Intelligent monitoring and operation and maintenance management system for agricultural mechanical equipment
By employing methods such as environmental compensation, latent fault monitoring, fault knowledge graphs, and remote diagnostics, the problems of low data acquisition adaptability and low operation and maintenance efficiency in agricultural machinery monitoring systems have been solved, achieving intelligent fault diagnosis and collaborative operation and maintenance management.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- DA NONG TECH CO LTD
- Filing Date
- 2026-04-15
- Publication Date
- 2026-05-15
AI Technical Summary
Existing agricultural machinery monitoring systems have deficiencies in data acquisition adaptability, intelligent fault diagnosis, and collaborative operation and maintenance management. Sensor data is prone to deviation, leading to false alarms or missed alarms. System alarms require manual investigation, resulting in low operation and maintenance management efficiency and a lack of initiative and collaboration.
The system employs a data acquisition module for environmental compensation, a latent fault module to monitor equipment sound and vibration data, an anomaly identification module to generate core parameter change curves, a cause reasoning module to construct a fault knowledge graph, a work order dispatch module to screen maintenance personnel, a proactive prevention module to generate preventative inspection work orders, and a remote diagnosis module to perform remote diagnosis using a lightweight digital twin model.
It improves the adaptability of data collection, the intelligence of fault diagnosis, and the initiative and collaboration of operation and maintenance management, reduces false alarms and missed alarms, improves operation and maintenance efficiency, and realizes intelligent fault diagnosis and preventive maintenance.
Smart Images
Figure CN122048332A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the technical field of equipment management, and in particular to an intelligent monitoring and operation and maintenance management system for agricultural machinery and equipment. Background Technology
[0002] Agricultural machinery is the core of modern agricultural production, and its operational status directly affects operational efficiency and the stability of agricultural production. With the development of intelligent technology, higher requirements are being placed on the monitoring and maintenance of agricultural machinery and equipment.
[0003] In existing technologies, common monitoring systems primarily collect equipment operating parameters through various sensors and then issue alarms based on preset fixed values. However, agricultural machinery typically operates in harsh environments, and the data collected by sensors is prone to inconsistencies, which can lead to sensor malfunctions in severe cases. Relying on a single threshold for alarms is susceptible to false alarms or missed alarms. Secondly, system alarms usually only indicate the surface situation, requiring further manual investigation for specific problems, resulting in low efficiency. At the operation and maintenance management level, the creation, allocation, and processing of work orders still heavily rely on manual labor, leading to extremely low efficiency and difficulty in comprehensively considering the actual situation of maintenance personnel, as well as a lack of proactive coordination with farmland operation plans and equipment maintenance cycles. Therefore, existing agricultural machinery monitoring and operation and maintenance systems have significant deficiencies in terms of data acquisition adaptability, intelligent fault diagnosis, and proactive and collaborative operation and maintenance management, which urgently need to be addressed. Summary of the Invention
[0004] The purpose of this invention is to provide an intelligent monitoring and operation and maintenance management system for agricultural machinery and equipment to solve the problems mentioned in the background art.
[0005] This application provides an intelligent monitoring and maintenance management system for agricultural machinery and equipment, the system comprising: Data acquisition module: used to collect data on agricultural machinery and equipment and their working environment, obtain equipment operating parameters and working environment parameters, and perform data compensation on the equipment operating parameters based on the working environment parameters to obtain target equipment parameters; Latent Fault Module: Used to extract equipment sound data and equipment vibration data from the equipment operating parameters, monitor the equipment sound data and equipment vibration data to obtain equipment performance change parameters, and obtain latent equipment anomalies based on the equipment performance change parameters; Anomaly identification module: used to obtain the equipment load and working time based on the target equipment parameters, monitor the target equipment parameters, generate core parameter change curves, identify anomalies on the core parameter change curves, and combine the anomalies, the equipment load, and the working time to obtain the abnormal parameters of the agricultural machinery equipment. Cause reasoning module: used to acquire historical maintenance data, employee experience data and equipment manuals, construct a fault knowledge graph, and perform anomaly reasoning in the fault knowledge graph based on the hidden equipment anomalies and the equipment anomaly parameters to obtain troubleshooting and maintenance suggestions; Work order dispatch module: used to obtain the fault location, current location of maintenance personnel, current task load and historical completion quality, generate employee information package for each maintenance personnel, and generate work orders by combining the employee information package and the troubleshooting and maintenance suggestions and dispatching them to the selected target maintenance personnel. Proactive prevention module: Used to obtain the operation plan of agricultural machinery and equipment. Before the agricultural machinery and equipment is operated, the historical maintenance records of the equipment are checked, and preventive inspection work orders are generated and dispatched in advance based on the historical maintenance records. Remote diagnostic module: used to extract the hidden equipment anomalies and the fault complexity and mechanical parameters of the equipment anomalies, establish a lightweight digital twin model of the fault location based on the fault complexity and the mechanical parameters, perform remote diagnosis and debugging based on the lightweight digital twin model, record maintenance instructions, and distribute them to local maintenance personnel.
[0006] Preferably, the data acquisition module includes: Acquisition unit and compensation unit; Data Acquisition Unit: Used to install sensors on agricultural machinery and equipment to collect equipment and environmental information, and to collect data on agricultural machinery and equipment and their working environment through the sensors to obtain equipment operating parameters and working environment parameters; Compensation unit: used to extract real-time ambient temperature and humidity, ambient vibration parameters, and air quality parameters based on the working environment parameters; Based on the real-time ambient temperature and humidity, the environmental vibration parameters, and the air quality parameters, the data offset of the target sensor for data acquisition from agricultural machinery equipment is evaluated to obtain the data offset amount. Based on the data offset, environmental compensation parameters are generated, and the target sensor is calibrated according to the environmental compensation parameters, and the device operating parameters are compensated.
[0007] Preferably, the latent fault module includes: Data monitoring unit and latent anomaly assessment unit; Data monitoring unit: used to extract device sound data and device vibration data from the device operating parameters; Based on the device sound data, the voiceprint information in the device sound data is extracted, and the voiceprint information is filtered to obtain the internal voiceprint information of the device. From the device sound data, extract the internal sound information corresponding to the internal voiceprint information, and generate an internal sound curve based on the internal sound information; Based on the equipment vibration data, the equipment vibration data is classified to obtain vibration source data and coordinated vibration data; Based on the vibration source data and the coordinated vibration data, the source vibration curve and the coordinated vibration curve are generated respectively. Latent anomaly assessment unit: used to combine the internal sound data, the source vibration curve, and the coordinated vibration curve to construct sound-vibration coordinated assessment parameters; The sound-vibration co-evaluation parameters are compared with preset standard evaluation parameters to obtain a comparison difference value. Based on the comparison difference value, the equipment performance is evaluated to obtain equipment performance change parameters. Based on the equipment performance change parameters, latent equipment anomalies are obtained.
[0008] Preferably, the anomaly detection module includes: Multi-source fusion unit and anomaly detection unit; Multi-source fusion unit: used to identify the target equipment parameters, extract the equipment load and working time of agricultural machinery equipment, and construct the equipment background parameters before the anomaly judgment based on the equipment load and the working time; Monitor the parameters of the target device for book search, generate an initial core change curve, substitute the device background parameters into the initial core change curve, and adaptively adjust the initial core change curve to obtain the core parameter change curve. Anomaly detection unit: used to traverse the core parameter change curve and identify the initial anomaly points on the core parameter change curve; The initial abnormal points are filtered based on the equipment background parameters to obtain abnormal points. The abnormal values of the abnormal points are extracted to generate equipment abnormal parameters for agricultural machinery.
[0009] Preferably, the cause reasoning module includes: Knowledge graph unit and reasoning unit; Knowledge graph unit: used to acquire historical maintenance data, employee experience data, and equipment manuals; Based on the equipment manual, the equipment structure data of the agricultural machinery equipment is obtained; based on the employee experience data, the fault handling preferences and fault occurrence preferences are obtained; based on the historical maintenance data, the causes of historical faults and historical fault handling solutions are obtained. By combining the equipment construction data, fault handling preferences, fault occurrence preferences, historical fault causes, and historical fault handling schemes, knowledge point information is extracted to obtain fault knowledge point data, and a fault knowledge graph is constructed based on the fault knowledge point data. Inference unit: used to generate a graph query query based on the hidden device anomaly and the device anomaly parameters; Based on the query information in the graph, deep reasoning is performed in the fault knowledge graph, and the reasoning logic value obtained in each reasoning is recorded. Determine whether the inference logic value exceeds a preset standard logic value; If it is determined that the reasoning logic value exceeds the standard logic value, then proceed to the next step of reasoning until troubleshooting and maintenance suggestions are generated; If the reasoning logic value is determined to be no more than the standard logic value, then the reasoning is stopped and the next knowledge point is queried in the knowledge graph for further reasoning.
[0010] Preferably, the work order dispatch module includes: Employee identification unit and distribution unit; Employee identification unit: used to obtain the location of the fault, the current location of the maintenance personnel, the current workload, and the historical completion quality; Based on the location of the fault and the current location, the relative distance between each maintenance personnel and the location of the fault is obtained, and the fastest arrival time is obtained based on the relative distance; By combining the fastest arrival time, the current task load, and the historical completion quality, an employee information package is generated for each employee. Dispatch unit: used to determine the difficulty and urgency of handling the fault based on the troubleshooting and maintenance recommendations. Based on the difficulty and urgency of the fault handling, multiple employee information packages are screened to obtain target maintenance personnel. Based on the troubleshooting and repair recommendations, a work order is generated and sent to the target repair personnel.
[0011] Preferably, the active prevention module includes: Pre-work inspection unit and preventative work order dispatch unit; Pre-work inspection unit: used to obtain the operation plan of agricultural machinery and equipment, and to determine the operation time and location of the agricultural machinery and equipment based on the operation plan; Within a preset time period before the operation time, check the historical maintenance records of the agricultural machinery and equipment, and based on the historical maintenance records, determine whether the equipment is close to its maintenance cycle and / or whether the equipment has experienced a malfunction in the area of the operation location; If it is determined that the equipment is close to its maintenance cycle and / or the equipment has experienced a failure in the area of the work site, then extract the equipment's maintenance item data and / or historical target failure information; Preventive work order dispatch unit: used to generate equipment maintenance work orders based on the maintenance item data, and to obtain the historical fault location and historical fault cause based on the historical target fault information; Based on the historical fault location and cause, a historical fault inspection work order is generated, and the equipment maintenance work order and / or the historical fault inspection work order are dispatched to maintenance personnel in advance.
[0012] Preferably, the remote diagnostic module includes: Digital modeling unit and remote processing unit; Digital model unit: used to extract the fault complexity and mechanical parameters of the hidden equipment anomaly and the equipment anomaly parameters; Based on the mechanical parameters of the equipment, a data twin model is created to construct a lightweight model framework for the abnormal equipment area; Based on the hidden device anomaly, the device anomaly parameters, and the fault complexity, the lightweight model framework is used to reproduce the fault, resulting in a lightweight digital twin model. Remote processing unit: used to send the lightweight digital twin model to a remote fault handling expert, and use the lightweight digital twin model to simulate faults and demonstrate fault phenomena; Fault handling experts perform online remote diagnosis and debugging based on the lightweight digital twin model, and record the diagnosis and debugging process. The data from the processing procedure is organized to generate maintenance instructions, which are then sent to local maintenance personnel.
[0013] In summary, this application includes at least one of the following beneficial technical effects: By collecting data on the equipment status and working environment during the operation of agricultural machinery, environmental parameter compensation is applied to the collected equipment operating parameters based on the environmental data to obtain more accurate target equipment parameters. Equipment sound and vibration data are extracted from the operating parameters and monitored to obtain equipment performance change parameters, further identifying hidden equipment anomalies. Based on the target equipment parameters, the equipment load and working duration are obtained and monitored to generate core parameter change curves. Anomalies are identified on these core parameter change curves using equipment load and working duration as background parameters, and the values of these anomalies are extracted to obtain the abnormal equipment parameters. Furthermore, historical maintenance data, employee experience, and equipment manuals are used to construct a fault knowledge graph. The abnormal equipment parameters are searched within this fault knowledge graph to obtain troubleshooting and maintenance suggestions. Then, work orders are generated based on these suggestions. The optimal target maintenance personnel are selected based on their current location, current workload, and historical performance, and the work order is then assigned to them. Before agricultural machinery is put into operation, its historical maintenance records are checked. Based on these records, it is determined whether preventative maintenance is required. If so, a preventative inspection work order is generated and dispatched to maintenance personnel. When a complex fault occurs that cannot be handled on-site, a lightweight digital twin model is generated and sent to a remote fault expert for remote diagnosis and troubleshooting. Repair instructions are then recorded and sent to local maintenance personnel. This improves the adaptability of data collection, the intelligence of fault diagnosis, and the proactiveness and collaboration of operation and maintenance management. Attached Figure Description
[0014] Figure 1 This is a block diagram of the intelligent monitoring and operation and maintenance management system for agricultural machinery and equipment provided in the embodiments of this application.
[0015] Explanation of reference numerals in the attached diagram: 1. Data acquisition module; 2. Latent fault module; 3. Anomaly identification module; 4. Cause reasoning module; 5. Work order dispatch module; 6. Proactive prevention module; 7. Remote diagnosis module. Detailed Implementation
[0016] The following is in conjunction with the appendix Figure 1 This application will be described in further detail, but the embodiments of the present invention are not limited thereto.
[0017] This application discloses an intelligent monitoring and operation and maintenance management system for agricultural machinery and equipment.
[0018] In this embodiment, an intelligent monitoring and operation and maintenance management system for agricultural machinery and equipment includes: Data acquisition module 1: Used to collect data on agricultural machinery and equipment and their working environment, obtain equipment operating parameters and working environment parameters, and perform data compensation on the equipment operating parameters based on the working environment parameters to obtain the target equipment parameters; Latent Fault Module 2: Used to extract equipment sound data and equipment vibration data from equipment operating parameters, monitor the equipment sound data and equipment vibration data to obtain equipment performance change parameters, and obtain latent equipment anomalies based on the equipment performance change parameters; Anomaly identification module 3: It is used to obtain the equipment load and working time based on the target equipment parameters, monitor the target equipment parameters, generate the core parameter change curve, identify the abnormal points on the core parameter change curve, and combine the abnormal points, equipment load and working time to obtain the abnormal parameters of the agricultural machinery equipment. Cause Reasoning Module 4: Used to acquire historical maintenance data, employee experience data and equipment manuals, construct a fault knowledge graph, and perform anomaly reasoning in the fault knowledge graph based on hidden equipment anomalies and equipment anomaly parameters to obtain troubleshooting and maintenance suggestions; Work order dispatch module 5: Used to obtain the location of the fault, the current location of the maintenance personnel, the current task load and historical completion quality, generate an employee information package for each maintenance personnel, and generate a work order by combining the employee information package and troubleshooting and maintenance suggestions and dispatching it to the selected target maintenance personnel. Proactive Prevention Module 6: Used to obtain the operation plan of agricultural machinery and equipment. Before the agricultural machinery and equipment is operated, the historical maintenance records of the equipment are checked, and preventive inspection work orders are generated and dispatched in advance based on the historical maintenance records. Remote diagnostic module 7: Used to extract the fault complexity and mechanical parameters of hidden equipment anomalies and equipment anomaly parameters, establish a lightweight digital twin model of the fault location based on the fault complexity and mechanical parameters, perform remote diagnosis and debugging based on the lightweight digital twin model, record maintenance instructions, and distribute them to local maintenance personnel.
[0019] The data acquisition module includes: Acquisition unit and compensation unit; Data Acquisition Unit: Used to install sensors on agricultural machinery and equipment to collect equipment and environmental information, and to collect data on agricultural machinery and equipment and their working environment through the sensors to obtain equipment operating parameters and working environment parameters; Compensation unit: used to extract real-time ambient temperature and humidity, ambient vibration parameters, and air quality parameters based on working environment parameters; Based on real-time environmental temperature and humidity, environmental vibration parameters, and air quality parameters, the target sensors for data acquisition of agricultural machinery and equipment are evaluated for data offset to obtain the data offset amount. Based on the data offset, environmental compensation parameters are generated, and the target sensor is calibrated according to the environmental compensation parameters, and the equipment operating parameters are compensated.
[0020] In practice, a John Deere S760 corn harvester operating in the Northeast Plain during autumn harvest is used as an example. The system incorporates vibration, temperature, and speed sensors on key components such as the harvester's engine, drive shaft, and header, as well as ambient temperature, humidity, and vibration sensors on the exterior of the machine. The data acquisition module collects real-time operating parameters from these sensors, including engine speed of 2100 rpm and hydraulic system oil temperature of 85 degrees Celsius, as well as environmental parameters such as ambient temperature of 35 degrees Celsius, relative humidity of 80%, and vibration amplitude of 0.5G due to ground bumps. The compensation unit then analyzes the collected environmental parameters, finding that the high temperature and humidity cause temperature sensor readings to be approximately 3 degrees Celsius higher than standard laboratory conditions, while continuous bumps cause a zero-point drift of approximately 0.1G in the vibration sensor. Based on this analysis, the compensation unit calculates the data offset and generates corresponding environmental compensation parameters: a temperature compensation coefficient of -3 and a vibration compensation coefficient of -0.1. Subsequently, the system used these compensation parameters to calibrate the original equipment operating parameters, correcting the engine oil temperature reading from 85 degrees to 82 degrees and the drive shaft vibration value from 2.8G to 2.7G, thereby obtaining more accurate target equipment parameters that eliminated environmental interference, laying the foundation for subsequent precise analysis.
[0021] The hidden fault module includes: Data monitoring unit and latent anomaly assessment unit; Data monitoring unit: used to extract equipment sound data and equipment vibration data from equipment operating parameters; Based on the device's sound data, the voiceprint information is extracted from the device's sound data, and the voiceprint information is filtered to obtain the internal voiceprint information of the device. In the device's sound data, extract the internal sound information corresponding to the internal voiceprint information, and generate an internal sound curve based on the internal sound information; Based on equipment vibration data, the equipment vibration data is classified to obtain vibration source data and coordinated vibration data; Based on vibration source data and coordinated vibration data, source vibration curves and coordinated vibration curves are generated respectively. Latent anomaly assessment unit: used to combine internal sound data, source vibration curves, and coordinated vibration curves to construct sound-vibration coordinated assessment parameters; The sound-vibration co-evaluation parameters are compared with the preset standard evaluation parameters to obtain the comparison difference value. Based on the comparison difference value, the equipment performance is evaluated to obtain the equipment performance change parameters. Based on the equipment performance change parameters, the hidden equipment anomalies are obtained.
[0022] In practice, a John Deere S760 corn harvester operating in the Northeast Plain during autumn harvest is used as an example. The data monitoring unit of the latent fault module extracts the sound data spectrum of the engine during continuous operation and the vibration acceleration data of multiple points on the frame from the target equipment parameters. Through in-depth analysis of the sound data, the unit successfully isolates external noises such as wind noise and crop friction noise, extracts specific acoustic signature information representing in-cylinder combustion and gear meshing within the engine, and generates an "internal sound curve" reflecting the internal state of the engine. Simultaneously, the vibration data is categorized, separating "vibration source data" directly from the engine body and "coordinated vibration data" excited by the engine and transmitted through the frame, and plotting "source vibration curves" and "coordinated vibration curves" respectively. The latent anomaly assessment unit performs a coordinated analysis of these three curves to construct a comprehensive "sound-vibration coordinated assessment parameter." The system compares this parameter with the standard evaluation parameter library recorded when the equipment was brand new. It finds that the energy in the frequency band representing gear meshing has increased by 15%, and there is asynchronous fluctuation with a certain high-frequency component in the vibration curve, with a difference value reaching 0.25 (the threshold is usually 0.15). Based on this difference, the system assesses that the equipment performance has undergone potential changes, thus determining a latent equipment anomaly of early wear in the engine gearbox. At this time, the equipment's surface operating parameters may still be within the normal range, but the potential problem has been detected in advance.
[0023] The anomaly detection module includes: Multi-source fusion unit and anomaly detection unit; Multi-source fusion unit: used to identify target equipment parameters, extract equipment load and working time of agricultural machinery, and construct equipment background parameters before anomaly judgment based on equipment load and working time; Monitor the parameters of the target device for book search, generate an initial core change curve, substitute the device background parameters into the initial core change curve, and adaptively adjust the initial core change curve to obtain the core parameter change curve. Anomaly detection unit: used to traverse the core parameter change curves and identify initial anomalies on the core parameter change curves; The initial anomalies are filtered based on the equipment background parameters to obtain the anomalies. The anomaly values of the anomalies are then extracted to generate the equipment anomaly parameters for the agricultural machinery.
[0024] In practice, a John Deere S760 corn harvester operating in the Northeast Plain during autumn harvest is used as an example. The multi-source fusion unit of the anomaly identification module first identifies from the compensated target equipment parameters that the equipment is currently in a "high-load threshing operation" state and has been working continuously for 7.5 hours, thus constructing equipment background parameters that include "high load" and "long working hours". Subsequently, the system monitors real-time data of engine oil pressure and generates an initial core parameter change curve. The unit substitutes the equipment background parameters and, considering that the oil pressure can fluctuate more under high load, the system adaptively adjusts the initial curve, for example, by widening the threshold range of the anomaly alarm by 5%, thus obtaining a "core parameter change curve" that better reflects the current working conditions. The anomaly judgment unit scans this adjusted curve and finds several "initial anomaly points" on the curve where the pressure drops sharply at instantaneous moments. The system then filters out fluctuations within the normal range caused by brief load changes, based on the context of "high load and long working hours." It ultimately identifies an "anomaly" where the pressure remained below the adjusted lower threshold for 30 seconds during a stable load phase. The system extracts the pressure value of this anomaly as 180 kPa (normal range 200-400 kPa). Combined with its duration, it generates the equipment anomaly parameter "persistently low oil pressure," indicating a potential malfunction such as decreased oil pump efficiency or oil circuit leakage.
[0025] The causal reasoning module includes: Knowledge graph unit and reasoning unit; Knowledge graph unit: used to acquire historical maintenance data, employee experience data, and equipment manuals; Based on the equipment manual, we obtain the equipment structure data of the agricultural machinery; based on the employee experience data, we obtain the fault handling preferences and fault occurrence preferences; based on the historical maintenance data, we obtain the historical fault causes and historical fault handling solutions. By combining equipment structure data, fault handling preferences, fault occurrence preferences, historical fault causes and historical fault handling solutions, knowledge point information is extracted to obtain fault knowledge point data, and a fault knowledge graph is constructed based on the fault knowledge point data. Inference unit: Used to generate graph query query information based on hidden equipment anomalies and equipment anomaly parameters; Based on the query information from the graph, deep reasoning is performed in the fault knowledge graph, and the reasoning logic value obtained in each reasoning is recorded. Determine whether the inference logic value exceeds the preset standard logic value; If the reasoning logic value is determined to exceed the standard logic value, proceed to the next step of reasoning until troubleshooting and repair suggestions are generated; If the reasoning logic value is determined to be less than the standard logic value, the reasoning will stop and the next knowledge point will be queried in the knowledge graph for further reasoning.
[0026] In practice, a John Deere S760 corn harvester operating in the Northeast Plain during autumn harvest is used as an example. For the anomalies of "hidden gearbox wear" and "persistently low oil pressure" observed in this John Deere S760 corn harvester, the cause reasoning module is activated. The knowledge graph unit retrieves the equipment manual for this model of harvester, obtaining detailed structural data of its transmission and lubrication systems; it retrieves historical maintenance data from the maintenance database, discovering that the equipment experienced low oil pressure due to filter clogging during the same period last year; and it extracts the fault occurrence preference from experienced maintenance personnel's experience records, finding that "high gearbox temperature is a common cause for this model after more than 8 hours of continuous operation." The reasoning unit integrates the current hidden equipment anomalies and equipment anomaly parameters to generate a graph query question: "Abnormal gearbox wear accompanied by low oil pressure." The system begins reasoning within the constructed fault knowledge graph: first, it associates "gearbox wear" and "low oil pressure," and the knowledge graph provides an intermediate node for "poor lubrication," with a high reasoning logic value (0.8). Starting with "poor lubrication," the system then queries "filter clogging (historical)" and "prolonged high-temperature operation (experience preference)." The calculated path logic value for "filter clogging causing oil circuit obstruction" is 0.9 (exceeding the standard value of 0.85), while the path logic value for "high temperature only causing oil deterioration" is 0.7. The system adopts the higher logic value path and continues to deduce the troubleshooting and repair suggestion of "prioritizing the inspection and replacement of the oil filter, while also checking the gearbox lubricating oil level and quality," and stops further deduction of the lower logic value path.
[0027] The work order dispatch module includes: Employee identification unit and distribution unit; Employee identification unit: used to obtain the location of the fault, the current location of the maintenance personnel, the current workload, and the historical completion quality; Based on the location of the fault and the current location, the relative distance between each maintenance personnel and the location of the fault is obtained, and the fastest arrival time is obtained based on the relative distance; By combining the fastest arrival time, current task load, and historical completion quality, an employee information package is generated for each employee; Dispatch unit: Used to determine the difficulty and urgency of handling the fault based on the troubleshooting and maintenance suggestions; Based on the difficulty and urgency of troubleshooting, multiple employee information packages are screened to obtain target maintenance personnel. Based on the troubleshooting and repair recommendations, a work order is generated and sent to the target repair personnel.
[0028] In practice, a John Deere S760 corn harvester operating in the Northeast Plain during autumn harvest is used as an example. After the system generates a troubleshooting and repair suggestion for this malfunctioning John Deere S760 corn harvester—"Check the filter and gearbox lubrication"—the work order dispatch module begins operation. The employee identification unit immediately takes action, identifying the location of the malfunction as a cornfield at "43.5 degrees North latitude, 125.3 degrees East longitude," and simultaneously acquiring the real-time GPS locations of three on-duty maintenance personnel. The unit calculates the straight-line distance between each maintenance personnel and the malfunction point, and, combined with current road conditions, estimates the fastest arrival time: Maintenance Personnel A (45 minutes), Maintenance Personnel B (60 minutes), and Maintenance Personnel C (90 minutes). The system also retrieves their current number of pending work orders (task load) and their maintenance success rate over the past year (historical completion quality). Combining these three pieces of information, an employee information package is generated for each repairman. For example, Repairman A's information package states, "Appointment time as early as 45 minutes, currently has one low-priority task, historical repair success rate is 95%." Based on the troubleshooting and repair suggestions, the dispatch unit determines the difficulty of handling this fault to be "medium" (requires component replacement) and the urgency to be "high" (affects critical operations). Subsequently, the system comprehensively filters the fault handling difficulty and urgency against the employee information packages of the three repairmen. Although Repairman C has the highest historical quality, his arrival time is too long; Repairman B has a moderate arrival time and task load, but a slightly lower historical success rate. Ultimately, the system selects Repairman A, who has the shortest arrival time, lightest task load, and highest historical quality, as the target repairman and automatically sends a work order containing fault details, location navigation, and troubleshooting suggestions to his smart terminal.
[0029] The proactive prevention module includes: Pre-work inspection unit and preventative work order dispatch unit; Pre-work inspection unit: used to obtain the work plan of agricultural machinery and equipment, and based on the work plan, to determine the work time and location of the agricultural machinery and equipment; Before the scheduled operation time, check the historical maintenance records of agricultural machinery and equipment. Based on the historical maintenance records, determine whether the equipment is close to its maintenance cycle and / or whether the equipment has experienced a malfunction in the area of the operation site. If it is determined that the equipment is close to its maintenance cycle and / or the equipment has experienced a failure in the area of the work site, then extract the equipment's maintenance item data and / or historical target failure information. Preventive work order dispatch unit: used to generate equipment maintenance work orders based on maintenance item data, and to obtain the location and cause of historical faults based on historical target fault information; Based on the location and cause of historical faults, generate historical fault inspection work orders and dispatch equipment maintenance work orders and / or historical fault inspection work orders to maintenance personnel in advance.
[0030] In practice, let's take a John Deere S760 corn harvester operating in the Northeast Plain during autumn harvest as an example. The night before this John Deere S760 corn harvester was scheduled to work on "High Standard Farmland No. 5" the following day, the proactive prevention module automatically activated. The pre-work inspection unit retrieved the harvester's work plan, confirming the work time and location. Subsequently, the system checked the equipment's historical maintenance records and found two key pieces of information: First, the equipment's last comprehensive maintenance was 280 hours ago, while the recommended maintenance cycle is 300 hours, indicating it was "approaching its maintenance cycle"; second, historical records showed that the equipment had experienced "minor damage to the header blades" last year while operating on "High Standard Farmland No. 5," due to hard rocks along the field ridges. Based on the first point, "approaching its maintenance cycle," the preventative work order dispatch unit generated an "Equipment Maintenance Work Order" including items such as changing the engine oil, oil filter, and air filter. Based on the second point, "historical fault information," a "Historical Fault Inspection Work Order" specifically targeting "inspecting the wear of the header blades and fasteners" was generated. The system merges these two preventative work orders and dispatches them to the maintenance team responsible for the area 12 hours before the start of the operation, reminding them to complete these inspections and maintenance before the equipment departs or during work breaks, thereby effectively preventing downtime due to untimely maintenance or recurrence of old problems.
[0031] The remote diagnostic module includes: Digital modeling unit and remote processing unit; Digital model unit: used to extract the fault complexity and mechanical parameters of hidden equipment anomalies and equipment anomaly parameters; Based on the mechanical parameters of the equipment, a data twin model is created to construct a lightweight model framework for the abnormal equipment area; Based on the hidden equipment anomalies, equipment anomaly parameters, and fault complexity, the lightweight model framework is used to reproduce the fault, resulting in a lightweight digital twin model. Remote processing unit: Used to send lightweight digital twin models to remote fault handling experts and use the lightweight digital twin models to simulate faults and demonstrate fault phenomena; Troubleshooting experts conduct online remote diagnosis and debugging based on a lightweight digital twin model, and record the diagnosis and debugging process. The data from the processing is organized, maintenance instructions are generated, and the maintenance instructions are sent to local maintenance personnel.
[0032] In practice, let's take a John Deere S760 corn harvester operating in the Northeast Plain during autumn harvest as an example. For this John Deere S760 corn harvester, if the fault manifests as a complex "intermittent hydraulic system pressure loss accompanied by abnormal noise," and local maintenance personnel are unable to resolve the issue after initial troubleshooting, the remote diagnostic module is activated. The digital model unit extracts the "hidden equipment anomalies" and "equipment anomaly parameters" of the fault, assessing its complexity as "high." Simultaneously, the unit retrieves precise three-dimensional mechanical parameters of the harvester's hydraulic system from the equipment database, including the hydraulic pump model, piping layout, and control valve group positions. Based on these mechanical parameters, the system quickly constructs a lightweight digital twin model framework in the cloud, containing only the relevant components of the hydraulic system. Subsequently, the fault characteristics of "intermittent pressure loss" and "abnormal noise at specific frequencies" are reproduced within this model framework, generating a lightweight digital twin model that simulates the fault phenomenon. The remote processing unit then sends this model to hydraulic system experts at headquarters. Experts can operate the model remotely, observing the pressure and flow dynamics of the virtual hydraulic system from multiple angles, listening to simulated abnormal noises, and performing virtual debugging operations, such as simulating the adjustment of the relief valve and checking the volumetric efficiency of the virtual pump. Through online diagnosis, the expert determined that the root cause of the fault might be internal wear of the main hydraulic pump. He recorded this diagnostic process, verification steps, and maintenance instructions, including "recommending a focused inspection of the main pump and measuring its flow and pressure curves," creating a detailed electronic guide that was then distributed in real-time to local maintenance personnel in the field to guide them in precise repairs.
[0033] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.
Claims
1. An intelligent monitoring and operation and maintenance management system for agricultural machinery and equipment, characterized in that, include: Data acquisition module: used to collect data on agricultural machinery and equipment and their working environment, obtain equipment operating parameters and working environment parameters, and perform data compensation on the equipment operating parameters based on the working environment parameters to obtain target equipment parameters; Latent Fault Module: Used to extract equipment sound data and equipment vibration data from the equipment operating parameters, monitor the equipment sound data and equipment vibration data to obtain equipment performance change parameters, and obtain latent equipment anomalies based on the equipment performance change parameters; Anomaly identification module: used to obtain the equipment load and working time based on the target equipment parameters, monitor the target equipment parameters, generate core parameter change curves, identify anomalies on the core parameter change curves, and combine the anomalies, the equipment load, and the working time to obtain the abnormal parameters of the agricultural machinery equipment. Cause reasoning module: used to acquire historical maintenance data, employee experience data and equipment manuals, construct a fault knowledge graph, and perform anomaly reasoning in the fault knowledge graph based on the hidden equipment anomalies and the equipment anomaly parameters to obtain troubleshooting and maintenance suggestions; Work order dispatch module: used to obtain the fault location, current location of maintenance personnel, current task load and historical completion quality, generate employee information package for each maintenance personnel, and generate work orders by combining the employee information package and the troubleshooting and maintenance suggestions and dispatching them to the selected target maintenance personnel. Proactive prevention module: Used to obtain the operation plan of agricultural machinery and equipment. Before the agricultural machinery and equipment is operated, the historical maintenance records of the equipment are checked, and preventive inspection work orders are generated and dispatched in advance based on the historical maintenance records. Remote diagnostic module: used to extract the hidden equipment anomalies and the fault complexity and mechanical parameters of the equipment anomalies, establish a lightweight digital twin model of the fault location based on the fault complexity and the mechanical parameters, perform remote diagnosis and debugging based on the lightweight digital twin model, record maintenance instructions, and distribute them to local maintenance personnel.
2. The intelligent monitoring and operation and maintenance management system for agricultural machinery and equipment according to claim 1, characterized in that, The data acquisition module includes: Acquisition unit and compensation unit; Data Acquisition Unit: Used to install sensors on agricultural machinery and equipment to collect equipment and environmental information, and to collect data on agricultural machinery and equipment and their working environment through the sensors to obtain equipment operating parameters and working environment parameters; Compensation unit: used to extract real-time ambient temperature and humidity, ambient vibration parameters, and air quality parameters based on the working environment parameters; Based on the real-time ambient temperature and humidity, the environmental vibration parameters, and the air quality parameters, the data offset of the target sensor for data acquisition from agricultural machinery equipment is evaluated to obtain the data offset amount. Based on the data offset, environmental compensation parameters are generated, and the target sensor is calibrated according to the environmental compensation parameters, and the device operating parameters are compensated.
3. The intelligent monitoring and operation and maintenance management system for agricultural machinery and equipment according to claim 2, characterized in that, The latent fault module includes: Data monitoring unit and latent anomaly assessment unit; Data monitoring unit: used to extract device sound data and device vibration data from the device operating parameters; Based on the device sound data, the voiceprint information in the device sound data is extracted, and the voiceprint information is filtered to obtain the internal voiceprint information of the device. From the device sound data, extract the internal sound information corresponding to the internal voiceprint information, and generate an internal sound curve based on the internal sound information; Based on the equipment vibration data, the equipment vibration data is classified to obtain vibration source data and coordinated vibration data; Based on the vibration source data and the coordinated vibration data, the source vibration curve and the coordinated vibration curve are generated respectively. Latent anomaly assessment unit: used to combine the internal sound data, the source vibration curve, and the coordinated vibration curve to construct sound-vibration coordinated assessment parameters; The sound-vibration co-evaluation parameters are compared with preset standard evaluation parameters to obtain a comparison difference value. Based on the comparison difference value, the equipment performance is evaluated to obtain equipment performance change parameters. Based on the equipment performance change parameters, latent equipment anomalies are obtained.
4. The intelligent monitoring and operation and maintenance management system for agricultural machinery and equipment according to claim 3, characterized in that, The anomaly detection module includes: Multi-source fusion unit and anomaly detection unit; Multi-source fusion unit: used to identify the target equipment parameters, extract the equipment load and working time of agricultural machinery equipment, and construct the equipment background parameters before the anomaly judgment based on the equipment load and the working time; Monitor the parameters of the target device for book search, generate an initial core change curve, substitute the device background parameters into the initial core change curve, and adaptively adjust the initial core change curve to obtain the core parameter change curve. Anomaly detection unit: used to traverse the core parameter change curve and identify the initial anomaly points on the core parameter change curve; The initial abnormal points are filtered based on the equipment background parameters to obtain abnormal points. The abnormal values of the abnormal points are extracted to generate equipment abnormal parameters for agricultural machinery.
5. The intelligent monitoring and operation and maintenance management system for agricultural machinery and equipment according to claim 4, characterized in that, The causal reasoning module includes: Knowledge graph unit and reasoning unit; Knowledge graph unit: used to acquire historical maintenance data, employee experience data, and equipment manuals; Based on the equipment manual, the equipment structure data of the agricultural machinery equipment is obtained; based on the employee experience data, the fault handling preferences and fault occurrence preferences are obtained; based on the historical maintenance data, the causes of historical faults and historical fault handling solutions are obtained. By combining the equipment construction data, fault handling preferences, fault occurrence preferences, historical fault causes, and historical fault handling schemes, knowledge point information is extracted to obtain fault knowledge point data, and a fault knowledge graph is constructed based on the fault knowledge point data. Inference unit: used to generate a graph query query based on the hidden device anomaly and the device anomaly parameters; Based on the query information in the graph, deep reasoning is performed in the fault knowledge graph, and the reasoning logic value obtained in each reasoning is recorded. Determine whether the inference logic value exceeds a preset standard logic value; If it is determined that the reasoning logic value exceeds the standard logic value, then proceed to the next step of reasoning until troubleshooting and maintenance suggestions are generated; If the reasoning logic value is determined to be no more than the standard logic value, then the reasoning is stopped and the next knowledge point is queried in the knowledge graph for further reasoning.
6. The intelligent monitoring and operation and maintenance management system for agricultural machinery and equipment according to claim 5, characterized in that, The work order dispatch module includes: Employee identification unit and distribution unit; Employee identification unit: used to obtain the location of the fault, the current location of the maintenance personnel, the current workload, and the historical completion quality; Based on the location of the fault and the current location, the relative distance between each maintenance personnel and the location of the fault is obtained, and the fastest arrival time is obtained based on the relative distance; By combining the fastest arrival time, the current task load, and the historical completion quality, an employee information package is generated for each employee. Dispatch unit: used to determine the difficulty and urgency of handling the fault based on the troubleshooting and maintenance recommendations. Based on the difficulty and urgency of the fault handling, multiple employee information packages are screened to obtain target maintenance personnel. Based on the troubleshooting and repair recommendations, a work order is generated and sent to the target repair personnel.
7. The intelligent monitoring and operation and maintenance management system for agricultural machinery and equipment according to claim 6, characterized in that, The active prevention module includes: Pre-work inspection unit and preventative work order dispatch unit; Pre-work inspection unit: used to obtain the operation plan of agricultural machinery and equipment, and to determine the operation time and location of the agricultural machinery and equipment based on the operation plan; Within a preset time period before the operation time, check the historical maintenance records of the agricultural machinery and equipment, and based on the historical maintenance records, determine whether the equipment is close to its maintenance cycle and / or whether the equipment has experienced a malfunction in the area of the operation location; If it is determined that the equipment is close to its maintenance cycle and / or the equipment has experienced a failure in the area of the work site, then extract the equipment's maintenance item data and / or historical target failure information; Preventive work order dispatch unit: used to generate equipment maintenance work orders based on the maintenance item data, and to obtain the historical fault location and historical fault cause based on the historical target fault information; Based on the historical fault location and cause, a historical fault inspection work order is generated, and the equipment maintenance work order and / or the historical fault inspection work order are dispatched to maintenance personnel in advance.
8. The intelligent monitoring and operation and maintenance management system for agricultural machinery and equipment according to claim 7, characterized in that, The remote diagnostic module includes: Digital modeling unit and remote processing unit; Digital model unit: used to extract the fault complexity and mechanical parameters of the hidden equipment anomaly and the equipment anomaly parameters; Based on the mechanical parameters of the equipment, a data twin model is created to construct a lightweight model framework for the abnormal equipment area; Based on the hidden device anomaly, the device anomaly parameters, and the fault complexity, the lightweight model framework is used to reproduce the fault, resulting in a lightweight digital twin model. Remote processing unit: used to send the lightweight digital twin model to a remote fault handling expert, and use the lightweight digital twin model to simulate faults and demonstrate fault phenomena; The troubleshooting experts performed online remote diagnosis and debugging based on the lightweight digital twin model, and recorded the diagnosis and debugging process. The data from the processing procedure is organized to generate maintenance instructions, which are then sent to local maintenance personnel.