device, a data compression algorithm (such as the Zstandard algorithm) is used to process the environmental parameter values. Redundancy is removed through encoding (e.g., for temperature values ​​with fluctuations ≤0.2℃ within 5 seconds, only the start and end data are retained), generating an initial compressed dataset. If the compression ratio of the initial compressed dataset (e.g., the current compression ratio of 50%) is lower than a preset threshold (e.g., a preset threshold of 65%), the sampling interval of the data compression algorithm is adjusted, for example, from 1 second / sample to 2 seconds / sample, reducing redundant data collection and optimizing the data compression process to obtain an optimized compressed dataset. Based on the optimized compressed dataset, principal component analysis is used to calculate the variance contribution of each environmental parameter value, such as 45% for temperature, 35% for rotational speed, and 20% for humidity. Temperature and rotational speed values ​​with variance contributions higher than a preset percentage (e.g., 30%) are extracted as primary features to obtain a feature dataset. If the feature dimensions of the feature dataset (e.g., the current 4 dimensions) are higher than the preset threshold (e.g., the preset threshold is 2 dimensions), then a dimensionality reduction algorithm (e.g., PCA dimensionality reduction) is used to process the feature dataset. Through linear transformation, the two-dimensional data of temperature and rotation speed values ​​are converted into a single comprehensive feature dimension to obtain a dimensionality-reduced dataset, which further reduces the data dimensionality, reduces the amount of computation in subsequent data processing, and adapts to the limited computing resources of the device.

[0020] Based on the dimensionality-reduced dataset, time series analysis was employed to analyze the changes in comprehensive feature data in chronological order (e.g., every 10 minutes as a time window). For example, the comprehensive feature value increased from 1.2 to 1.8 between 10:00 and 10:10, and decreased from 1.8 to 1.5 between 10:10 and 10:20. The temporal trend of environmental parameter values ​​was extracted, reflecting the fluctuation pattern of the equipment's operating status in different time periods. Using the trend analysis, K-means clustering was applied, with the magnitude and rate of change of the comprehensive feature value as the clustering basis. The operating status in different time periods was divided into stable (change magnitude ≤ 0.3), fluctuating (change magnitude 0.3-0.6), and abnormal (change magnitude > 0.6) groups, thus enabling the classification and identification of equipment operating status and providing a basis for anomaly detection.

[0021] If a status group shows an anomaly (e.g., a change in the comprehensive feature value of 0.7 between 10:20 and 10:30, classified as an anomaly group), the corresponding comprehensive feature data is extracted from the anomaly group. Combined with the original environmental parameter values ​​(e.g., temperature rising to 36℃ or speed dropping to 1100 r / min during this time period), the source of the anomaly is determined (e.g., temperature anomaly due to heat dissipation module blockage, speed anomaly due to unstable motor power supply). This yields the anomaly analysis results, ultimately determining a compressed dataset containing key features and anomaly information. This compressed dataset retains key data while controlling its size through compression and dimensionality reduction, enabling efficient transmission to subsequent stages. Furthermore, the anomaly analysis results provide direction for equipment fault diagnosis, addressing the problem of unclear fault location in existing technologies.

[0022] Step S102: If the size of the compressed dataset is less than a preset bandwidth threshold, the compressed dataset is transmitted through the communication channel, and the transmission completion status of the compressed dataset is determined; the compressed dataset that has been transmitted is obtained, and the compressed dataset is restored using a decompression algorithm to obtain the restored dataset.

[0023] In one specific embodiment, step S102 may specifically include the following steps: If the storage capacity of the compressed dataset is lower than a preset capacity threshold, the compressed dataset is transmitted through the communication channel, the transmission completion status is determined, and the transmission result is obtained. Based on the transmission results, the compressed dataset is divided into multiple data packets using data packetization technology, and the packet sequence is determined. If the transmission rate of the packet sequence is lower than the bandwidth limit of the communication channel, the size of the data packet is adjusted, and an optimized packet sequence is regenerated to obtain an optimized packet. The integrity of the compressed dataset is verified using the optimized packet segmentation and a verification algorithm to determine whether data loss occurs during transmission and obtain the verification result. If the verification result shows that the compressed dataset has been transmitted completely, the optimized packets are merged using the transmission protocol to obtain the complete compressed dataset. The complete compressed dataset is decoded using a decompression algorithm to restore the environmental parameter values. Verify whether the restored data matches the original dataset. After removing outliers, determine the restored dataset.

[0024] Specifically, please refer to Figure 2For the generated compressed dataset (e.g., a compressed package containing environmental parameter values ​​such as temperature, humidity, and the rotation speed of the sugarcane leaf crushing module, along with corresponding parameter deviations, with a storage capacity of approximately 0.8MB), the first step is to determine if its storage capacity is lower than a preset threshold (e.g., a preset threshold of 1.0MB). If the condition is met, the compressed dataset is transmitted through a communication channel between devices (e.g., industrial Ethernet), while simultaneously monitoring the data transmission progress in real time (e.g., the number of bytes transmitted / total number of bytes), determining whether the compressed dataset has been completely transmitted to the target end, and obtaining the transmission result (e.g., "transmission complete" or "transmission interrupted"). Based on the transmission result, if it is "transmission complete," data packetization technology is used to divide the compressed dataset into multiple data packets of a fixed size (e.g., 200KB per packet) (e.g., 0.8MB compressed data divided into 4 packets), and each packet is labeled with a sequence number (e.g., 1-4) according to the division order, thus determining the packet sequence. If the transmission rate of the packet sequence during transmission (e.g., currently 100KB / s) is lower than the bandwidth limit of the communication channel (e.g., a channel bandwidth limit of 200KB / s), the size of the data packets is adjusted (e.g., from 200KB to 300KB per packet). This reduces the number of packets, lowers transmission interaction time, and regenerates the optimized packet sequence, resulting in optimized packets. This adjustment process addresses the problem of data transmission delay caused by insufficient transmission rate, improving transmission efficiency by adapting to bandwidth.

[0025] By optimizing packet segmentation, a verification algorithm (such as CRC32) is used to calculate the data stream of each packet, generating a check value (such as a unique 32-bit checksum for each packet). The check value calculated by the receiving end is compared with the check value attached by the sending end to determine whether data loss has occurred during transmission (if the check values ​​are inconsistent, it is determined that data loss has occurred), and a verification result (such as "data complete" or "data missing") is obtained. If the verification result shows that the compressed dataset is complete, the optimized packets are merged sequentially according to the packet sequence number using a transmission protocol (such as TCP protocol) to form a complete compressed dataset consistent with the initial transmission. After obtaining the complete compressed dataset after transmission, a decompression algorithm (such as Zstandard decompression algorithm) matching the data compression algorithm in step S101 is used to decode it, and the original environmental parameter values ​​(such as temperature 25~35℃, humidity 50%~70%, rotation speed 1200~1500r / min) and parameter deviations (such as temperature deviation -5~5℃, rotation speed deviation -150~150r / min) are restored in reverse. Through consistency verification, the restored environmental parameter values ​​and parameter deviations are compared one by one with the corresponding data in the original dataset generated in step S101 (e.g., if the absolute value of the difference between the restored temperature value and the original temperature value is ≤0.1℃, a match is determined). Abnormal values ​​that exceed the preset allowable range are removed (e.g., abnormal data where the difference between the restored temperature value and the original value reaches 2℃), and the restored dataset is finally determined. Consistency verification solves the data distortion problem that may occur during decompression and ensures the consistency between the restored data and the original data.

[0026] Step S103: For the restored dataset, the parameter formats corresponding to different modules in the restored dataset are converted using a standardized interface to obtain a standardized dataset; clustering analysis algorithm is used to group the data with abnormal environmental parameter values ​​in the standardized dataset to obtain fault mode groups.

[0027] In one specific embodiment, obtaining the standardized dataset in step S103 may specifically include the following steps: The restored dataset is parsed using a standardized interface to obtain the environmental parameter values ​​and parameter formats, and an initial conversion dataset is generated. Based on the initial conversion dataset, data mapping technology is used to match the environmental parameter values ​​with preset format specifications to obtain a mapped dataset; If the mapped dataset conforms to the format specification, the data consistency of each module is integrated through interface call technology to generate an integrated dataset; Based on the integrated dataset, data verification technology is used to determine the data integrity and consistency of the environmental parameter values ​​and parameter deviations, resulting in a verification dataset. Based on the verification dataset, a clustering analysis algorithm is used to group the environmental parameter values ​​to determine the parameter group set; If the parameter grouping set meets the preset grouping threshold, the parameter mapping is merged through data aggregation technology to generate a standardized dataset.

[0028] Specifically, for the obtained restored dataset (which includes environmental parameter values ​​such as temperature, humidity, rotation speed of the sugarcane leaf crushing module, and motor current of the conveying module, as well as the corresponding deviation values ​​of each parameter, and the parameter formats of different modules are different), a standardized interface (such as the general data format interface for industrial equipment) is used to parse the restored dataset, extract the environmental parameter values ​​and the format information (such as unit type and decimal places) of each parameter, and generate an initial converted dataset. This dataset still retains the original format differences of the parameters of each module, but the association mapping between parameters and formats has been completed.

[0029] By initially transforming the dataset, data mapping technology is used to match the values ​​of various environmental parameters with preset format specifications (such as uniformly setting temperature and humidity to retain two decimal places, speed and current to retain integers, and uniformly placing the unit label after the parameter value). For example, the original "32.5℃" in the temperature module is mapped to "32.50℃", and the original "1450r / min" in the speed module remains in the format of "1450r / min", resulting in a mapped dataset. If the format of all parameters in the mapped dataset conforms to the preset format specifications (such as no missing units or deviations in decimal places), then the mapped data of each module is associated through interface calling technology (such as calling the device data integration interface) to ensure the correspondence of parameters of different modules at the same time node (such as temperature 32.50℃, speed 1450r / min, and current 15.20A at 10:00), integrating data consistency and generating an integrated dataset.

[0030] Based on the integrated dataset, data validation techniques (such as field non-empty validation and numerical range validation) are used to determine the integrity and consistency of environmental parameter values ​​and parameter deviations. For example, it verifies whether the temperature value is within a reasonable range of 0-60℃ and whether the parameter deviation calculation conforms to the logic of "deviation = target value - measured value". Null values ​​and logically erroneous data are removed to obtain a validation dataset. Using the validation dataset, clustering analysis algorithms (such as K-means clustering) are employed to group the environmental parameter values ​​based on their numerical range. For example, temperatures of 28–32℃, rotational speeds of 1400–1500 r / min, and currents of 14–16A are classified as normal parameter groups, while temperatures >35℃ or <25℃ and rotational speeds >1600 r / min or <1300 r / min are classified as abnormal parameter groups, thus determining the parameter grouping set. The clustering analysis algorithm enables the classification and management of parameters, solving the problem of the lack of a unified standard for parameter status judgment in existing technologies. If the number of groups and the amount of data in each group of the parameter grouping set meet the preset thresholds (e.g., the proportion of normal group data ≥ 80%, and abnormal group data can be clearly distinguished), then the mapping relationships of related parameters within the same group are merged using data aggregation technology (e.g., merging and recording the format mapping rules of abnormal temperature parameters and abnormal speed parameters in the abnormal group), generating a standardized dataset. All parameters in this dataset have a unified format and clear grouping. Based on the standardized dataset, condition monitoring technology (e.g., monitoring the stability after data format conversion and the repeatability of grouping results) is used to analyze the stability of the data conversion, obtaining stable analysis results (e.g., format uniformity ≥ 99% after 10 consecutive conversions, and consistent grouping results), ensuring the effectiveness of the standardized dataset. Condition monitoring technology solves the problems of format rollback and grouping confusion that easily occur after data conversion, ensuring that the standardized dataset can provide reliable data support for subsequent fault mode analysis, and further solving the problems of low reliability and inability to support fault identification after data processing in existing technologies.

[0031] In one specific embodiment, obtaining the fault mode grouping in step S103 may specifically include the following steps: The environmental parameter values ​​in the standardized dataset are cleaned to generate a preprocessed dataset; The preprocessed dataset was format-converted using standardization techniques. The environmental parameter values ​​in the transformed preprocessed dataset are grouped using the K-means clustering algorithm to generate initial parameter groups; If the initial parameter grouping meets the anomaly threshold setting, then the abnormal parameters in the environmental parameter values ​​in the preprocessed dataset are identified by anomaly detection technology to obtain the abnormal parameter grouping; Based on the grouping of the abnormal parameters, the decision tree algorithm is used to classify the abnormal parameters and generate a fault mode classification. Based on the aforementioned fault mode classification, condition monitoring technology is used to analyze the distribution characteristics of the abnormal parameters and obtain the abnormal distribution results. Based on the abnormal distribution results, the fault modes are merged using data aggregation technology to obtain fault mode groups.

[0032] Specifically, for the obtained standardized dataset (containing environmental parameter values ​​such as temperature, humidity, crushing module speed, and conveying module pressure in a uniform format, e.g., temperature is retained to two decimal places in °C, and speed is retained as an integer in r / min), the environmental parameter values ​​are first cleaned to remove outliers caused by sensor fluctuations (such as sudden temperature changes or negative speed data) and duplicate records (such as the same speed data collected repeatedly at the same time point), generating a preprocessed dataset. Standardization techniques (such as Z-score standardization) are used to convert the format of the preprocessed dataset, calculating the mean and standard deviation of each environmental parameter value. The parameter values ​​are then converted into dimensionless data that meets the analytical requirements using the formula "standardized value = (original value - mean) / standard deviation," ensuring the data is compatible with the subsequent K-means clustering algorithm. This conversion process eliminates the influence of differences in parameter magnitudes on the clustering results, solving the problem of inaccurate clustering due to different magnitudes of multiple parameters.

[0033] The environmental parameter values ​​in the transformed preprocessed dataset are grouped using the K-means clustering algorithm. The number of clusters is set (e.g., 3: normal, slightly abnormal, severely abnormal). The Euclidean distance between the standardized value of each parameter and the cluster center is calculated. Parameters with the closest distance are grouped together to generate initial parameter groups. For example, parameters with standardized values ​​between -1 and 1 are grouped into the normal group, those between 1 and 2 into the slightly abnormal group, and those greater than 2 into the severely abnormal group. If the proportion of data in the slightly abnormal and severely abnormal groups in the initial parameter groups meets the anomaly threshold (e.g., ≥5%), anomaly detection techniques (e.g., the Isolation Forest algorithm) are used to further identify anomalous parameters in the environmental parameter values. Parameters that deviate from the normal data distribution (e.g., stress parameters with standardized values ​​greater than 2.5) are filtered out, resulting in anomalous parameter groups.

[0034] Based on the abnormal parameter grouping, a decision tree algorithm is used to classify the abnormal parameters. A decision tree model is constructed using parameter type (temperature, speed, pressure, etc.) and abnormality degree (standardized value size) as classification features. For example, when the parameter type is temperature and the standardized value is >2.5, it is determined as "heat dissipation fault-related anomaly"; when the parameter type is speed and the standardized value is <-2, it is determined as "power deficiency-related anomaly," generating a fault mode classification. Through fault mode classification, condition monitoring technology is used to analyze the distribution characteristics of abnormal parameters, statistically analyzing the occurrence time of parameters under different fault modes (e.g., "heat dissipation fault-related anomaly" mostly appears after 2 hours of equipment operation) and the number of modules involved (e.g., involving 2 temperature monitoring modules), obtaining anomaly distribution results. Based on the anomaly distribution results, data aggregation technology is used to merge related fault modes (e.g., "heat dissipation fault-related anomaly" and "temperature sensor anomaly-related anomaly" are both concentrated in the heat dissipation module area, merged into "heat dissipation system related fault mode"), generating fault mode groupings. This aggregation process achieves accurate correspondence between fault modes and equipment problems through decision tree classification and status monitoring, solving the technical defects of existing technologies such as ambiguous fault mode identification and inability to locate specific equipment problems, and providing clear fault basis for subsequent optimization of restart trigger conditions.

[0035] Step S104: By grouping the fault modes and fusing the parameter deviations in the restored dataset, determine the high-frequency fault location of the sugarcane leaf recycling equipment; based on the high-frequency fault location, obtain the optimized restart trigger condition; and adjust the threshold parameter in the restart mechanism according to the optimized restart trigger condition to obtain the adjusted restart mechanism.

[0036] In one specific embodiment, obtaining the optimized restart trigger condition in step S104 may specifically include the following steps: The parameter deviation data is preprocessed using data cleaning techniques to remove noisy data and obtain a cleaned deviation dataset. Time series analysis techniques are used to perform time series modeling on the post-cleaning deviation dataset to generate time series data. Based on the time series data, the K-means clustering algorithm is used to extract features from the fault mode groups to obtain the location distribution features. If the distribution characteristics of the points meet the preset distribution threshold, then the high-frequency fault points are identified by anomaly detection technology, and the locations of the high-frequency fault points are obtained. Based on the location of the high-frequency fault points, a decision tree algorithm is used to classify the equipment status information and generate status change trends; By analyzing the state change trends, data aggregation technology is used to fuse the fault occurrence frequency and trigger condition thresholds to obtain optimized restart trigger conditions.

[0037] Specifically, the obtained parameter deviation data is preprocessed using data cleaning techniques (such as the moving average method). The average value of five consecutive collection points is calculated to replace the original fluctuation data, and noisy data exceeding reasonable ranges (such as data with sudden temperature deviations of ±20℃ or speed deviations of ±100r / min) are removed to obtain a cleaned deviation dataset. Time series analysis techniques (such as the ARIMA model) are used to perform time series modeling on the cleaned deviation dataset. With time as the horizontal axis and deviation value as the vertical axis, a curve showing the change of deviation data over time is fitted. For example, the temperature deviation remains at -2 to 2℃ for 1 to 2 hours of equipment operation, and gradually expands to -3 to 3℃ for 2 to 3 hours, generating time series data. Based on the time series data, the K-means clustering algorithm is used to extract features from fault mode groups (such as fault modes related to the heat dissipation system and fault modes related to the power module). The deviation value and the time point of occurrence at the time of fault are used as clustering features to calculate the cluster centers of deviation data under different fault modes, obtaining the point distribution characteristics. If the proportion of clustered data for the same fault mode in the location distribution characteristics meets a preset threshold (e.g., accounting for 70% of the total data for that fault mode), then density analysis of the location distribution characteristics is performed using anomaly detection technology (e.g., the DBSCAN algorithm) to identify fault occurrence areas with high data density (e.g., areas with a temperature deviation of 4℃ and a time of approximately 2.5 hours have significantly higher data density than other areas), high-frequency fault points are identified, and their locations are determined (e.g., the temperature monitoring point of the heat dissipation module, the speed monitoring point of the power module). Based on the location of the high-frequency fault points, the equipment status information of the corresponding points is collected (e.g., the cooling fan speed of the temperature monitoring point, the motor current of the speed monitoring point), and a decision tree algorithm is used to classify the equipment status information, using the numerical range of the status information as the classification node. For example, when the cooling fan speed is <1200r / min and the motor current is >16A, it is determined as a "high fault risk state," and a status change trend is generated (e.g., the time interval between the equipment status at the high-frequency fault point location changing from "normal" to "high fault risk state" is approximately 30 minutes).

[0038] By analyzing state change trends, data aggregation technology is used to fuse fault occurrence frequency (e.g., a fault occurring once every 2 hours at the temperature monitoring point of the heat dissipation module) and trigger condition thresholds (e.g., the original restart trigger conditions were temperature deviation ≥5℃ and speed deviation ≥50r / min). The fault occurrence frequency is then correlated with and adjusted to these thresholds. For example, for a high-frequency fault occurring once every 2 hours, the temperature deviation trigger threshold is adjusted to ≥4℃ and the speed deviation trigger threshold to ≥40r / min, resulting in optimized restart trigger conditions. The application of data aggregation technology addresses the mismatch between restart trigger conditions and actual fault frequencies in existing technologies, making the trigger conditions more closely aligned with equipment fault patterns. Furthermore, through decision tree classification and anomaly detection, it combines high-frequency fault location with state trend analysis, further resolving the technical deficiency of existing restart mechanisms that cannot specifically respond to high-frequency faults.

[0039] In one specific embodiment, step S104, obtaining the adjusted restart mechanism, may specifically include the following steps: Extract key parameters from the optimized restart trigger conditions to generate an initial threshold set; Distribution characteristics are obtained from the initial threshold set, and a threshold distribution model is constructed using statistical analysis techniques; Based on the threshold distribution model, determine the parameter adjustment range and obtain the adjustment range set; If the set of adjustment ranges meets the preset constraints, then the threshold parameters in the initial threshold set are optimized by a linear regression algorithm to generate an optimized threshold set. Based on the optimized threshold set, the logic of the restart mechanism is updated to generate a restart mechanism after initial adjustment; Operation log data is obtained from the initial adjustment and restart mechanism, and time series analysis technology is used to process the operation log data to generate operation status trends; Based on the operating status trend, the stability of the initial adjustment and restart mechanism is determined. If the stability meets the requirements, it is determined to be an adjustment and restart mechanism.

[0040] Specifically, for the obtained optimized restart trigger conditions, key parameters are extracted, namely the trigger threshold values ​​corresponding to each parameter, to generate an initial threshold set (e.g., {temperature deviation: 4℃, speed deviation: 40r / min, humidity deviation: 5%}). Distribution characteristics are obtained from the initial threshold set, and the frequency of each threshold occurrence and the proportion of associated fault types in historical fault data are statistically analyzed. For example, a temperature deviation of 4℃ corresponds to a 60% proportion of "heat dissipation faults," and a speed deviation of 40r / min corresponds to a 55% proportion of "power faults." Statistical analysis techniques (such as frequency distribution statistics and correlation analysis) are used to construct a threshold distribution model, which includes the mapping relationship between each threshold and fault type and fault frequency. Based on the threshold distribution model, and combined with the hardware tolerance range of each module of the equipment (e.g., temperature deviation adjustment cannot be lower than 2℃, otherwise frequent restarts are likely; speed deviation adjustment cannot be higher than 50r / min, otherwise fault warnings are likely to be missed), the parameter adjustment range is determined, resulting in an adjustment range set. This range setting solves the problem of boundless threshold adjustment and easy exceeding of the equipment's adaptability range in existing technologies, ensuring the feasibility of the adjusted thresholds.

[0041] If the set of adjustment ranges meets preset constraints (e.g., each parameter adjustment range covers more than 90% of the effective thresholds in historical faults, and adjacent parameter adjustment ranges do not conflict), then the threshold parameters in the initial threshold set are optimized using a linear regression algorithm. Using the fault occurrence frequency as the dependent variable and each threshold parameter as the independent variable, a linear regression model is constructed (e.g., fault frequency = α × temperature deviation + β × speed deviation + γ × humidity deviation + ε, where α, β, and γ are regression coefficients, and ε is the error term). The optimal regression coefficients are solved using the least squares method, and then the threshold parameters that control the fault frequency within a preset reasonable range are calculated, generating an optimized threshold set (e.g., {temperature deviation: 3.5℃, speed deviation: 38 r / min, humidity deviation: 4.5%}). The application of the linear regression algorithm solves the problem of threshold adjustment relying on experience and lacking data support in existing technologies, making threshold optimization more closely aligned with actual fault patterns.

[0042] Based on the optimized threshold set, the restart mechanism logic is updated. The original trigger logic of "temperature deviation ≥ 4℃, speed deviation ≥ 40r / min, or humidity deviation ≥ 5% triggers restart" is adjusted to "temperature deviation ≥ 3.5℃, speed deviation ≥ 38r / min, or humidity deviation ≥ 4.5% triggers restart." A delay judgment is added after threshold triggering (e.g., triggering only after three consecutive data collection cycles reach the threshold), generating a preliminary adjusted restart mechanism. Operation log data (including the trigger time for each threshold, parameter values ​​at trigger, and the device's recovery status after restart, such as a 3.5℃ temperature deviation triggering restart at 10:00, and the device returning to normal operation at 10:05) is obtained from the preliminary adjusted restart mechanism. Time series analysis is used to process the operation log data to generate an operation status trend. Based on the operation status trend, the stability of the preliminary adjusted restart mechanism is determined. If the stability meets the requirements (e.g., restart frequency fluctuation ≤ 1 time within 24 hours, and device recovery time ≤ 5 minutes), it is confirmed as an adjusted restart mechanism. This stability verification addresses the issues of unverified effects and mechanism instability in existing technologies after threshold adjustment. It ensures that the restart mechanism can operate reliably after adjustment, further improving the effectiveness of communication and collaborative control of sugarcane leaf recycling equipment and avoiding production interruptions or missed fault detections due to unreasonable restart mechanisms.

[0043] Step S105: Using the adjusted restart mechanism, collect the updated environmental parameter values ​​and updated parameter deviations from each module to obtain the updated original dataset and generate a running status log.

[0044] In one specific embodiment, step S105 may specifically include the following steps: Based on the aforementioned adjustment and restart mechanism, updated high-frequency data is collected from each module. The collected high-frequency data is then filtered using data filtering technology to remove invalid data and generate the updated original dataset. If the updated original dataset meets the preset integrity conditions, redundant information in the updated original dataset is removed by data cleaning technology to obtain the cleaned dataset. Based on the cleaned dataset, a clustering analysis algorithm is used to divide the data into categories to obtain a categorized dataset. Key features are extracted from the classification dataset, and the extracted key features are filtered using feature selection technology to generate a feature dataset; If the feature dataset meets the preset correlation threshold, then the feature dataset is analyzed using time series analysis technology to generate data change trends; Based on the data change trend, a decision tree algorithm is used to determine the operating status of each module and obtain the status evaluation results; Abnormal state data is extracted from the state assessment results, the abnormal state data is recorded, and an operation status log is generated.

[0045] Specifically, please refer to Figure 3 Based on the adjusted restart mechanism, updated high-frequency data is collected from each module of the device at a frequency of 2 times per second. The data includes temperature, rotation speed, humidity, current, and corresponding deviation values. Data filtering techniques (such as setting numerical ranges for filtering; temperatures exceeding 0-60℃ and rotation speeds exceeding 800-2000 r / min are considered invalid) are used to filter the collected high-frequency data, removing invalid data caused by sensor malfunctions, thus generating an updated original dataset. If the updated original dataset meets preset integrity conditions (such as a data missing rate of less than 5% and no interruptions exceeding 10 seconds in continuous data collection from each module), data cleaning techniques (such as deleting duplicate records at the same timestamp and replacing isolated outliers with the mean of five adjacent valid data points) are used to remove redundant information from the updated original dataset, resulting in a cleaned dataset. Based on the cleaned dataset, clustering analysis algorithms (such as K-means clustering, with 3 clusters representing normal, minor anomalies, and severe anomalies) are used to categorize the data. Each parameter value and deviation is used as a clustering feature. The Euclidean distance between data points and cluster centers is calculated, and the closest points are grouped into the same cluster, resulting in a categorized dataset. Key features (such as temperature deviation, speed deviation, and current values, which are more correlated with equipment failures than other parameters) are extracted from the categorized dataset. Feature selection techniques (such as correlation analysis, calculating the Pearson correlation coefficient between each feature and the fault label, and retaining features with an absolute correlation coefficient greater than 0.6) are used to filter the extracted key features, generating a feature dataset. Feature selection addresses the problem of excessive computation due to too many parameter dimensions, focusing on key influencing factors.

[0046] If the feature dataset meets a preset correlation threshold (e.g., the average correlation coefficient between key features and fault labels ≥ 0.7), time series analysis techniques (such as the ARIMA model) are used to analyze the feature dataset, sorting out the changes in feature data by timestamp to generate data change trends. Based on the data change trends, a decision tree algorithm is used to classify the operating status of each module, using the magnitude and rate of change of feature data as classification nodes to obtain status assessment results. Abnormal status data (e.g., temperature value of 33℃ and deviation value of 3℃ corresponding to a slight abnormality in the temperature module, and current value of 17A and deviation value of 2A corresponding to a severe abnormality in the current module) are extracted from the status assessment results. The timestamp of the abnormality, module identifier, abnormal parameter value, deviation value, and corresponding status assessment results are recorded to generate an operating status log. The log records solve the problem of no traceability for abnormal data, and at the same time update the original dataset to integrate effective, cleaned, and classified key data, providing reliable data support for subsequent equipment collaborative control and fault handling, and further solving the problem of equipment control response lag caused by imperfect data processing links in existing technologies.

[0047] The above describes the communication and collaborative control system method between sugarcane leaf recycling equipment in the embodiments of this application. Please refer to [link / reference]. Figure 4 The following describes the communication and collaborative control system 400 between sugarcane leaf recycling equipment in the embodiments of this application, including: The acquisition and compression module 401 is used to acquire environmental parameter values ​​and high-frequency data of parameter deviations from each module through a restart mechanism to obtain the original dataset, and to process the high-frequency part of the environmental parameter values ​​according to the original dataset using a data compression algorithm to obtain a compressed dataset. The transmission restoration module 402 is used to transmit compressed datasets smaller than a preset bandwidth threshold through a communication channel and determine the transmission completion status. At the same time, it uses a decompression algorithm to restore environmental parameter values ​​and parameter deviations to determine the restored dataset. The standardization grouping module 403 is used to convert the environmental parameter values ​​and other parameter formats of different modules into a standardized dataset using a standardized interface, and to obtain a fault mode group by grouping the abnormal environmental parameter values ​​based on the standardized dataset using a clustering analysis algorithm. The optimization and adjustment module 404 is used to determine the location of high-frequency fault points by the fault mode grouping fusion parameter deviation to obtain the optimized restart trigger condition, and adjust the threshold parameter in the restart mechanism according to the optimized restart trigger condition to obtain the adjusted restart mechanism; Update the acquisition module 405 to obtain the updated original dataset by acquiring and updating high-frequency data from each module using an adjustment and restart mechanism.

[0048] Through the collaborative efforts of the aforementioned components, a closed-loop management system is achieved, encompassing multi-module high-frequency data acquisition, compression processing, compliant transmission and restoration of the sugarcane leaf recycling equipment; parameter format standardization conversion; abnormal data clustering and grouping; high-frequency fault point location and restart trigger condition optimization; and high-frequency data update acquisition and operation status log generation based on the adjusted restart mechanism. Specifically: the acquisition and compression module 401 collects environmental parameter values ​​and deviations from various functional modules such as the temperature acquisition module, humidity monitoring module, and power speed module through a restart mechanism. After forming the original dataset, it is processed by a data compression algorithm to generate a compressed dataset, completing the initial high-frequency data acquisition and volume optimization, laying the foundation for bandwidth adaptation in subsequent transmission stages. The transmission and restoration module 402 determines the capacity of the compressed dataset and transmits it through the communication channel when it meets the preset capacity threshold. Combining data packetization technology, verification algorithms, and decompression algorithms, it achieves complete data transmission and accurate restoration, avoiding data loss or distortion, ensuring the reliability of data from the acquisition end to the processing end, and compensating for the shortcomings of traditional transmission methods in guaranteeing data integrity. The standardization and grouping module 403 uses a standardized interface to unify the parameter formats of different modules. After generating a standardized dataset, it uses clustering analysis algorithms, anomaly detection technology, and decision tree algorithms. Abnormal data is grouped and fault mode identified to eliminate the analytical complexity caused by differences in parameter formats across multiple modules, providing a clear grouping basis for fault location. The optimization and adjustment module 404 integrates fault mode grouping and parameter deviation data, uses time series analysis, K-means clustering, and linear regression algorithms to locate high-frequency fault points and optimize restart trigger conditions, adjusting threshold parameters in the restart mechanism to adapt to the dynamic operating status of the equipment. The update and acquisition module 405 continuously collects high-frequency data updated by each module based on the adjusted restart mechanism, generates an operating status log after screening, cleaning, feature extraction, and status evaluation, and simultaneously determines the updated original dataset to ensure that data acquisition is synchronized with the equipment operating status, providing real-time data support for subsequent collaborative control, forming a complete closed loop from data acquisition to mechanism optimization to data update, ensuring the stability and efficiency of communication and collaborative control of the sugarcane leaf recycling equipment.

[0049] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for communication coordination control between cane leaf recycling equipment, characterized in that, The method comprises: Step S101, collecting environment parameter values and parameter deviation data from each module of the sugarcane leaf recycling equipment through a restart mechanism to generate an original data set; processing the environment parameter values in the original data set using a data compression algorithm to obtain a compressed data set; Step S102, if the size of the compressed data set is smaller than a preset bandwidth threshold, transmitting the compressed data set through a communication channel and judging the transmission completion state of the compressed data set; obtaining the compressed data set that has been transmitted, restoring the compressed data set using a decompression algorithm to obtain a restored data set; Step S103, for the restored data set, converting the parameter formats corresponding to different modules in the restored data set using a standardized interface to obtain a standardized data set; grouping the data with abnormal environment parameter values in the standardized data set using a clustering analysis algorithm to obtain a fault mode grouping; Step S104, fusing the parameter deviations in the restored data set through the fault mode grouping to determine the high-frequency fault point position of the sugarcane leaf recycling equipment, obtaining an optimized restart trigger condition based on the high-frequency fault point position, adjusting the threshold parameter in the restart mechanism according to the optimized restart trigger condition to obtain an adjusted restart mechanism; Step S105, collecting updated environment parameter values and updated parameter deviations from each module using the adjusted restart mechanism to obtain an updated original data set and generate a running state log.

2. The method of claim 1, wherein, In the step S101, the original data set is generated, comprising: Obtaining environment parameters and parameter deviations from each module through a restart mechanism to generate a high-frequency original data set; Judging whether the collection frequency of the high-frequency original data set is lower than a preset frequency threshold, if yes, adjusting the restart mechanism to increase the collection frequency to obtain a stable high-frequency original data set; According to the stable high-frequency original data set, using time series analysis to extract the change trend of the environment parameters and parameter deviations in the stable high-frequency original data set to determine a parameter fluctuation mode; According to the parameter fluctuation mode, using a K-means clustering algorithm to classify the running states of different modules of the sugarcane leaf recycling equipment to obtain a module state grouping; Judging whether the module state grouping shows an abnormality, if yes, extracting environment parameters and parameter deviation data from the abnormal module state grouping to determine an abnormal source; According to the abnormal source, using a support vector machine algorithm to predict potential fault points of the sugarcane leaf recycling equipment to generate a fault prediction result; According to the fault prediction result, adjusting the parameter configuration of the restart mechanism to optimize the data collection process to obtain an improved data set, which is the original data set.

3. The method of claim 2, wherein, In the step S101, the compressed data set is obtained, comprising: Processing the environment parameter values in the original data set according to a data compression algorithm to generate an initial compressed data set; If the compression rate of the initial compressed data set is lower than a preset compression rate threshold, adjusting the sampling interval of the data compression algorithm to optimize the data compression process to obtain an optimized compressed data set; According to the optimized compressed data set, the principal component analysis algorithm is used to extract the main features of the environmental parameter values, and a feature data set is obtained; If the feature dimension of the feature data set is higher than the preset dimension threshold, the feature data set is processed by dimension reduction to obtain a dimension-reduced data set; According to the dimension-reduced data set, the time series analysis is used to extract the time variation trend of the environmental parameter values, and the variation trend is obtained; Based on the variation trend, the K-means clustering algorithm is used to classify the running states of different time periods to obtain a state grouping; If the state grouping shows an anomaly, the corresponding feature data is extracted from the abnormal grouping and the source of the anomaly is judged to generate an anomaly analysis result, and the anomaly analysis result is combined with the dimension-reduced data set to obtain a compressed data set.

4. The method of claim 1, wherein, The step S102 comprises: If the storage capacity of the compressed data set is lower than the preset capacity threshold, the compressed data set is transmitted through a communication channel, the transmission completion state is judged, and a transmission result is obtained; According to the transmission result, the data packet technology is used to divide the compressed data set into a plurality of data packets, and a packet sequence is determined; If the transmission rate of the packet sequence is lower than the bandwidth limit of the communication channel, the size of the data packet is adjusted, an optimized packet sequence is regenerated, and an optimized packet is obtained; Through the optimized packet, the integrity of the compressed data set is verified by using a verification algorithm, whether there is data loss in the transmission process is judged, and a verification result is obtained; If the verification result shows that the compressed data set is transmitted completely, the optimized packet is combined through a transmission protocol, and the complete compressed data set is obtained; The environment parameter values are restored by using a decompression algorithm to decode the complete compressed data set; Whether the restored data matches the original data set is verified, and after eliminating the abnormal values, a restored data set is determined.

5. The method of claim 1, wherein, In the step S103, the standardized data set is obtained, comprising: The format of the restored data set is analyzed by using a standardized interface to obtain the environment parameter values and parameter format, and an initial conversion data set is generated; Based on the initial conversion data set, the data mapping technology is used to match the environment parameter values with the preset format specification to obtain a mapping data set; If the mapping data set meets the format specification, the data consistency of each module is integrated through the interface calling technology to generate an integrated data set; According to the integrated data set, the data integrity and consistency of the environment parameter values and the parameter deviation are judged by using the data verification technology to obtain a verification data set; Based on the verification data set, the clustering analysis algorithm is used to group the environment parameter values to determine a parameter grouping set; If the parameter grouping set meets the preset grouping threshold, the parameter mapping is combined by using the data aggregation technology to generate a standardized data set.

6. The method of claim 5, wherein, In the step S103, the fault mode grouping is obtained, comprising: The environment parameter values in the standardized data set are cleaned to generate a preprocessed data set; The preprocessed data set is converted by using a standardization technology; The environment parameter values in the converted preprocessed data set are grouped by using the K-means clustering algorithm to generate an initial parameter grouping. If the initial parameter group meets the abnormal threshold setting, an abnormal parameter in the preprocessed data set is identified by an abnormality detection technique to obtain an abnormal parameter group; According to the abnormal parameter group, a decision tree algorithm is used to classify the abnormal parameters to generate a fault mode classification; Based on the fault mode classification, a state monitoring technique is used to analyze the distribution characteristics of the abnormal parameters to obtain an abnormal distribution result; According to the abnormal distribution result, a data aggregation technique is used to merge the fault modes to obtain a fault mode group.

7. The method of claim 1, wherein, In step S104, an optimized restart trigger condition is obtained, including: The parameter deviation data is preprocessed by a data cleaning technique to remove noise data and obtain a cleaned deviation data set; A time series analysis technique is used to model the time series of the cleaned deviation data set to generate time series data; According to the time series data, a K-means clustering algorithm is used to extract features from the fault mode group to obtain point distribution characteristics; If the point distribution characteristics meet the preset distribution threshold, an abnormal detection technique is used to determine the high-frequency fault point to obtain the high-frequency fault point position; According to the high-frequency fault point position, a decision tree algorithm is used to classify the device state information to generate a state change trend; Through the state change trend, a data aggregation technique is used to fuse the fault occurrence frequency and trigger condition threshold to obtain the optimized restart trigger condition.

8. The method of claim 7, wherein, In step S104, an adjusted restart mechanism is obtained, including: Key parameters are extracted from the optimized restart trigger condition to generate an initial threshold set; Distribution characteristics are obtained from the initial threshold set, and a statistical analysis technique is used to construct a threshold distribution model; According to the threshold distribution model, the parameter adjustment range is determined to obtain an adjustment range set; If the adjustment range set meets the preset constraint condition, the threshold parameters in the initial threshold set are optimized by a linear regression algorithm to generate an optimized threshold set; According to the optimized threshold set, the logic of the restart mechanism is updated to generate a preliminary adjusted restart mechanism; The running log data is obtained from the preliminary adjusted restart mechanism, and a time series analysis technique is used to process the running log data to generate a running state trend; According to the running state trend, the stability of the preliminary adjusted restart mechanism is determined, and if the stability meets the requirements, it is determined as the adjusted restart mechanism.

9. The method of claim 1, wherein, The step S105 includes: Based on the adjusted restart mechanism, updated high-frequency data is collected from each module, and data screening techniques are used to screen the collected high-frequency data to remove invalid data and generate an updated original data set; If the updated original data set meets the preset integrity condition, a data cleaning technique is used to remove redundant information in the updated original data set to obtain a cleaned data set; According to the cleaned data set, a clustering analysis algorithm is used to classify the data according to categories to obtain a classified data set; Key features are extracted from the classified data set, and a feature selection technique is used to screen the extracted key features to generate a feature data set; If the feature dataset meets a preset correlation threshold, the feature dataset is analyzed by a time series analysis technique to generate a data change trend; According to the data change trend, a decision tree algorithm is used to judge the running state of each module to obtain a state evaluation result; Abnormal state data is extracted from the state evaluation result, and the abnormal state data is recorded to generate a running state log.

10. A communication coordination control system between cane leaf recycling equipment, characterized by, The cane leaf recycling equipment intercommunication collaborative control system for implementing the method of any one of claims 1-9 comprises: A collection compression module is configured to collect environment parameter values and parameter deviation high-frequency data from the modules to obtain an original dataset by a restart mechanism, and to process the high-frequency part of the environment parameter values by a data compression algorithm to obtain a compressed dataset according to the original dataset; A transmission restoration module is configured to transmit the compressed dataset with a size smaller than a preset bandwidth threshold through a communication channel and to judge a transmission completion state, and to restore the environment parameter values and the parameter deviation by a decompression algorithm to determine a restored dataset; A standardization grouping module is configured to convert the parameter formats of the environment parameter values of different modules to obtain a standardized dataset by a standardized interface according to the restored dataset, and to group the environment parameter values according to a clustering analysis algorithm to obtain a fault mode group; An optimization adjustment module is configured to judge the positions of high-frequency fault points by fusing the parameter deviations of the fault mode group to obtain an optimized restart trigger condition, and to adjust the threshold parameters in the restart mechanism to obtain an adjusted restart mechanism according to the optimized restart trigger condition; An update collection module is configured to collect updated high-frequency data from the modules to obtain an updated original dataset by the adjusted restart mechanism.

Images