A device whole cycle health diagnosis and collaborative operation big data management system

Abstract

This invention discloses a big data management system for full-cycle health diagnosis and collaborative operation of equipment, belonging to the field of ranch equipment management technology. The equipment monitoring terminal monitors ranch equipment and sends the obtained ranch perception data to the edge terminal. The edge terminal processes the ranch perception data and sends the processed ranch perception data to the user cloud. The user cloud aggregates the equipment characteristic information of each ranch device in real time and shares this information with the platform cloud in real time. It loads the equipment analysis model configured on the platform cloud, analyzes the ranch perception data based on the equipment analysis model, and obtains the health analysis results of each ranch device. Furthermore, it performs collaborative operation correlation analysis based on the ranch perception data and the health analysis results of each ranch device to obtain collaborative analysis results. Finally, it performs corresponding processing based on the health analysis results and the collaborative analysis results.

Description

Technical Field

[0001] This invention belongs to the field of ranch equipment management technology, specifically a big data management system for full-cycle health diagnosis and collaborative operation of equipment. Background Technology

[0002] As the core infrastructure of modern livestock production, farm equipment covers multiple categories such as milking, feeding, manure treatment, ventilation and temperature control. Its operation directly affects the quality of milk source, production efficiency and breeding costs. Currently, farm equipment management generally adopts the TnPM (Total Productive Maintenance) system, but the integration of existing equipment health management systems with farm TnPM systems is poor, making it difficult to adapt to the special working conditions of farms, such as dust, temperature fluctuations, and high humidity. Furthermore, it suffers from incomplete data collection, weak fault diagnosis, and disconnected collaborative management processes. For example, traditional farm equipment data collection focuses on single operating parameters, lacking comprehensive perception of multimodal data such as vibration, sound patterns, oil levels, and ambient temperature and humidity. Older farm equipment lacks native data interfaces, and the addition of sensors is susceptible to interference from the complex farm environment, resulting in low data accuracy. Fault diagnosis for farm equipment often uses general models, failing to consider farm-specific operating conditions such as milking machine turntable speed, feed cart load changes, and manure scraper operating resistance, leading to low diagnostic accuracy under varying conditions. Additionally, existing equipment health management systems lack in-depth analysis and coordinated control of the equipment-production activity relationship, focusing only on the equipment's own operating status, resulting in low production collaboration efficiency and significant resource waste.

[0003] In order to solve the above problems, this invention provides a big data management system for equipment full-cycle health diagnosis and collaborative operation. Summary of the Invention

[0004] To address the problems of the above solutions, this invention provides a big data management system for full-cycle health diagnosis and collaborative operation of equipment.

[0005] The objective of this invention can be achieved through the following technical solutions: A big data management system for full-cycle health diagnosis and collaborative operation of equipment, including platform cloud, user cloud, edge terminal and equipment monitoring terminal; Furthermore, the platform cloud communicates with each user cloud, the user cloud communicates with the edge terminal, and the edge terminal communicates with the device monitoring terminal.

[0006] The equipment monitoring terminal is used to monitor the ranch equipment in the ranch, obtain equipment monitoring data and equipment coordination data of each ranch equipment, summarize the equipment monitoring data and equipment coordination data of each ranch equipment into ranch perception data, and send the ranch perception data to the edge terminal.

[0007] The cloud-based platform includes a model repository and a configuration module. The model repository is used to store equipment analysis models for various ranch equipment at different equipment stages.

[0008] Furthermore, the equipment analysis model includes a fault diagnosis model, anomaly identification model, and operation and maintenance analysis model.

[0009] Furthermore, the establishment of the model repository includes: Step SA1: Obtain the various ranch equipment and the historical analysis data set of various ranch equipment. The historical analysis data set consists of historical unit data corresponding to multiple data sources; Classify the historical unit data in the historical analysis dataset to obtain the historical material data of the ranch equipment; Step SA2: Establish corresponding equipment analysis models based on various historical data, and conduct simulation verification of the corresponding equipment analysis models using historical data to obtain the equipment stages to which the equipment analysis models are applicable; Based on the equipment lifecycle and applicable equipment stages, identify unsuitable equipment stages and mark them as optimization and adjustment stages; Step SA3: Obtain historical material data corresponding to the optimization and adjustment phase, mark it as optimization material data, adjust the corresponding equipment analysis model based on the optimization material data, and obtain new equipment system models; determine the equipment phase applicable to the new equipment analysis model, and determine the new optimization and adjustment phase according to the equipment life cycle and the equipment phase applicable to each equipment analysis model; Step SA4: Repeat step SA2 until there is no more optimization and adjustment stage, to obtain the equipment analysis model of the ranch equipment at each equipment stage; Step SA5: Establish a model reserve based on the equipment analysis models of various ranch equipment at different equipment stages.

[0010] Furthermore, the historical data units within the historical analysis dataset are categorized, including: The equipment simulation analysis model of the ranch equipment is invoked, and the simulation analysis model is used to simulate and analyze the data of each historical unit in the historical analysis dataset to obtain the simulation calibration data of each historical unit. Based on the simulation calibration data, the historical unit data is classified to obtain the historical material data.

[0011] The configuration module is used to configure device analysis models for users in the cloud, identify device feature information of various ranch devices shared by users in the cloud in real time, and determine the ranch devices that need to be configured based on the device feature information. The platform matches the corresponding equipment analysis model from the model reserve based on the ranch equipment, optimizes and adjusts the equipment analysis model according to the equipment feature information, and configures the optimized and adjusted equipment analysis model in the user's cloud.

[0012] Furthermore, based on equipment characteristic information, the ranch equipment requiring model configuration is determined, including: Identify whether the ranch equipment corresponding to the equipment feature information is equipped with an equipment analysis model; When the ranch equipment does not have an equipment analysis model configured, determine that the corresponding ranch equipment needs to be configured with an equipment analysis model. When a ranch is equipped with an equipment analysis model, the model analysis features of the ranch equipment are identified based on the equipment feature information, and the ranch equipment is evaluated to determine whether model configuration is required based on the model analysis features.

[0013] Furthermore, the optimized and adjusted equipment analysis model is lightweighted to obtain a lightweight equipment analysis model, which is then deployed at the edge.

[0014] The edge device is used to receive ranch sensing data, process the ranch sensing data according to a preset data processing method, and send the processed ranch sensing data to the user's cloud.

[0015] The user cloud includes a device information module and a device analysis module; The equipment information module is used to summarize the equipment feature information of each ranch's equipment in real time and share the equipment feature information to the platform cloud in real time.

[0016] The equipment analysis module is used to perform health analysis on ranch equipment, load the equipment analysis model configured in the cloud of the platform, analyze the ranch perception data based on the equipment analysis model, and obtain the health analysis results of each ranch equipment; and perform collaborative operation correlation analysis based on the ranch perception data and the health analysis results of each ranch equipment to obtain collaborative analysis results; and perform corresponding processing based on the health analysis results and collaborative analysis results.

[0017] Furthermore, based on the pasture perception data and the health analysis results of various pasture equipment, a collaborative operation correlation analysis is conducted, including: The equipment production collaboration database is set up based on the historical production data of the ranch, and is dynamically updated according to the updates of the historical production data. The equipment production collaboration database is used to store the collaborative production activities of various ranch equipment, the production scope, the impact of different equipment status on collaborative production activities, and the deterioration impact of different production activity loads on ranch equipment. Based on the equipment production collaboration library, real-time feature extraction is performed on the ranch perception data to obtain the collaborative feature data of each ranch's equipment. Based on the health analysis results, production health impact data of ranch equipment are matched from the equipment production collaboration library; Production collaboration analysis is performed on ranch equipment based on collaboration feature data and production health impact data to obtain production collaboration analysis results, and corresponding production collaboration processing is carried out based on the production collaboration analysis results.

[0018] Compared with the prior art, the beneficial effects of the present invention are: A dedicated model library covering the entire lifecycle of equipment is pre-built and combined with accurately uploaded ranch equipment information from users, achieving a high degree of matching between the model and the actual equipment. This allows the equipment analysis model to completely break free from the limitations of general-purpose models. The model can accurately capture the operational characteristics and failure modes of different ranches and different equipment at specific lifecycle stages, effectively improving the targeting of fault diagnosis and the sensitivity of anomaly identification. Based on the adaptively adjusted dedicated model, it can provide early warnings of equipment degradation trends and potential micro-defects, achieving a fundamental shift from traditional reactive maintenance to predictive maintenance. This mechanism transforms operation and maintenance work from a passive response to proactive, regular management. It quantitatively analyzes the interaction between equipment status and production activities and establishes a dynamic linkage scheduling mechanism. The system can automatically optimize production scheduling strategies based on equipment health status, avoiding production quality decline or efficiency loss caused by operating equipment with defects. Conversely, it can also dynamically adapt equipment management strategies based on changes in production load, avoiding forced overload operation that causes accelerated equipment wear and tear. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a block diagram illustrating the principle of the present invention. Detailed Implementation

[0021] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0022] like Figure 1 As shown, a big data management system for full-cycle health diagnosis and collaborative operation of equipment includes a platform cloud, a user cloud, an edge terminal, and an equipment monitoring terminal; The platform cloud communicates with each user cloud, the user cloud communicates with the edge terminal, and the edge terminal communicates with the device monitoring terminal.

[0023] The platform cloud is set up by the platform provider, which can be a corporate headquarters or other operators; the user cloud is used by users, such as ranch managers.

[0024] The equipment monitoring terminal is used to monitor various farm equipment that need to be managed within the farm, and to obtain equipment monitoring data and equipment coordination data for each farm equipment. The equipment coordination data is data related to the production activities of the equipment, such as milking volume, milking planning, and other related data. The equipment monitoring data and equipment coordination data of each farm equipment are aggregated into farm perception data, and the farm perception data is sent to the edge terminal.

[0025] It integrates equipment status sensors such as vibration, sound pattern, current, pressure, running resistance, and load, as well as production activity data acquisition units such as milking volume, feeding volume, manure treatment volume, and ambient temperature and humidity / ammonia concentration, to achieve synchronous acquisition of equipment status data and production activity data. For complex outdoor working conditions in pastures, it adopts anti-interference variable frequency sparse coding sampling technology to ensure data acquisition accuracy. Through a heterogeneous data access gateway, it is compatible with old pasture equipment and production management systems, realizing unified access and standardized integration of multi-source data such as equipment, production, feeding, and energy consumption, laying a data foundation for collaborative analysis of equipment production activities.

[0026] The cloud-based platform includes a model repository and a configuration module. The model repository stores equipment analysis models for various ranch equipment at different equipment stages. Specifically, these models can be categorized according to analysis needs, such as fault diagnosis models, anomaly identification models, and operation and maintenance analysis models, for fault diagnosis, anomaly identification, and operation and maintenance analysis, respectively. The different equipment stages are determined based on the changes in the analysis accuracy of the equipment analysis model. If the accuracy meets the diagnostic analysis requirements for the entire life cycle, then the ranch equipment has only one equipment stage corresponding to the entire life cycle. However, due to factors such as equipment aging, the equipment analysis model generally needs to be optimized and adjusted after a certain stage. If users perform optimization and adjustment themselves, they need to have certain technical capabilities and incur high optimization costs; otherwise, the analysis accuracy will decrease, affecting the credibility of the user's experience.

[0027] In one embodiment, the equipment analysis model is set up by the platform based on existing technologies, such as common machine learning, deep learning algorithms, random forest algorithms, etc., but it is an intelligent model built to fit the ranch equipment.

[0028] Furthermore, by continuously providing services to various users, we can compile equipment analysis models for various ranch equipment and different equipment stages, making full use of the platform's accumulated resources.

[0029] In one embodiment, the establishment of a model repository includes: Step SA1: Obtain the various ranch equipment and the historical analysis data set of various ranch equipment. The historical analysis data set consists of historical unit data corresponding to multiple data sources. Multiple data sources refer to the historical analysis data generated by the same ranch equipment in different ranches. Because the differences in environment and other factors of different ranches will lead to the application of different equipment analysis models, the historical unit data includes the corresponding monitoring data and real standard fault diagnosis results, abnormal results, operation and maintenance results and other related data. Based on the historical analysis data set, various historical material data of ranch equipment are obtained. For the same type of ranch equipment, due to differences in ranch environment and other factors, there are different patterns in the historical analysis data. There are multiple historical material data, which are used to establish differentiated equipment analysis models applicable to the ranch equipment, while reducing the amount of analysis. Step SA2: Establish corresponding equipment analysis models based on historical data, and simulate and verify the corresponding equipment analysis models using historical data to obtain the applicable equipment stages for the equipment analysis model, that is, the life cycle corresponding to the required analysis accuracy, such as an accuracy rate greater than 99%, which is set by the platform. Based on the equipment life cycle and the applicable equipment stages, determine the inapplicable equipment stages and mark them as optimization and adjustment stages. This is determined based on all applicable equipment stages. If there are two equipment analysis models corresponding to two applicable equipment stages, the equipment life cycle outside of those two applicable equipment stages is considered. Step SA3: Obtain historical material data corresponding to the optimization and adjustment phase, mark it as optimization material data, adjust the corresponding equipment analysis model based on the optimization material data, and obtain a new equipment system model; determine the equipment phase to which the new equipment analysis model is applicable, and determine the new optimization and adjustment phase according to the equipment life cycle and the applicable equipment phase; Step SA4: Repeat step SA2 until there is no more optimization and adjustment stage, to obtain the equipment analysis model of the ranch equipment at each equipment stage; Step SA5: Establish a model reserve based on the equipment analysis models of various ranch equipment at different equipment stages.

[0030] In one embodiment, historical material data of ranch equipment is obtained based on historical analysis data sets. It can be classified in an existing manner, and historical unit data applicable to the same equipment analysis model can be grouped into one category and marked as historical material data. It can be classified based on clustering algorithms, etc. Alternatively, it can be classified manually to form training materials under different conditions and marked as historical material data.

[0031] In one embodiment, historical material data of the ranch equipment is obtained based on a historical analysis dataset, including: The equipment simulation analysis model of the ranch equipment is invoked, which can be the existing fault diagnosis model, anomaly identification model, operation and maintenance analysis model, etc. for the ranch equipment, or the general equipment analysis model available on the platform. The equipment simulation analysis model is used to simulate and analyze the data of each historical unit in the historical analysis data set to obtain the simulation calibration data of each historical unit data. That is, the simulation analysis results are compared with the corresponding standard results to determine the simulation calibration data. Based on the simulation calibration data, the historical unit data is classified. That is, based on the simulation calibration, it is determined whether the same equipment analysis model is applicable. Applicable data are classified into one category, and inapplicable data are classified into different categories, thus forming various data categories. The historical unit data within each category is summarized and marked as historical material data. If a single equipment simulation analysis model cannot completely classify all historical unit data, other equipment simulation analysis models can be used for simulation. That is, simulation analysis is performed through multiple equipment simulation analysis models to achieve the corresponding classification.

[0032] In one embodiment, the corresponding equipment analysis model can be adjusted based on optimized material data. This can be done by establishing a new equipment analysis model to increase the overall applicable equipment stages; or by adjusting the original equipment analysis model to adjust the applicable equipment stages.

[0033] The configuration module is used to configure the device analysis model for the user cloud, identify the device feature information of each ranch device shared by the user cloud in real time, and determine the ranch devices that need to be configured based on the device feature information. This is mainly for new ranch devices and situations where the device feature information is updated (causing the original device analysis model to need to be configured and adjusted). Based on the equipment feature information of the ranch equipment, the platform matches the corresponding ranch equipment at the corresponding equipment stage from the model reserve. The platform optimizes and adjusts the equipment analysis model based on the equipment feature information and configures the optimized and adjusted equipment analysis model in the user's cloud.

[0034] In one embodiment, the platform optimizes and adjusts the device analysis model based on device feature information. This can be done manually by the platform or intelligently based on existing intelligent technologies, such as optimization based on self-learning technology.

[0035] For example, the model analyzes the differences between the actual operating conditions of the equipment and the preset operating conditions of the model, and makes real-time fine-tuning of the model's threshold parameters and feature weights (e.g., if the actual load of the feeding cart is consistently 10% higher than the rated value, the load anomaly threshold of the hydraulic system fault diagnosis model is automatically adjusted to improve the sensitivity of fault identification); it calls up historical maintenance records, inspection data, and fault data uploaded by users, and combines them with big data analysis results of similar equipment to deeply optimize the model's algorithm parameters and feature extraction rules (e.g., if historical data of the milking machine shows "frequent failures of the turntable bearing after 3 years of operation", the wear coefficient of the remaining life prediction model is automatically adjusted to improve the accuracy of life prediction); it combines the TnPM maintenance records (AM / PM / BM records) of the ranch equipment to optimize the model's anomaly identification rules and early warning triggering conditions (e.g., if the user's equipment performs strict weekly lubrication maintenance, the model automatically reduces the false alarm rate of "insufficient lubrication" type anomalies); during the model adjustment process, the core algorithm framework is retained, and only parameters and thresholds are optimized to ensure model stability. Adjustment logs are automatically recorded, supporting backtracking and rollback.

[0036] In one embodiment, the ranch equipment that needs model configuration can be determined based on the equipment feature information. This can be done by the user, who can then add the configuration result to the equipment feature information, and the platform cloud can directly identify it. Alternatively, intelligent models can be built based on machine learning, deep learning algorithms, etc., to make intelligent judgments.

[0037] In one embodiment, determining the ranch equipment requiring model configuration based on equipment characteristic information includes: Identify whether the ranch equipment corresponding to the equipment feature information is equipped with an equipment analysis model; When the ranch equipment does not have an equipment analysis model configured, determine that the corresponding ranch equipment needs to be configured with an equipment analysis model. When a ranch is equipped with an equipment analysis model, the model analysis characteristics of the ranch equipment are identified based on the equipment feature information, such as the analysis accuracy, efficiency, and life cycle of the equipment analysis model. Based on the model analysis characteristics, it is assessed whether the ranch equipment needs to be configured with a model; for example, if the life cycle determines that different equipment analysis models need to be matched, or if the model analysis accuracy does not reach the preset value.

[0038] In one embodiment, for a device analysis model configured in the user's cloud, after optimization and adjustment, the platform performs lightweight processing on the device analysis model, including pruning, knowledge distillation, and other operations, to obtain a lightweight device analysis model, which is then configured in the edge device.

[0039] Generally, the configuration is first performed in the user's cloud, and then the user's cloud configures it in the edge device; if the user grants permissions, the configuration can also be performed directly from the platform cloud to the edge device.

[0040] In one embodiment, the lightweight device analysis model is configured according to user needs. For example, only the device analysis model that ensures the basic operation of the equipment is configured. That is, when the cloud and the edge terminal are connected, the basic operation of the ranch equipment can be ensured by the edge terminal. This can be determined in accordance with the conventional methods in the industry or by the user manually.

[0041] The edge device is used to receive ranch sensing data, process the ranch sensing data according to a preset data processing method, and send the processed ranch sensing data to the user's cloud.

[0042] In one embodiment, the ranch-sensing data is processed according to a preset data processing method, which is a conventional edge-end preprocessing of the ranch-sensing data, such as data cleaning, data standardization, data filtering and compression, feature extraction, data format conversion and other related data processing.

[0043] In one embodiment, the ranch-sensing data is processed according to a preset data processing method, including analyzing the data through a pre-set lightweight device analysis model when the user cloud cannot perform analysis (e.g., connection interruption), and taking corresponding actions based on the analysis results, such as stopping operation due to fault.

[0044] The user cloud includes a device information module and a device analysis module; The equipment information module is used to summarize the equipment feature information of various ranch equipment that need to be diagnosed and analyzed in real time, and to share the equipment feature information to the platform cloud in real time.

[0045] Equipment characteristic information mainly includes information related to matching equipment analysis models from the platform cloud and adapting and adjusting equipment analysis models. This includes basic attribute data (equipment core identifiers: name, model specifications, manufacturer, serial number, unique equipment code; key technical parameters: rated capacity / power / load / speed, core component list and specifications, structural type; installation information: installation location, installation date, installation method; technical standards: energy efficiency rating, protection rating, applicable operating conditions), lifecycle stage data (current lifecycle stage, years of operation, cumulative operating time, annual operating days), and operating condition data (actual operating parameters: average daily operating load, average daily operating hours). Number of load fluctuations; environmental parameters: operating environment temperature and humidity, dust concentration, ammonia concentration, whether it is deployed outdoors; operating condition characteristics: longest continuous operation time, start-up and shutdown frequency, production linkage requirements), historical management data (fault records: past fault types, fault locations, fault occurrence time, handling methods, replacement parts list; maintenance records: maintenance cycle, maintenance methods, lubrication / inspection / maintenance records; abnormal data: historical abnormal alarm records, micro-defect discovery status; operating data: historical energy consumption data, production capacity data); model analysis characteristics (accuracy, efficiency, equipment life cycle, etc., if an equipment analysis model is configured, used to analyze whether it needs optimization and adjustment), etc.

[0046] The equipment analysis module is used to perform health analysis on ranch equipment, load the equipment analysis model configured in the cloud of the platform, analyze the ranch perception data based on the equipment analysis model, and obtain the health analysis results of each ranch equipment; and perform collaborative operation correlation analysis based on the ranch perception data and the health analysis results of each ranch equipment to obtain collaborative analysis results. Based on the health analysis results and collaborative analysis results, appropriate processing will be carried out, such as data display and fault warning.

[0047] In one embodiment, a collaborative operation correlation analysis is performed based on pasture sensing data and health analysis results of various pasture equipment, including: The equipment production collaboration library is set up based on the historical production data of the ranch, and is dynamically updated according to the updates of the historical production data. The equipment production collaboration library is used to store relevant data such as the collaborative production activities of various ranch equipment, production scope, the impact of different equipment status on collaborative production activities, and the deterioration impact of different production activity loads on ranch equipment. The equipment production collaboration library is set up and updated according to the real-time generated historical production data of the ranch. Based on the equipment production collaboration library, real-time feature extraction is performed on the ranch perception data to obtain the collaboration feature data of each ranch equipment, such as the production load requirements of production activities within the corresponding production range; and based on the health analysis results, the production health impact data of the ranch equipment is matched from the equipment production collaboration library, that is, the impact data of the ranch equipment in this health state on production. Production collaboration analysis is performed on ranch equipment based on collaboration feature data and production health impact data to obtain production collaboration analysis results, and corresponding production collaboration processing is carried out based on the production collaboration analysis results.

[0048] For example, a collaborative library for equipment manufacturing: Based on historical production data of the ranch, the core supporting production activities of various ranch equipment are identified (such as milking machine → milking operation, feeding cart → precision feeding operation, manure scraper → manure cleaning operation, ventilation fan → barn environment control operation, biogas generator → ranch energy supply operation). Then, the production scope of each ranch equipment is determined according to the actual production area it is responsible for (milking operation in area A). The impact of farm equipment in different health states on the execution of production activities is quantified. For example, when the milking machine health score is ≥80, the milking efficiency is ≥95%; when the score is 50-80 (sub-healthy), the milking efficiency decreases by 10%-20% and the milk source qualification rate decreases by 5%; when the score is <50 (warning), the milking operation cannot be performed normally; when the hydraulic system of the feeding cart fails, the execution rate of precision feeding operations decreases by 30% and the timeliness of feeding the herd decreases by 40%; at the same time, a production impact threshold is set, and when the impact of equipment status on production activities exceeds the threshold, an early warning and production scheduling adjustment suggestion are automatically triggered. The impact of changes in production load and execution rhythm on equipment degradation rate and failure probability is quantified. For example, when the load on the feeding cart exceeds the rated value by 10%, the degradation rate of the hydraulic system increases by 25% and the failure probability increases by 30%. When the milking machine runs continuously for more than 8 hours during peak hours, the fatigue wear of the turntable bearing increases by 40% and the remaining life is shortened by 15%. At the same time, equipment wear thresholds are set. When adjustments to production activities cause equipment wear to exceed the threshold, equipment maintenance warnings and production load optimization suggestions are automatically triggered.

[0049] Data aggregation was conducted to establish a collaborative database for equipment production.

[0050] In one embodiment, production collaboration analysis is performed on ranch equipment based on collaborative feature data and production health impact data. A collaborative analysis model is established based on machine learning, deep learning algorithms, etc., and the collaborative feature data and production health impact data are analyzed through the collaborative analysis model to obtain production collaboration analysis results.

[0051] In one embodiment, production collaboration analysis is performed on ranch equipment based on collaboration feature data and production health impact data. Collaboration analysis is performed using various existing methods, such as commonly preset production collaboration analysis results corresponding to different collaboration feature data and production health impact data; alternatively, corresponding thresholds can be set directly, and when the health status or load exceeds the threshold, an early warning is issued, which is regarded as a collaboration anomaly.

[0052] For example, when the health status of equipment drops to the sub-health / warning level, or when equipment malfunctions, the system automatically adjusts the production activity plan based on the quantitative results of the equipment's impact on production activities. For instance, when a milking machine is in a sub-healthy state, the system automatically optimizes milking batches and reduces the amount of milk produced per batch to prevent a decline in milk quality. When a feeding cart malfunctions, the system automatically dispatches backup feeding equipment and adjusts the feeding route and time to ensure the continuity of precise feeding operations. At the same time, the system synchronizes the production activity adjustment information to the feeding and maintenance departments to achieve cross-departmental collaboration. When production activities require load adjustments (such as a surge in feed intake due to expanding the cattle herd or an increase in milking volume due to an increase in lactating cows), the system automatically optimizes equipment allocation and maintenance plans based on the quantified impact of production activities on equipment. For example, it may add well-maintained feeding carts to feed operations to avoid overloading individual machines; perform preventative maintenance on milking machines during peak milking periods, replacing easily worn parts to improve equipment load-bearing capacity; and provide equipment capacity constraint suggestions for production activities, clarifying the adjustable load range of production activities to avoid excessive equipment wear and tear. Incorporate production activity plans into the basis for equipment TnPM maintenance plans. For example, schedule preventative maintenance of milking machines during off-peak milking periods and spot checks and maintenance of feeding carts during feeding intervals. This achieves the synergistic goal of "maintenance not affecting production and production adapting to maintenance," reducing production downtime caused by maintenance.

[0053] Example 1: When the health score of one of the milking machines drops to 75 points (sub-healthy, triggering the production impact threshold), the system automatically analyzes its impact on milking operations (efficiency decrease of 15%) and pushes scheduling suggestions to the production department: optimize the original 6 milking batches into 8 smaller batches, reduce the milking volume per batch of this milking machine, and at the same time, group lactating cows, with healthy cows giving priority to using another healthy milking machine to ensure milk quality; the operation and maintenance department receives an early warning at the same time and performs preventive maintenance on the milking machine during the off-peak milking period.

[0054] Example 2: Due to the expansion of the cattle herd, the feeding volume for precision feeding operations needs to be increased by 20%. If only the original 6 feeding carts are used, the load on a single cart will exceed the rated value by 12% (triggering the equipment wear threshold). Optimization suggestions are then pushed: add 2 backup feeding carts (health score ≥90 points), distribute the feeding volume evenly among the 8 feeding carts, and conduct advance inspection and lubrication of the hydraulic systems of the original 6 feeding carts to improve their load-bearing capacity. Based on equipment capacity constraints, the production department adjusts the feeding volume increase to a gradual increase over 3 days to avoid sudden overload operation of the equipment.

[0055] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of a device lifecycle health diagnosis and collaborative operation big data management system as described in the above embodiments.

[0056] The above formulas are all numerical calculations after removing dimensions. The formulas are obtained by software simulation based on a large amount of data and are closest to the real situation. The preset parameters and preset thresholds in the formulas are set by those skilled in the art according to the actual situation or obtained by simulation based on a large amount of data.

[0057] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A big data management system for full-cycle health diagnosis and collaborative operation of equipment, characterized in that, This includes the platform cloud, user cloud, edge computing, and device monitoring. The device monitoring terminal is used to monitor the ranch equipment in the ranch, obtain ranch perception data, and send the ranch perception data to the edge terminal. The edge terminal is used to receive ranch sensing data, process the ranch sensing data according to a preset data processing method, and send the processed ranch sensing data to the user's cloud. The platform cloud includes a model repository and a configuration module; the user cloud includes a device information module and a device analysis module. The model repository is used to store equipment analysis models of various ranch equipment at different equipment stages; The configuration module is used to configure device analysis models for users in the cloud, identify device feature information of various ranch devices shared by users in the cloud in real time, and determine the ranch devices that need to be configured based on the device feature information. According to the ranch equipment, the corresponding equipment analysis model is matched from the model reserve. The platform optimizes and adjusts the equipment analysis model based on the equipment feature information and configures the optimized and adjusted equipment analysis model in the user's cloud. The equipment information module is used to summarize the equipment feature information of each ranch's equipment in real time and share the equipment feature information to the platform cloud in real time; The equipment analysis module is used to perform health analysis on ranch equipment, load the equipment analysis model configured in the cloud of the platform, analyze the ranch perception data based on the equipment analysis model, and obtain the health analysis results of each ranch equipment. Based on the pasture perception data and the health analysis results of each pasture's equipment, a collaborative operation correlation analysis is conducted to obtain the collaborative analysis results; Appropriate actions will be taken based on the results of the health analysis and collaborative analysis.

2. The big data management system for full-cycle health diagnosis and collaborative operation of equipment according to claim 1, characterized in that, The platform cloud communicates with each user cloud, the user cloud communicates with the edge terminal, and the edge terminal communicates with the device monitoring terminal.

3. The big data management system for full-cycle health diagnosis and collaborative operation of equipment according to claim 1, characterized in that, The equipment analysis model includes a fault diagnosis model, anomaly identification model, and operation and maintenance analysis model.

4. The big data management system for full-cycle health diagnosis and collaborative operation of equipment according to claim 1, characterized in that, The establishment of the model repository includes: Step SA1: Obtain the various ranch equipment and the historical analysis data set of various ranch equipment. The historical analysis data set consists of historical unit data corresponding to multiple data sources; Classify the historical unit data in the historical analysis dataset to obtain the historical material data of the ranch equipment; Step SA2: Establish corresponding equipment analysis models based on various historical data, and conduct simulation verification of the corresponding equipment analysis models using historical data to obtain the equipment stages to which the equipment analysis models are applicable; Based on the equipment lifecycle and applicable equipment stages, identify unsuitable equipment stages and mark them as optimization and adjustment stages; Step SA3: Obtain historical material data corresponding to the optimization and adjustment phase, mark it as optimization material data, adjust the corresponding equipment analysis model based on the optimization material data, and obtain new equipment system models; determine the equipment phase applicable to the new equipment analysis model, and determine the new optimization and adjustment phase according to the equipment life cycle and the equipment phase applicable to each equipment analysis model; Step SA4: Repeat step SA2 until there is no more optimization and adjustment stage, to obtain the equipment analysis model of the ranch equipment at each equipment stage; Step SA5: Establish a model reserve based on the equipment analysis models of various ranch equipment at different equipment stages.

5. The big data management system for full-cycle health diagnosis and collaborative operation of equipment according to claim 4, characterized in that, The historical data in the historical analysis dataset is categorized into various historical units, including: The equipment simulation analysis model of the ranch equipment is invoked, and the simulation analysis model is used to simulate and analyze the data of each historical unit in the historical analysis dataset to obtain the simulation calibration data of each historical unit. Based on the simulation calibration data, the historical unit data is classified to obtain the historical material data.

6. The big data management system for full-cycle health diagnosis and collaborative operation of equipment according to claim 1, characterized in that, Based on equipment characteristic information, identify the ranch equipment that requires model configuration, including: Identify whether the ranch equipment corresponding to the equipment feature information is equipped with an equipment analysis model; When the ranch equipment does not have an equipment analysis model configured, determine that the corresponding ranch equipment needs to be configured with an equipment analysis model. When a ranch is equipped with an equipment analysis model, the model analysis features of the ranch equipment are identified based on the equipment feature information, and the ranch equipment is evaluated to determine whether model configuration is required based on the model analysis features.

7. The big data management system for full-cycle health diagnosis and collaborative operation of equipment according to claim 1, characterized in that, In the configuration module, the optimized and adjusted device analysis model is lightweighted to obtain a lightweight device analysis model, which is then deployed at the edge.

8. The big data management system for full-cycle health diagnosis and collaborative operation of equipment according to claim 7, characterized in that, Pre-defined data processing methods at the edge include: When the edge device loses connection with the user's cloud, it is analyzed using a pre-deployed lightweight device analysis model, and appropriate actions are taken based on the analysis results.

9. The big data management system for full-cycle health diagnosis and collaborative operation of equipment according to claim 1, characterized in that, Based on the farm's sensor data and the health analysis results of various farm equipment, a collaborative operation correlation analysis is conducted, including: The equipment production collaboration database is set up based on the historical production data of the ranch, and is dynamically updated according to the updates of the historical production data. The equipment production collaboration database is used to store the collaborative production activities of various ranch equipment, the production scope, the impact of different equipment status on collaborative production activities, and the deterioration impact of different production activity loads on ranch equipment. Based on the equipment production collaboration library, real-time feature extraction is performed on the ranch perception data to obtain the collaborative feature data of each ranch's equipment. Based on the health analysis results, production health impact data of ranch equipment are matched from the equipment production collaboration library; Production collaboration analysis is performed on ranch equipment based on collaboration feature data and production health impact data to obtain production collaboration analysis results, and corresponding production collaboration processing is carried out based on the production collaboration analysis results.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a big data management system for full-cycle health diagnosis and collaborative operation of a device as described in any one of claims 1 to 9.

Images