A method and system for detecting electrical faults in park facilities

Abstract

The present invention provides a method and system for detecting electrical faults in park facilities, which relates to the field of electrical fault detection. The method includes: predicting electrical faults based on multi-dimensional health data; determining multiple early warning conveyor belts based on a number of predicted fault probability analyses, obtaining multiple predicted maintenance durations based on a number of high-frequency fault type sets, and generating multiple initial maintenance plans; optimizing multiple initial maintenance plans with the goal of maximizing package throughput and minimizing maintenance cycles to determine the optimal maintenance plan; and performing operation and maintenance according to the optimal maintenance plan in a preset time zone. This method aims to solve the technical problem that traditional methods lack maintenance plans based on risk prediction and resource scheduling optimization, resulting in the difficulty of operation and maintenance strategies in balancing system security and package transportation efficiency, and easily causing business interruptions and reduced throughput capacity. It can maximize package throughput and improve the overall operating efficiency of the logistics park conveyor line system while ensuring system security.

Description

Technical Field

[0001] The present invention relates to the field of electrical fault detection, and in particular to a method and system for detecting electrical faults in park facilities. Background Art

[0002] With the rapid development of the logistics industry, automated conveyor systems within logistics parks have become core infrastructure supporting modern logistics operations. The efficient operation of these systems directly impacts the throughput and operational efficiency of the entire logistics park. However, the electrical equipment widely used in conveyor systems often faces the risk of electrical failures during long-term operation, including component aging, overloads, power outages, and signal failures. These failures can not only cause equipment downtime but also impact overall logistics operations, leading to business interruptions or reduced throughput.

[0003] Currently, traditional electrical fault detection methods typically rely on regular manual inspections or passive responses based on fault alarms, which are subject to lags and the risk of missed detection, and are unable to identify potential faults in advance. In addition, these traditional methods lack intelligent prediction and risk assessment of the operating conditions of different conveyor line equipment, and fail to effectively combine fault prediction information with the scheduling of maintenance resources. Therefore, it is impossible to scientifically plan operation and maintenance strategies, resulting in a lack of flexibility in operation and maintenance plans, making it difficult to balance equipment safety and operational efficiency, and prone to problems such as excessive downtime and the inability to repair equipment in a timely manner. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for detecting electrical faults in park facilities. This method addresses the technical issues that traditional electrical fault detection methods for logistics park conveyor line systems lack maintenance plans based on risk prediction and resource scheduling optimization, resulting in an inability to strike a balance between system security and package delivery efficiency in operation and maintenance strategies, and easily leading to business interruptions and reduced throughput. The system includes the following:

[0005] In a first aspect, the present invention provides a method for detecting electrical faults in park facilities, comprising: monitoring and acquiring a plurality of multidimensional health data of a plurality of conveyor belts in a conveyor line system in a logistics park, performing electrical fault predictions respectively based on the plurality of multidimensional health data, and outputting a plurality of predicted fault probabilities and a plurality of high-frequency fault type sets; determining a plurality of early warning conveyor belts based on the plurality of predicted fault probabilities, and calculating and acquiring a plurality of predicted maintenance durations based on the plurality of high-frequency fault type sets; generating a plurality of initial maintenance plans within a preset time zone based on the plurality of early warning conveyor belts and the plurality of predicted maintenance durations; optimizing the plurality of initial maintenance plans to determine an optimal maintenance plan with a constraint of being less than a conveyor belt failure probability threshold value and with the goal of maximizing package throughput and minimizing maintenance cycle; and performing operation and maintenance on the plurality of early warning conveyor belts in accordance with the optimal maintenance plan in the preset time zone.

[0006] Preferably, the method for detecting electrical faults in park facilities also includes: monitoring and obtaining a number of multidimensional health data of a number of conveyor belts in a conveyor line system in a logistics park within a preset historical time window, wherein the multidimensional health data includes at least electrical parameters, control signals and operating parameters; pre-training an electrical fault predictor; using the electrical fault predictor to perform electrical fault predictions based on the number of multidimensional health data, and outputting a number of predicted fault probabilities and a number of high-frequency fault type sets.

[0007] Preferably, the method for detecting electrical faults in campus facilities further includes: collecting a sample multidimensional health data set based on the historical operation and maintenance records of the conveyor line system, and counting the proportion of electrical faults and high-frequency fault types of different sample multidimensional health data in the historical time zone, setting them as sample fault probabilities and sample high-frequency fault types, obtaining a sample fault probability set and a sample high-frequency fault type set, wherein the high-frequency fault type includes at least two fault types; using the sample multidimensional health data set as input and the sample fault probability set and the sample high-frequency fault type set as supervision, training the BP neural network until the model converges to obtain the electrical fault predictor.

[0008] Preferably, the method for detecting electrical faults in campus facilities further includes: configuring a warning probability threshold, setting a conveyor belt with a predicted fault probability greater than the warning probability threshold as a warning conveyor belt, determining multiple warning conveyor belts based on the several predicted fault probabilities, and matching and obtaining multiple high-frequency fault type sets of the multiple warning conveyor belts; analyzing and obtaining the average historical fault maintenance time of high-frequency fault types within a preset historical time range based on the historical operation and maintenance records of the conveyor line system; setting weights based on the proportion of fault types in high-frequency fault types, and calculating the maintenance time of the multiple high-frequency fault type sets based on the average historical fault maintenance time, and outputting multiple predicted maintenance times, wherein the weights are positively correlated with the proportion of fault types.

[0009] Preferably, the method for detecting electrical faults in campus facilities further includes: configuring the maximum number of collaborative maintenance of conveyor belts; taking the maximum number of collaborative maintenance as a constraint, within a preset time zone, enumerating the maintenance order of the warning conveyor belts according to the multiple warning conveyor belts and multiple predicted maintenance durations, and outputting multiple initial maintenance plans, wherein in the initial maintenance plan, the working time interval between adjacent warning conveyor belts is greater than or equal to 0.

[0010] Preferably, the method for detecting electrical faults in park facilities further comprises: randomly selecting a first maintenance plan from the multiple initial maintenance plans, respectively calculating the time intervals between the maintenance time nodes of multiple early warning conveyor belts in the first maintenance plan and the current time nodes, and obtaining multiple maintenance time intervals; performing a correlation analysis on the transport speed of the conveyor belt and the growth trend of the failure probability based on the historical operation and maintenance records of the conveyor line system, and constructing a speed-trend coefficient comparison table, wherein the transport speed and the growth trend coefficient of the failure probability are positively correlated; based on the speed-trend coefficient comparison table, with the constraint of being less than the conveyor belt failure probability threshold value, Based on multiple predicted failure probabilities and the multiple maintenance time intervals, multiple first maximum transport speeds of multiple early warning conveyor belts are calculated; a first maintenance cycle of the first maintenance plan is obtained, wherein the maintenance cycle is the time interval between the current time node and the time node when all the early warning conveyor belts in the maintenance plan complete the maintenance; with the goal of maximizing the package throughput and minimizing the maintenance cycle, the plan fitness is calculated according to the multiple first maximum transport speeds and the first maintenance cycle, and the first plan fitness is output; the multiple plan fitnesses of the multiple initial maintenance plans are continued to be analyzed and obtained, and the initial maintenance plan with the maximum plan fitness is selected as the optimal maintenance plan.

[0011] Preferably, the method for detecting electrical faults in campus facilities further includes: performing a parcel transport simulation based on the multiple first maximum transport speeds and the first maintenance plan, and outputting a first parcel throughput; constructing a scheme fitness evaluation function with the goal of maximizing the parcel throughput and minimizing the maintenance cycle, wherein the scheme fitness evaluation function includes an initial weight ratio, and the scheme fitness is positively correlated with the parcel throughput and negatively correlated with the maintenance cycle; optimizing the initial weight ratio based on the first parcel throughput to determine a first optimized weight ratio, updating the scheme fitness evaluation function to determine an updated scheme fitness evaluation function; and using the updated scheme fitness evaluation function to calculate the first scheme fitness based on the first parcel throughput and the first maintenance cycle.

[0012] Preferably, the method for detecting electrical faults in campus facilities further includes: configuring an initial weight ratio, wherein the initial weight ratio includes a preset throughput weight and a preset maintenance cycle weight, and the sum of the two is 1; obtaining the throughput of packages to be transported in a preset time zone; if the throughput of packages to be transported is less than or equal to the first package throughput, setting the initial weight ratio to a first optimized weight ratio; if the throughput of packages to be transported is greater than the first package throughput, calculating the throughput deviation ratio between the throughput of packages to be transported and the first package throughput; multiplying the throughput deviation ratio by a constant, and adding the sum to 1 to obtain a first adjustment coefficient, wherein the constant is greater than 1 and less than 5; multiplying the first adjustment coefficient by the preset throughput weight, and setting the product of the two as the optimized throughput weight; subtracting the optimized throughput weight from 1 to obtain an optimized maintenance cycle weight, and setting the optimized throughput weight and the optimized maintenance cycle weight as the first optimized weight ratio.

[0013] In a second aspect, the present invention further provides a park facility electrical fault detection system for executing a park facility electrical fault detection method as described in the first aspect, comprising: an electrical fault prediction module for monitoring and obtaining a plurality of multidimensional health data of a plurality of conveyor belts in a conveyor line system within a logistics park, performing electrical fault predictions based on the plurality of multidimensional health data, and outputting a plurality of predicted fault probabilities and a plurality of high-frequency fault type sets; a predicted maintenance duration calculation module for determining a plurality of early warning conveyor belts based on the plurality of predicted fault probabilities, and calculating and obtaining a plurality of predicted maintenance durations based on the plurality of high-frequency fault type sets; an initial maintenance plan generation module for generating a plurality of initial maintenance plans within a preset time zone based on the plurality of early warning conveyor belts and the plurality of predicted maintenance durations; an initial maintenance plan optimization module for optimizing the plurality of initial maintenance plans with a constraint of being less than a conveyor belt failure probability threshold value and with the goal of maximizing package throughput and minimizing maintenance cycle to determine an optimal maintenance plan; an operation and maintenance control module for performing operation and maintenance on the plurality of early warning conveyor belts in the preset time zone according to the optimal maintenance plan.

[0014] The embodiments of the present invention include the following advantages:

[0015] By monitoring and acquiring a plurality of multi-dimensional health data of a plurality of conveyor belts in a conveyor line system within a logistics park, electrical fault prediction is performed based on the plurality of multi-dimensional health data, and a plurality of predicted fault probabilities and a plurality of high-frequency fault type sets are outputted. Then, based on the plurality of predicted fault probabilities, a plurality of early warning conveyor belts are determined, and a plurality of predicted maintenance durations are calculated based on the plurality of high-frequency fault type sets. Furthermore, based on the plurality of early warning conveyor belts and the plurality of predicted maintenance durations, a plurality of initial maintenance plans are generated within a preset time zone. Then, with a constraint of being less than a conveyor belt failure probability threshold, and with the goal of maximizing package throughput and minimizing maintenance cycles, the plurality of initial maintenance plans are optimized to determine the optimal maintenance plan. Finally, within the preset time zone, the plurality of early warning conveyor belts are operated and maintained according to the optimal maintenance plan. In other words, by combining electrical fault prediction with resource scheduling optimization, dynamic operation and maintenance scheduling based on fault risk can be achieved. This can maximize package throughput while ensuring system safety, and reduce downtime through precise maintenance planning, thereby improving the overall operating efficiency of the conveyor line system in the logistics park and significantly improving system throughput and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a flow chart of the steps of a method for detecting electrical faults in park facilities according to the present invention;

[0017] Figure 2 This is a structural diagram of a park facility electrical fault detection system according to the present invention.

[0018] Description of reference numerals:

[0019] Electrical fault prediction module 11, predicted maintenance time calculation module 12, initial maintenance plan generation module 13, initial maintenance plan optimization module 14, operation and maintenance control module 15. DETAILED DESCRIPTION

[0020] The present invention provides a method and system for detecting electrical faults in park facilities, addressing the technical issues of traditional logistics park conveyor line systems lacking a maintenance plan based on risk prediction and resource scheduling optimization, resulting in an operation and maintenance strategy that is difficult to balance system security with package transportation efficiency, and easily causing business interruptions and reduced throughput. By combining electrical fault prediction with resource scheduling optimization, dynamic operation and maintenance scheduling based on fault risks can be achieved. This can maximize package throughput while ensuring system security, and reduce downtime through precise maintenance plans, thereby improving the overall operating efficiency of the logistics park conveyor line system and significantly enhancing the system's throughput and reliability.

[0021] Below, the technical solutions of the present invention will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments described herein. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. It should also be noted that, for the convenience of description, only the parts related to the present invention, rather than all, are shown in the accompanying drawings.

[0022] For example, see the attached Figure 1 The present invention provides a method for detecting electrical faults in park facilities, which is applied to a system for detecting electrical faults in park facilities and specifically comprises the following steps:

[0023] S10: Monitor and obtain a plurality of multi-dimensional health data of a plurality of conveyor belts of a conveyor line system in a logistics park, perform electrical fault predictions based on the plurality of multi-dimensional health data, and output a plurality of predicted fault probabilities and a plurality of high-frequency fault type sets.

[0024] Furthermore, step S10 of the present invention further includes:

[0025] S11: Within a preset historical time window, monitor and obtain a plurality of multi-dimensional health data of a plurality of conveyor belts of a conveyor line system in a logistics park, wherein the multi-dimensional health data at least includes electrical parameters, control signals and operating parameters.

[0026] Specifically, a preset historical time window is configured, and the preset historical time window can be adjusted according to actual operational needs to ensure the timeliness and accuracy of fault prediction. For example, the preset historical time window is set to 12 hours, that is, the health data of multiple conveyor belts in the last 12 hours are monitored and recorded. Then, within the preset historical time window, a number of multi-dimensional health data of several conveyor belts in the conveyor line system in the logistics park are obtained through multi-source sensor monitoring, that is, different types of sensors are used to monitor and collect relevant data, such as current sensors, voltage sensors, temperature sensors, vibration sensors, load sensors, etc.; multi-source sensors provide data sources of different dimensions, including electrical signals (voltage, current, etc.), mechanical signals (vibration, load, etc.) and control signals (conveyor belt start, stop, speed regulation signals, etc.), to ensure comprehensive coverage of various health data; among them, multi-dimensional health data at least includes electrical parameters, control signals and operation parameters. Electrical parameters include, but are not limited to, key electrical indicators such as current, voltage, power, frequency, and temperature. By monitoring these electrical parameters, electrical component anomalies such as overload, short circuit, and aging of electrical equipment can be detected. Control signals involve control commands such as the start, stop, and speed adjustment of the conveyor belt. These signals are crucial for determining whether the control system is operating normally. If the control signal is abnormal, it may cause equipment failure or abnormal stop. Operating parameters include the conveyor belt's operating speed, load, operating time, fault records, etc. Operating parameters can help identify problems such as mechanical wear and excessive load, thereby conducting risk assessments on the system. Through the continuous monitoring of multi-dimensional health data, data-driven intelligent fault prediction can be achieved, thereby improving the timeliness and accuracy of electrical fault detection.

[0027] S12: Pre-trained electrical fault predictor.

[0028] Furthermore, step S12 of the present invention further includes:

[0029] S121: Based on the historical operation and maintenance records of the conveyor line system, a sample multidimensional health data set is collected, and the proportion of electrical faults and high-frequency fault types of different sample multidimensional health data in the historical time zone are counted, which are set as sample fault probabilities and sample high-frequency fault types, and a sample fault probability set and a sample high-frequency fault type set are obtained, wherein the high-frequency fault type includes at least two fault types; S122: With the sample multidimensional health data set as input and the sample fault probability set and the sample high-frequency fault type set as supervision, a BP neural network is trained until the model converges to obtain the electrical fault predictor.

[0030] Specifically, first, based on the historical operation and maintenance records of the conveyor line system, sample multidimensional health data is collected. The health data set of each sample includes multidimensional data such as electrical parameters, control signals, load information, and mechanical status collected from the operation of the conveyor belt, thereby obtaining a sample multidimensional health data set. Next, the proportion of electrical faults and high-frequency fault types in the historical time zone of different sample multidimensional health data are counted. That is, by analyzing the health data of each sample in different historical time zones, the electrical fault incidence rate (i.e., the proportion of electrical faults) of each sample is counted. These statistical values ​​reflect the frequency of electrical faults in the conveyor line system in a specific time zone, providing an important reference for fault prediction. Frequently occurring fault types are extracted from historical data, such as overload faults, power failures, loss of control signals, etc. Statistical results of at least two fault types are included to help identify common fault types in the system.

[0031] Next, an electrical fault predictor is constructed based on a BP neural network. The BP neural network typically consists of an input layer, a hidden layer, and an output layer. The input layer receives multidimensional health data, while the output layer outputs the electrical fault prediction results, including fault probability and fault type. The number of nodes and layers in the hidden layer can be adjusted based on the actual data volume and problem complexity. The electrical fault predictor is then supervised and trained using the sample multidimensional health data set as input and the sample fault probability set and sample high-frequency fault type set as supervision. During the training process, the input data is first passed into the network in each iteration. After the weights and activation functions of each layer are calculated, the prediction result is finally output. Next, the network's performance is gradually optimized by calculating the gradient of the loss function and using the gradient descent algorithm to adjust the weights in the network. The loss function is a function used to evaluate the gap between the model's prediction results and the actual target, with the goal of minimizing this gap and thus improving the model's accuracy. By calculating the gradient of the loss function and performing gradient descent, the network can gradually adjust its weights so that the prediction results are closer to the true value. Then, through continuous iterative training, the model adjusts its parameters based on the input data until the neural network's loss function stabilizes over multiple iterations, indicating that the network has reached convergence. Typically, convergence can be determined by setting a maximum number of iterations or an accuracy threshold. After training, the electrical fault predictor can predict the probability of future electrical faults and the types of high-frequency faults that are likely to occur based on new multi-dimensional health data, providing intelligent early warning and support for operational and maintenance decisions.

[0032] By training the BP neural network model, the patterns and characteristics of electrical faults can be learned from historical data, thereby achieving accurate fault prediction. This predictor can not only predict the probability of electrical faults, but also identify potential high-frequency fault types, providing data-driven decision support for operation and maintenance personnel.

[0033] S13: Utilizing the electrical fault predictor, respectively perform electrical fault predictions based on the plurality of multi-dimensional health data, and output a plurality of predicted fault probabilities and a plurality of high-frequency fault type sets.

[0034] Specifically, the electrical fault predictor is finally used to perform electrical fault predictions based on the multiple multidimensional health data. The electrical fault predictor compares and analyzes the input multidimensional health data with the previously trained model. The model predicts the possibility of a fault (fault probability) and the possible type of fault based on the learned historical data patterns. The fault probability value is between 0 and 1, and a larger value represents a higher risk of a fault.

[0035] S20: Determine a plurality of early warning conveyor belts based on the plurality of predicted fault probabilities, and calculate and obtain a plurality of predicted maintenance durations based on the plurality of high-frequency fault type sets.

[0036] Furthermore, step S20 of the present invention further includes:

[0037] S21: Configure a warning probability threshold, set the conveyor belt with a predicted failure probability greater than the warning probability threshold as a warning conveyor belt, determine multiple warning conveyor belts based on the several predicted failure probabilities, and match and obtain multiple high-frequency fault type sets of the multiple warning conveyor belts; S22: According to the historical operation and maintenance records of the conveyor line system, analyze and obtain the average historical fault maintenance time of the high-frequency fault type within the preset historical time range; S23: Set weights based on the proportion of fault types in the high-frequency fault types, calculate the maintenance time for the multiple high-frequency fault type sets according to the average historical fault maintenance time, and output multiple predicted maintenance times, wherein the weight is positively correlated with the proportion of the fault type.

[0038] Specifically, first, configure the warning probability threshold. The warning probability threshold is a set value used to determine which conveyor belts have a higher risk of failure. It is usually set by operation and maintenance personnel based on the equipment's failure history, operating environment, and safety requirements. It can be adjusted according to specific operation and maintenance strategies to optimize the allocation of operation and maintenance resources. For example, the warning probability threshold is set to 50%. Then, by comparing the predicted failure probability of each conveyor belt with the warning probability threshold, the conveyor belt with a predicted failure probability greater than the warning probability threshold is set as a warning conveyor belt; and based on several high-frequency fault type sets, multiple high-frequency fault type sets of the multiple warning conveyor belts are matched and obtained. By setting the warning probability threshold, it is possible to intelligently screen out the warning conveyor belts that need to be focused on, and perform targeted operation and maintenance processing based on the fault type.

[0039] Next, based on the historical operation and maintenance records of the conveyor line system, multiple historical fault repair times of high-frequency fault types within a preset historical time range (such as within the last month) are collected. That is, for each high-frequency fault type, the repair time in the historical records is analyzed. The repair time for each fault will be different, usually determined by factors such as the severity of the fault, the difficulty of repair, and the supply of spare parts; then, the average of multiple historical fault repair times is calculated to obtain the average of multiple historical fault repair times for multiple high-frequency fault types, where the high-frequency fault types and the average of historical fault repair times correspond one to one.

[0040] Then, through statistical analysis of historical fault data, the proportion of each high-frequency fault type in the total fault population is determined. This proportion reflects the frequency of each fault type, that is, the probability of a fault occurring. Furthermore, a weight is set based on the proportion of each fault type. The weight is positively correlated with the proportion of the fault type, that is, the higher the proportion of the fault type, the greater the weight assigned. In this way, the impact of different fault types on the overall maintenance time can be more accurately reflected. Next, based on the weights of multiple high-frequency fault types, a weighted calculation is performed on the mean historical fault maintenance duration corresponding to each fault type in each high-frequency fault type set, resulting in multiple predicted maintenance durations corresponding to multiple high-frequency fault type sets. By combining the proportion of fault types and the mean historical maintenance duration, the maintenance duration of future faults can be more accurately predicted, providing stronger data support for scheduling and resource allocation.

[0041] S30: generating a plurality of initial maintenance plans within a preset time zone according to the plurality of early warning conveyor belts and the plurality of predicted maintenance durations.

[0042] Furthermore, step S30 of the present invention further includes:

[0043] S31: Configure the maximum number of collaborative maintenance of the conveyor belts; S32: With the maximum number of collaborative maintenance as a constraint, within a preset time zone, enumerate the maintenance order of the warning conveyor belts according to the multiple warning conveyor belts and multiple predicted maintenance durations, and output multiple initial maintenance plans, wherein in the initial maintenance plan, the working time interval between adjacent warning conveyor belts is greater than or equal to 0.

[0044] Specifically, first, configure the maximum number of collaborative maintenance of conveyor belts. The maximum number of collaborative maintenance refers to the maximum number of conveyor belts allowed to be maintained simultaneously at any time. For example, assuming that the system is configured with a maximum collaborative maintenance number of 3, it means that at most 3 groups of conveyor belts can be maintained simultaneously at any time. Then, with the maximum number of coordinated maintenance as a constraint, within a preset time zone (such as the next 12 hours), the maintenance order of the warning conveyor belts is enumerated according to the multiple warning conveyor belts and multiple predicted maintenance durations, that is, according to the preset maximum number of coordinated maintenance and the predicted maintenance durations of multiple warning conveyor belts, multiple maintenance orders are enumerated to determine possible initial maintenance plans. In each initial maintenance plan, maintenance is arranged in a certain order, and it is ensured that the number of simultaneous maintenances in the same time zone does not exceed the set maximum number of coordinated maintenance. In order to ensure that each maintenance does not interfere with the operation of other conveyor belts, the working time interval between adjacent warning conveyor belts is required to be greater than or equal to 0. That is to say, in some cases, the maintenance work of the conveyor belt can be carried out in sections, that is, one conveyor belt is inspected first, and the other conveyor belts continue to run, and the next conveyor belt is inspected when it is idle; multiple initial maintenance plans are output. By constraining the maximum number of coordinated maintenance tasks and enumerating the maintenance sequence, maintenance tasks can be efficiently planned to avoid resource waste and long-term system downtime. While ensuring that the maximum number of coordinated maintenance tasks is not exceeded, the maintenance time of each conveyor belt can be reasonably arranged to ensure that the package transportation within the logistics park is not overly disturbed, thereby improving system throughput. Through the flexible configuration of working time intervals, maintenance plans can be adjusted according to real-time operation and maintenance needs, maximizing equipment utilization and maintaining a good business operation status.

[0045] S40: With the probability of failure of the conveyor belt being less than a threshold value as a constraint and with the goal of maximizing the package throughput and minimizing the maintenance cycle, the multiple initial maintenance plans are optimized to determine the optimal maintenance plan.

[0046] Furthermore, step S40 of the present invention further includes:

[0047] S41: Randomly select a first maintenance plan from the multiple initial maintenance plans, calculate the time intervals between the maintenance time nodes of the multiple warning conveyor belts in the first maintenance plan and the current time node, and obtain multiple maintenance time intervals; S42: According to the historical operation and maintenance records of the conveyor line system, perform a correlation analysis on the transportation speed of the conveyor belt and the growth trend of the failure probability, and construct a speed-trend coefficient comparison table, wherein the transportation speed and the failure probability growth trend coefficient are positively correlated; S43: Based on the speed-trend coefficient comparison table, with a constraint of being less than the conveyor belt failure probability threshold value, calculate multiple first maximum transportation speeds of the multiple warning conveyor belts according to the multiple predicted failure probabilities of the multiple warning conveyor belts and the multiple maintenance time intervals; S44: Obtain the first maintenance cycle of the first maintenance plan, wherein the maintenance cycle is the time interval between the current time node and the time node when all the warning conveyor belts in the maintenance plan complete the maintenance.

[0048] Specifically, first, any one of the multiple initial maintenance plans is randomly selected as the first maintenance plan. Then, the time intervals between the maintenance time nodes of the multiple warning conveyor belts in the first maintenance plan and the current time nodes are calculated respectively. That is, for each warning conveyor belt in the first maintenance plan, according to its arrangement in the plan, its corresponding planned maintenance start time node is obtained, and compared with the current system time (that is, the time for executing optimized scheduling), and the time difference is calculated and set as the maintenance time interval of each warning conveyor belt, thereby obtaining multiple maintenance time intervals.

[0049] Next, based on the historical operation and maintenance records of the conveyor line system, a correlation analysis was conducted between the conveyor belt's transportation speed and the growth trend of the failure probability. For example, the hourly growth rate of the electrical failure probability at speeds of 0.8 m / s, 1.0 m / s, and 1.2 m / s was calculated. The corresponding relationship between the transportation speed and the failure growth rate was obtained through methods such as linear fitting and sliding window regression. The transportation speed was then divided into multiple intervals (such as low speed, medium speed, and high speed). A trend coefficient (failure growth coefficient) was assigned to each speed interval to reflect its weighted value for failure risk. The transportation speed and the failure probability growth trend coefficient were positively correlated. That is, the higher the transportation speed, the faster the risk growth and the larger the failure probability growth trend coefficient. For example, for a transportation speed interval of 0.5 to 0.8 m / s, the failure probability growth trend coefficient per unit time (such as 6 hours) was 0.01; for a transportation speed interval of 0.8 to 1.2 m / s, the failure probability growth trend coefficient per unit time (such as 6 hours) was 0.02, and so on. A speed-trend coefficient comparison table was constructed.

[0050] Then, a conveyor belt failure probability threshold is obtained. The conveyor belt failure probability threshold can be set according to the actual scenario. For example, the failure probability threshold is set to 75%. That is, when the predicted failure probability of a conveyor belt is greater than 75%, it is regarded as an unacceptable risk and maintenance must be arranged. Next, with the constraint of being less than the conveyor belt failure probability threshold, based on the speed-trend coefficient comparison table, and according to the multiple predicted failure probabilities of the multiple warning conveyor belts and the multiple maintenance time intervals, multiple first maximum transport speeds of the multiple warning conveyor belts are calculated. That is, with "predicted failure probability + failure growth trend * maintenance time interval < failure probability threshold" as the constraint condition, based on the speed-trend coefficient comparison table, the first maximum transport speed allowed for each warning conveyor belt during this period is calculated in combination with the current predicted failure probability and the corresponding maintenance time interval. For example, assuming that the predicted failure probability of the first warning conveyor belt is 30%, the maintenance time interval is 24 hours, the maximum transport speed of the conveyor belt is 3 meters per second, and at the maximum transport speed, the failure probability growth trend coefficient per unit time (e.g., 6 hours) is 0.04, then the failure probability of the first warning conveyor belt after the maintenance time interval is 0.3 + 24 / 6 * 0.04, which is equal to 0.46, which is less than 75% of the failure probability threshold. Therefore, the first maximum transport speed of the warning conveyor belt is 3 meters per second.

[0051] Next, the first maintenance cycle of the first maintenance plan is obtained, wherein the maintenance cycle is the time interval between the current time node and the time node when all warning conveyor belts in the maintenance plan complete maintenance, which is used to measure the operation and maintenance duration of the maintenance plan within the preset time zone, as well as the urgency of conveyor belt operation and maintenance. The shorter the maintenance cycle, the more stable the overall operating status of the warning conveyor belt, the more timely the fault risk control, and the higher the potential stability of the system.

[0052] S45: With the goal of maximizing the package throughput and minimizing the maintenance cycle, calculate the solution fitness according to the multiple first maximum transportation speeds and the first maintenance cycle, and output the first solution fitness.

[0053] Furthermore, step S45 of the present invention further includes:

[0054] S451: Perform a parcel transportation simulation based on the multiple first maximum transportation speeds and the first maintenance plan, and output a first parcel throughput; S452: Construct a scheme fitness evaluation function with the goal of maximizing the parcel throughput and minimizing the maintenance cycle, wherein the scheme fitness evaluation function includes an initial weight ratio, and the scheme fitness is positively correlated with the parcel throughput and negatively correlated with the maintenance cycle.

[0055] Specifically, parcel throughput refers to the number of parcels transported through the conveyor line system within a certain period of time, usually measured in "parcels / hour" or "parcels / day". In a logistics park, parcel throughput is one of the key indicators for measuring the efficiency of the system operation and directly affects the operational benefits of the entire park. First, a parcel transportation simulation is performed based on the multiple first maximum transportation speeds and the first maintenance plan. A mathematical model or simulation system is established to simulate the operation of the conveyor line system within a certain period of time. During the simulation, the number of parcels that can be processed per unit time is estimated based on the set maximum transportation speed of each conveyor belt. At the same time, the impact of the maintenance cycle and transportation speed on parcel transportation is considered. For example, when the conveyor belt is undergoing maintenance, the transportation speed drops to zero or other lower speeds, affecting the speed of the parcels passing through the path, thereby reducing the overall throughput. The first parcel throughput is output, that is, the number of parcels that the system can handle within the set time period under the current maintenance plan and the maximum transportation speed constraints. This value reflects the operational efficiency of the system under the current plan and can provide a basis for subsequent optimization.

[0056] Next, with the goal of maximizing package throughput and minimizing maintenance cycles, a solution fitness evaluation function was constructed. This function aims to maximize the number of packages that can be transported per unit time. Higher package throughput means more efficient system operation. The time required for each maintenance plan was reduced, meaning the system downtime was minimized while minimizing the impact of maintenance on transportation. The solution fitness evaluation function includes initial weight ratios, assigning different weight ratios to package throughput and maintenance cycles. These weights can be adjusted based on actual needs. For example, if package throughput is considered more important than maintenance cycles, the weight of package throughput will be higher, and vice versa. The fitness value is positively correlated with package throughput, meaning that higher package throughput increases the solution's fitness. In other words, better transportation efficiency leads to a better solution. The fitness value is negatively correlated with maintenance cycles, meaning that shorter maintenance cycles increase the solution's fitness and achieve better results.

[0057] S453: Optimizing the initial weight ratio according to the first package throughput to determine a first optimized weight ratio, updating the solution fitness evaluation function to determine an updated solution fitness evaluation function.

[0058] Furthermore, step S453 of the present invention further includes:

[0059] S4531: Configure an initial weight ratio, wherein the initial weight ratio includes a preset throughput weight and a preset maintenance cycle weight, and the sum of the two is 1; S4532: Obtain the throughput of packages to be transported in a preset time zone; S4533: If the throughput of packages to be transported is less than or equal to the first package throughput, set the initial weight ratio to a first optimized weight ratio; S4534: If the throughput of packages to be transported is greater than the first package throughput, calculate the throughput deviation ratio between the throughput of packages to be transported and the first package throughput; S4535: Multiply the throughput deviation ratio by a constant, and add the sum to 1 to obtain a first adjustment coefficient, wherein the constant is greater than 1 and less than 5; S4536: Multiply the first adjustment coefficient by the preset throughput weight, and set the product of the two as the optimized throughput weight; S4537: Subtract the optimized throughput weight from 1 to obtain the optimized maintenance cycle weight, and set the optimized throughput weight and the optimized maintenance cycle weight as the first optimized weight ratio.

[0060] Specifically, first, configure the initial weight ratio. The initial weight ratio includes a preset throughput weight (used to measure the importance of package throughput in optimization) and a preset maintenance cycle weight (used to measure the importance of maintenance cycle in optimization). The sum of the two is 1. For example, if the preset throughput weight is set to 0.5, the preset maintenance cycle weight is also set to 0.5. Next, obtain the throughput of packages to be transported within a preset time zone. Then, the first package throughput is judged based on the throughput of packages to be transported. If the throughput of packages to be transported is less than or equal to the first package throughput, it indicates that the transportation demand is small. The optimization weight of the current solution can directly use the initial weight ratio without adjustment. In this case, the first optimization weight ratio is set equal to the initial weight ratio.

[0061] If the throughput of packages to be transported is greater than the first throughput, it indicates increased transport demand, and the system needs to adjust the weights to optimize the solution. The system then calculates the throughput deviation ratio between the throughput of packages to be transported and the first throughput. The throughput deviation ratio is the ratio of the difference between the throughput of packages to be transported and the first throughput to the throughput of packages to be transported. For example, if the current demand is 1,000 packages and the preset throughput is 800 packages, the throughput deviation ratio is 1.25, indicating that the demand is 25% higher than the preset solution. The throughput deviation ratio is then multiplied by a constant, and the sum is added to 1, where the constant is greater than 1 and less than 5. The sum of the two is used as the first adjustment coefficient. For example, if the throughput deviation ratio is 0.25 and the constant is 1, the first adjustment coefficient is 1 + 0.25 * 1, which equals 1.25. Therefore, the first adjustment coefficient is 1.25. Next, the first adjustment coefficient is multiplied by the preset throughput weight, and the product of the two is set as the optimized throughput weight. The optimized throughput weight is then subtracted from 1 to obtain the optimized maintenance cycle weight. For example, assuming the first adjustment coefficient is 1.25 and the preset throughput weight is 0.5, the optimized throughput weight is 1.25*0.5, which equals 0.625. The optimized maintenance cycle weight is 1-0.625, which equals 0.375. Finally, the optimized throughput weight and the optimized maintenance cycle weight are set as the first optimized weight ratio.

[0062] By setting dynamic weights, when actual demand exceeds the preset throughput, the operation and maintenance weights can be adjusted according to the throughput deviation, giving priority to increasing the weight of throughput (i.e. optimizing transportation speed) while correspondingly reducing the weight of the maintenance cycle. This ensures that when transportation demand increases, the throughput is maximized while avoiding excessive burden on the maintenance cycle. This dynamic adjustment ensures the efficient operation of the system and avoids efficiency losses caused by too much or too little maintenance.

[0063] Finally, the scheme fitness evaluation function is updated using the first optimized weight ratio to obtain an updated scheme fitness evaluation function.

[0064] S454: Using the updated solution fitness evaluation function, calculate the first solution fitness according to the first package throughput and the first maintenance cycle.

[0065] Specifically, the updated solution fitness evaluation function is then used to calculate the solution fitness according to the first package throughput and the first maintenance cycle to obtain the first solution fitness.

[0066] S46: Continue analyzing to obtain multiple scheme fitnesses of the multiple initial maintenance schemes, and select the initial maintenance scheme with the maximum scheme fitness as the optimal maintenance scheme.

[0067] Specifically, the same method used to calculate the fitness of the first solution is used to continue analyzing and obtaining the fitness of multiple solutions of the multiple initial maintenance solutions, and the initial maintenance solution with the maximum solution fitness is selected as the optimal maintenance solution. In this way, the maintenance solution that best suits the current transportation needs and operation and maintenance resources can be selected, thereby achieving the optimal balance between package throughput and maintenance cycle, and meeting different business needs and operation and maintenance requirements.

[0068] S50: In the preset time zone, the plurality of early warning conveyor belts are operated and maintained according to the optimal maintenance plan.

[0069] Specifically, finally, within the preset time zone, the multiple warning conveyor belts are operated and maintained according to the optimal maintenance plan to ensure that the system maximizes the parcel throughput while ensuring safety, while reducing downtime and improving the overall operating efficiency of the logistics park conveyor line system.

[0070] In summary, the method for detecting electrical faults in campus facilities provided by the present invention has the following technical effects:

[0071] By monitoring and acquiring a plurality of multi-dimensional health data of a plurality of conveyor belts in a conveyor line system within a logistics park, electrical fault prediction is performed based on the plurality of multi-dimensional health data, and a plurality of predicted fault probabilities and a plurality of high-frequency fault type sets are outputted. Then, based on the plurality of predicted fault probabilities, a plurality of early warning conveyor belts are determined, and a plurality of predicted maintenance durations are calculated based on the plurality of high-frequency fault type sets. Furthermore, based on the plurality of early warning conveyor belts and the plurality of predicted maintenance durations, a plurality of initial maintenance plans are generated within a preset time zone. Then, with a constraint of being less than a conveyor belt failure probability threshold, and with the goal of maximizing package throughput and minimizing maintenance cycles, the plurality of initial maintenance plans are optimized to determine the optimal maintenance plan. Finally, within the preset time zone, the plurality of early warning conveyor belts are operated and maintained according to the optimal maintenance plan. In other words, by combining electrical fault prediction with resource scheduling optimization, dynamic operation and maintenance scheduling based on fault risk can be achieved. This can maximize package throughput while ensuring system safety, and reduce downtime through precise maintenance planning, thereby improving the overall operating efficiency of the conveyor line system in the logistics park and significantly improving system throughput and reliability.

[0072] In the second embodiment, based on the same inventive concept as the method for detecting electrical faults in a park facility in the above embodiment, the present invention also provides a system for detecting electrical faults in a park facility. Figure 2, including: an electrical fault prediction module 11, which is used to monitor and obtain a number of multi-dimensional health data of a number of conveyor belts in the conveyor line system in the logistics park, and perform electrical fault predictions based on the said several multi-dimensional health data, and output a number of predicted fault probabilities and a number of high-frequency fault type sets; a predicted maintenance time calculation module 12, which is used to determine a number of warning conveyor belts based on the said several predicted fault probabilities, and calculate and obtain a number of predicted maintenance time periods based on the said several high-frequency fault type sets; an initial maintenance plan generation module 13, which is used to generate a number of initial maintenance plans within a preset time zone based on the said multiple warning conveyor belts and the multiple predicted maintenance time periods; an initial maintenance plan optimization module 14, which is used to optimize the said multiple initial maintenance plans with a constraint of being less than a conveyor belt failure probability threshold value, with the goal of maximizing the package throughput and minimizing the maintenance cycle, and determine the optimal maintenance plan; an operation and maintenance control module 15, which is used to perform operation and maintenance on the said multiple warning conveyor belts according to the said optimal maintenance plan in the said preset time zone.

[0073] Furthermore, the park facility electrical fault detection system is also used to: monitor and obtain a number of multidimensional health data of a number of conveyor belts in the conveyor line system in the logistics park within a preset historical time window, wherein the multidimensional health data includes at least electrical parameters, control signals and operating parameters; pre-train an electrical fault predictor; and use the electrical fault predictor to perform electrical fault predictions based on the number of multidimensional health data, and output a number of predicted fault probabilities and a number of high-frequency fault type sets.

[0074] Furthermore, the campus facility electrical fault detection system is also used to: collect a sample multidimensional health data set based on the historical operation and maintenance records of the conveyor line system, and count the electrical fault proportions and high-frequency fault types of different sample multidimensional health data in the historical time zone, set them as sample fault probabilities and sample high-frequency fault types, and obtain a sample fault probability set and a sample high-frequency fault type set, wherein the high-frequency fault type includes at least two fault types; use the sample multidimensional health data set as input and the sample fault probability set and the sample high-frequency fault type set as supervision to train the BP neural network until the model converges to obtain the electrical fault predictor.

[0075] Furthermore, the campus facility electrical fault detection system is also used to: configure a warning probability threshold, set a conveyor belt with a predicted failure probability greater than the warning probability threshold as a warning conveyor belt, determine multiple warning conveyor belts based on the several predicted failure probabilities, and match and obtain multiple high-frequency fault type sets of the multiple warning conveyor belts; analyze and obtain the average historical fault maintenance time of high-frequency fault types within a preset historical time range based on the historical operation and maintenance records of the conveyor line system; set weights based on the proportion of fault types in high-frequency fault types, and calculate the maintenance time of the multiple high-frequency fault type sets based on the average historical fault maintenance time, and output multiple predicted maintenance times, wherein the weights are positively correlated with the proportion of fault types.

[0076] Furthermore, the campus facility electrical fault detection system is also used to: configure the maximum number of collaborative maintenance of conveyor belts; with the maximum number of collaborative maintenance as a constraint, within a preset time zone, enumerate the maintenance order of the warning conveyor belts according to the multiple warning conveyor belts and multiple predicted maintenance durations, and output multiple initial maintenance plans, wherein in the initial maintenance plan, the working time interval between adjacent warning conveyor belts is greater than or equal to 0.

[0077] Furthermore, the electrical fault detection system for park facilities is also used to: randomly select a first maintenance plan from the multiple initial maintenance plans, calculate the time intervals between the maintenance time nodes of multiple early warning conveyor belts in the first maintenance plan and the current time nodes, and obtain multiple maintenance time intervals; perform a correlation analysis on the transportation speed of the conveyor belt and the growth trend of the failure probability based on the historical operation and maintenance records of the conveyor line system, and construct a speed-trend coefficient comparison table, wherein the transportation speed and the growth trend coefficient of the failure probability are positively correlated; based on the speed-trend coefficient comparison table, with the value less than the conveyor belt failure probability threshold as a constraint, Based on multiple predicted failure probabilities and the multiple maintenance time intervals, multiple first maximum transport speeds of multiple early warning conveyor belts are calculated; a first maintenance cycle of the first maintenance plan is obtained, wherein the maintenance cycle is the time interval between the current time node and the time node when all the early warning conveyor belts in the maintenance plan complete the maintenance; with the goal of maximizing the package throughput and minimizing the maintenance cycle, the plan fitness is calculated according to the multiple first maximum transport speeds and the first maintenance cycle, and the first plan fitness is output; the multiple plan fitnesses of the multiple initial maintenance plans are continued to be analyzed and obtained, and the initial maintenance plan with the maximum plan fitness is selected as the optimal maintenance plan.

[0078] Furthermore, the campus facility electrical fault detection system is also used to: perform a parcel transportation simulation based on the multiple first maximum transportation speeds and the first maintenance plan, and output a first parcel throughput; construct a scheme fitness evaluation function with the goal of maximizing the parcel throughput and minimizing the maintenance cycle, wherein the scheme fitness evaluation function includes an initial weight ratio, and the scheme fitness is positively correlated with the parcel throughput and negatively correlated with the maintenance cycle; optimize the initial weight ratio based on the first parcel throughput to determine a first optimized weight ratio, update the scheme fitness evaluation function, and determine an updated scheme fitness evaluation function; use the updated scheme fitness evaluation function to calculate the first scheme fitness based on the first parcel throughput and the first maintenance cycle.

[0079] Furthermore, the campus facility electrical fault detection system is also used to: configure an initial weight ratio, wherein the initial weight ratio includes a preset throughput weight and a preset maintenance cycle weight, and the sum of the two is 1; obtain the throughput of packages to be transported in a preset time zone; if the throughput of packages to be transported is less than or equal to the first package throughput, set the initial weight ratio to a first optimized weight ratio; if the throughput of packages to be transported is greater than the first package throughput, calculate the throughput deviation ratio between the throughput of packages to be transported and the first package throughput; multiply the throughput deviation ratio by a constant, and add the sum to 1 to obtain a first adjustment coefficient, wherein the constant is greater than 1 and less than 5; multiply the first adjustment coefficient by the preset throughput weight, and set the product of the two as the optimized throughput weight; subtract the optimized throughput weight from 1 to obtain the optimized maintenance cycle weight, and set the optimized throughput weight and the optimized maintenance cycle weight as the first optimized weight ratio.

[0080] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The method and specific examples of a campus facility electrical fault detection method in the aforementioned embodiment 1 are also applicable to a campus facility electrical fault detection system in this embodiment. Through the aforementioned detailed description of a campus facility electrical fault detection method, those skilled in the art can clearly understand a campus facility electrical fault detection system in this embodiment, so for the sake of brevity of the specification, it will not be described in detail here. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method section.

[0081] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

[0082] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalents, the present invention is intended to include these modifications and variations.

Claims

1. A method for detecting electrical faults in park facilities, characterized in that: include: Monitor and obtain a plurality of multi-dimensional health data of a plurality of conveyor belts of a conveyor line system in a logistics park, perform electrical fault prediction based on the plurality of multi-dimensional health data, and output a plurality of predicted fault probabilities and a plurality of high-frequency fault type sets; Determining a plurality of early warning conveyor belts based on the plurality of predicted fault probabilities, and calculating and obtaining a plurality of predicted maintenance durations based on the plurality of high-frequency fault type sets; generating a plurality of initial maintenance plans within a preset time zone according to the plurality of early warning conveyor belts and the plurality of predicted maintenance durations; Under the constraint of being less than a conveyor belt failure probability threshold value, and with the goal of maximizing package throughput and minimizing maintenance cycle, the multiple initial maintenance plans are optimized to determine the optimal maintenance plan; In the preset time zone, operating and maintaining the plurality of early warning conveyor belts according to the optimal maintenance plan; The optimization of the multiple initial maintenance plans is performed with the constraint of being less than the conveyor belt failure probability threshold and the goal of maximizing the package throughput and minimizing the maintenance cycle to determine the optimal maintenance plan, including: Randomly selecting a first maintenance plan from the multiple initial maintenance plans, and respectively calculating the time intervals between the maintenance time nodes of the multiple warning conveyor belts in the first maintenance plan and the current time node to obtain multiple maintenance time intervals; Based on the historical operation and maintenance records of the conveyor line system, a correlation analysis was conducted between the conveyor belt's transportation speed and the growth trend of the failure probability, and a speed-trend coefficient comparison table was constructed. Among them, the transportation speed and the growth trend coefficient of the failure probability are positively correlated. Based on the speed-trend coefficient comparison table, with the value being less than a conveyor belt failure probability threshold as a constraint, and according to the multiple predicted failure probabilities of the multiple early warning conveyor belts and the multiple maintenance time intervals, multiple first maximum transport speeds of the multiple early warning conveyor belts are calculated; Obtaining a first maintenance cycle of the first maintenance plan, wherein the maintenance cycle is the time interval between the current time node and the time node when all warning conveyor belts in the maintenance plan are completed; With the goal of maximizing package throughput and minimizing maintenance cycle, calculate the solution fitness based on the multiple first maximum transportation speeds and first maintenance cycles, and output a first solution fitness; Continuing to analyze and obtain multiple scheme fitnesses of the multiple initial maintenance schemes, and selecting the initial maintenance scheme with the maximum scheme fitness as the optimal maintenance scheme; The goal is to maximize the package throughput and minimize the maintenance cycle, calculate the solution fitness based on the multiple first maximum transportation speeds and the first maintenance cycle, and output the first solution fitness, including: performing a parcel transport simulation according to the plurality of first maximum transport speeds and the first maintenance plan, and outputting a first parcel throughput; With the goal of maximizing package throughput and minimizing maintenance cycle, a scheme fitness evaluation function is constructed. The scheme fitness evaluation function includes the initial weight ratio. The scheme fitness is positively correlated with package throughput and negatively correlated with maintenance cycle. Optimizing the initial weight ratio according to the first package throughput to determine a first optimized weight ratio, updating the solution fitness evaluation function to determine an updated solution fitness evaluation function; Utilizing the update solution fitness evaluation function, the fitness of the first solution is calculated according to the first package throughput and the first maintenance cycle; Optimizing the initial weight ratio according to the first package throughput to determine a first optimized weight ratio includes: Configuring an initial weight ratio, wherein the initial weight ratio includes a preset throughput weight and a preset maintenance cycle weight, the sum of which is 1; Get the throughput of packages to be transported in the preset time zone; If the throughput of the packages to be transported is less than or equal to the first throughput of the packages, setting the initial weight ratio to a first optimized weight ratio; If the throughput of the packages to be transported is greater than the first throughput of the packages, calculating a throughput deviation ratio between the throughput of the packages to be transported and the first throughput of the packages; Multiplying the throughput deviation ratio by a constant, and adding the result to 1 to obtain a first adjustment coefficient, wherein the constant is greater than 1 and less than 5; multiplying the first adjustment coefficient by the preset throughput weight, and setting the product of the two as the optimized throughput weight; The optimized throughput weight is subtracted from 1 to obtain the optimized maintenance cycle weight, and the optimized throughput weight and the optimized maintenance cycle weight are set as a first optimized weight ratio.

2. A method for detecting electrical faults in park facilities according to claim 1, characterized in that: Monitor and obtain multiple multi-dimensional health data of multiple conveyor belts in the conveyor line system in the logistics park, perform electrical fault prediction based on the multiple multi-dimensional health data, and output multiple predicted fault probabilities and multiple high-frequency fault type sets, including: Within a preset historical time window, monitor and obtain a plurality of multi-dimensional health data of a plurality of conveyor belts of a conveyor line system in a logistics park, wherein the multi-dimensional health data includes at least electrical parameters, control signals, and operating parameters; Pre-trained electrical fault predictor; The electrical fault predictor is used to perform electrical fault predictions based on the plurality of multi-dimensional health data, and outputs a plurality of predicted fault probabilities and a plurality of high-frequency fault type sets.

3. A method for detecting electrical faults in park facilities according to claim 2, characterized in that: Pre-trained electrical fault predictor, including: Based on the historical operation and maintenance records of the conveyor line system, a sample multidimensional health data set is collected, and the proportion of electrical faults and high-frequency fault types in different sample multidimensional health data in the historical time zone are counted. These are set as sample fault probability and sample high-frequency fault type, and a sample fault probability set and a sample high-frequency fault type set are obtained. The high-frequency fault type includes at least two fault types. The sample multidimensional health data set is used as input, and the sample fault probability set and the sample high-frequency fault type set are used as supervision to train a BP neural network until the model converges, thereby obtaining the electrical fault predictor.

4. A method for detecting electrical faults in park facilities according to claim 1, characterized in that: Determining multiple early warning conveyor belts based on the multiple predicted fault probabilities and calculating and obtaining multiple predicted maintenance durations based on the multiple high-frequency fault type sets includes: Configuring a warning probability threshold, setting a conveyor belt with a predicted failure probability greater than the warning probability threshold as a warning conveyor belt, determining multiple warning conveyor belts based on the multiple predicted failure probabilities, and matching and obtaining multiple high-frequency fault type sets for the multiple warning conveyor belts; Based on the historical operation and maintenance records of the conveyor line system, analyze and obtain the average historical fault repair time of high-frequency fault types within a preset historical time range; Weights are set based on the proportion of fault types in high-frequency fault types, and according to the average historical fault repair duration, repair durations are calculated for the multiple high-frequency fault type sets respectively, and multiple predicted repair durations are output, wherein the weights are positively correlated with the proportion of fault types.

5. A method for detecting electrical faults in park facilities according to claim 1, characterized in that: Based on the multiple early warning conveyor belts and the multiple predicted maintenance durations, multiple initial maintenance plans within a preset time zone are generated, including: Configure the maximum number of coordinated overhauls for the conveyor belt; With the maximum number of collaborative maintenance as a constraint, within a preset time zone, the maintenance order of the warning conveyor belts is enumerated according to the multiple warning conveyor belts and the multiple predicted maintenance durations, and multiple initial maintenance plans are output, wherein in the initial maintenance plan, the working time interval between adjacent warning conveyor belts is greater than or equal to 0.

6. A campus facility electrical fault detection system, characterized in that: The steps for implementing the method for detecting electrical faults in campus facilities according to any one of claims 1 to 5 include: An electrical fault prediction module is used to monitor and obtain a plurality of multi-dimensional health data of a plurality of conveyor belts in the conveyor line system of the logistics park, perform electrical fault prediction based on the plurality of multi-dimensional health data, and output a plurality of predicted fault probabilities and a plurality of high-frequency fault type sets; A predicted maintenance duration calculation module is used to determine multiple early warning conveyor belts based on the multiple predicted fault probabilities and calculate multiple predicted maintenance durations based on the multiple high-frequency fault type sets; An initial maintenance plan generating module, configured to generate a plurality of initial maintenance plans within a preset time zone according to the plurality of warning conveyor belts and the plurality of predicted maintenance durations; An initial maintenance plan optimization module is used to optimize the multiple initial maintenance plans and determine the optimal maintenance plan, with the constraint of being less than a conveyor belt failure probability threshold value and the goal of maximizing package throughput and minimizing maintenance cycle; An operation and maintenance control module is used to perform operation and maintenance on the multiple early warning conveyor belts according to the optimal maintenance plan in the preset time zone.

Images