Mobile mechanical equipment fault maintenance service system based on big data

Abstract

The invention relates to the technical field of maintenance service management, in particular to a mobile mechanical equipment fault maintenance service system based on big data, which comprises a state recognition module, a migration judgment module, an offset correction module, a task aggregation module and a path sorting module. According to the method, the fault feature mapping precision is improved by screening the linear correlation parameters of the rotating speed, the load, the temperature and the fault number, the migration decision adaptability is enhanced by numerical distance conversion and interval dynamic judgment, the temperature difference and the load correction coefficient are fused, and the working condition compatibility and the anti-interference capability of load calculation are optimized; the method comprises the following steps: cooperating with equipment position, fault degree and association strength, generating an optimal maintenance combination under space-time constraint, strengthening task aggregation efficiency, combining traffic data and remaining time to grade urgency, dynamically adjusting priority, and forming closed-loop optimization through a full chain from feature extraction to decision execution by a data-driven feedback mechanism. And the fault handling efficiency and the resource configuration accuracy are systematically improved.

Description

Technical Field

[0001] The present invention relates to the technical field of maintenance service management, and in particular to a mobile mechanical equipment fault maintenance service system based on big data. Background Art

[0002] The field of maintenance service management technology encompasses equipment maintenance, fault diagnosis, and maintenance process optimization. Its core focus is ensuring stable equipment operation and reducing downtime risks through technical means such as condition monitoring, fault prediction, and resource scheduling. This area systematically integrates sensor data acquisition, anomaly detection algorithms, and maintenance task allocation strategies. It leverages IoT devices to enable remote monitoring and real-time data transmission, optimizes equipment maintenance cycles through data analysis, and adjusts maintenance plans based on historical fault data and real-time operational feedback, forming a closed-loop management system from data acquisition to decision-making and execution.

[0003] The big data-based mobile mechanical equipment fault repair service system refers to a technical solution for fault feature identification, repair resource matching, and service route planning for rotating and transmission mechanical devices. This patent covers technical matters such as the integration of multi-source sensor data streams, detection of abnormal patterns in time series data, and synchronization of repair personnel and equipment locations. Specifically, it uses a distributed computing framework to process highly concurrent vibration and temperature signals, employs a sliding window mechanism to extract time and frequency domain features, constructs a mapping relationship library between fault types and repair solutions, and combines real-time positioning data with traffic condition information to generate dynamic dispatch instructions to match repair resources with fault events.

[0004] Existing technologies have limitations in the fusion of multi-source heterogeneous data. They fail to effectively identify the cross-correlations between parameters such as speed, load, and temperature. This leads to the neglect of synergistic effects between parameters when constructing fault signature libraries, which can easily lead to misjudgments or missed detections. Anomaly detection algorithms rely on preset thresholds and fixed time windows, lacking dynamic adjustment capabilities and adapting to real-time fluctuations in equipment operating conditions. This can cause delayed maintenance instructions or redundant scheduling. Resource matching mechanisms focus on single geographic location information and fail to simultaneously integrate the severity of equipment failures, the effectiveness of maintenance resources, and the impact range of associated equipment. This results in maintenance route planning that is localized rather than globally optimal. The integration of historical data and real-time operating status analysis is insufficient, and maintenance cycle prediction models fail to fully consider the fault transmission effects between equipment groups, potentially triggering cascading downtime incidents. Maintenance resource scheduling systems are not deeply embedded in real-time traffic data streams, making it impossible to accurately calculate the match between arrival times and remaining available time, impacting the accuracy of emergency response strategies. These defects lead to problems such as high resource mismatch rates, delayed fault handling, and rising operation and maintenance costs in traditional systems in complex scenarios. For example, when rotating machinery suddenly suffers from complex faults, the critical maintenance window may be delayed due to the failure to timely identify multi-parameter coupling anomalies. Summary of the Invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a mobile mechanical equipment fault maintenance service system based on big data.

[0006] In order to achieve the above objectives, the present invention adopts the following technical solutions: A mobile mechanical equipment fault repair service system based on big data includes:

[0007] The state recognition module collects equipment speed, load, and temperature data, calculates the speed change amplitude, load fluctuation amplitude, and temperature extremes, analyzes the linear correlation between the value change and the number of faults, selects the correlation values ​​that exceed the set standard, and generates a set of highly correlated values;

[0008] The migration judgment module extracts the value range in the migration record based on the highly correlated value set, calculates the difference between the current value and the median value, performs conversion, determines whether the result is within the set range, and generates a migration start identifier set;

[0009] The offset correction module compares the current temperature difference with the standard temperature based on the migration start identifier set, converts the adjustment amplitude, and applies it to the load fluctuation calculation process to generate a corrected matching value;

[0010] The task aggregation module extracts the location and fault degree of the maintenance equipment and peripheral equipment based on the modified matching value, calculates the distance and connection strength, screens the combinations that meet the conditions, and adjusts the time data to generate a superposition time limit list;

[0011] The path sorting module calculates the estimated arrival time of the vehicle to the maintenance equipment according to the superimposed time limit list, compares the remaining available time, determines whether it is close to the time limit, adjusts the departure order, and generates an urgent response sorting value.

[0012] As a further solution of the present invention, the high-correlation numerical value set includes speed change amplitude correlation values, load fluctuation amplitude correlation values ​​and temperature extreme difference correlation values, the migration start identification set includes executable status mark, numerical interval judgment result, the correction matching value includes temperature difference adjustment amplitude, load fluctuation correction value, the superposition time limit list includes equipment combination screening results, time adjustment parameters, and the urgent response ranking value includes priority order adjustment results and urgent status judgment results.

[0013] As a further solution of the present invention, the state recognition module includes:

[0014] The data calculation submodule collects equipment speed, load, and temperature data, calculates the speed change amplitude at adjacent time points, the load fluctuation amplitude at adjacent time points, and the temperature extreme difference, and integrates the calculation results to generate a change feature set;

[0015] The correlation analysis submodule calls the change feature set to obtain the number of equipment failures in the same time period, calculates the linear correlation between the speed change amplitude, load fluctuation amplitude, temperature extreme difference and the number of failures, and generates a correlation strength set;

[0016] The screening processing submodule calls the correlation strength set, compares the speed change amplitude correlation value, load fluctuation amplitude correlation value, and temperature extreme difference correlation value with the preset judgment criteria, retains the original features corresponding to the correlation values ​​that exceed the judgment criteria, and integrates the retained items to generate a high correlation value set.

[0017] As a further solution of the present invention, the migration judgment module includes:

[0018] The value extraction submodule extracts the value range in the migration record based on the highly correlated value set, obtains the upper and lower limits of the values ​​of the corresponding fields in the record, calculates the middle value interval according to the upper and lower limits, forms a value median sequence, and generates a value median sequence value;

[0019] The distance conversion submodule calls the median sequence value of the numerical value, calculates the difference with the current numerical value, performs proportional conversion based on the difference and the median value, and generates a numerical offset proportional value;

[0020] The interval determination submodule determines whether the numerical value is within the set upper limit interval and the lower limit interval according to the numerical value offset ratio value. If the interval boundary conditions are met, the current numerical state is marked as executable and a migration start identifier set is obtained.

[0021] As a further solution of the present invention, the specific calculation formula for calculating the intermediate value interval according to the upper and lower limits is:

[0022]

[0023] Among them, U i Represents the upper limit of the value of the field in the i-th migration record, L i Represents the lower limit of the value of the field in the i-th migration record, S i represents the fluctuation weight coefficient of the field in the i-th migration record before and after migration, D ij Represents the value of this field at the jth sampling in the i-th migration record. represents the average value of all sampled values ​​of the field in the i-th migration record, n represents the number of samples of the field in the i-th migration record, N i The normalized number of segments representing the difference between the upper and lower bounds associated with the field in the i-th migration record.

[0024] As a further solution of the present invention, the offset correction module includes:

[0025] The identification reading submodule extracts the current temperature value and the standard temperature value in the corresponding state based on the migration start identification set, obtains the numerical difference and records it as initial difference data, and generates a temperature difference ratio sequence;

[0026] The amplitude calculation submodule obtains the difference amplitude between the difference item and the reference value according to the temperature difference ratio sequence and compares it with the adjustment amplitude reference value, and converts the difference amplitude proportionally in combination with the adjustment amplitude trimming ratio to obtain the adjustment applicable amplitude value;

[0027] The load matching submodule calls the adjustment applicable amplitude value and converts the segment ratio with the current load fluctuation value, forms a grouping result according to the load interval and the adjustment amplitude, filters the matching segment position based on the ratio grouping order, and obtains the corrected matching value.

[0028] As a further solution of the present invention, the specific calculation formula for forming the grouping result according to the load interval and the adjustment range is:

[0029]

[0030] Among them, Δ i,j represents the segment adjustment characteristic value corresponding to the combination of load interval i and adjustment amplitude j, α i,k Represents the amplitude adjustment factor of the kth sample data in the i-th load segment, δ k,j represents the load deviation value of the kth sampling data under the current adjustment range j, β k Represents the original load value of the kth sampling data, μ j represents the load average of all sampling points under the current adjustment amplitude j, γ j Represents the standard deviation of the load data of the sampling points in the section corresponding to the current adjustment amplitude j, i,j represents the expected matching value corresponding to the historical combination of the i-th load interval and the j-th adjustment amplitude, and z represents the total number of sampling points in the i-th load segment.

[0031] As a further solution of the present invention, the task aggregation module includes:

[0032] The position extraction submodule obtains the corrected matching value, collects the spatial position information and corresponding fault severity values ​​of the maintenance equipment and surrounding equipment, calculates the spatial distance value between the maintenance equipment and surrounding equipment based on the position coordinates, and uses the fault severity value to construct a device association information table to generate a device spatial distance set;

[0033] The strength screening submodule performs numerical mapping based on the device spatial distance set and the distance value and fault severity value in the device association information table, calculates the correlation strength between the distance value and the fault severity value of each device group, and performs conditional judgment on the correlation strength and distance value of all device combinations, screening device combinations that meet the strength threshold and distance limit range to obtain a target device combination list;

[0034] The time limit construction submodule calls the target equipment combination list, extracts the original maintenance time value of the equipment and compares it in sequence, reorganizes the time for overlapping or adjacent time periods in all combinations, and superimposes the task execution time limit of the construction combination based on the time reorganization result to obtain the superimposed time limit list.

[0035] As a further solution of the present invention, the specific calculation formula for calculating the correlation strength between the distance value and the fault severity value between each group of devices is:

[0036]

[0037] Among them, R ij Represents the strength of the association between device i and device j, Q ij Represents the spatial distance between device i and device j, F ik represents the fault severity value of device i under fault feature k dimension, F jk represents the fault severity value of device j under fault feature k dimension, W k represents the weight coefficient of fault feature k, m represents the total number of dimensions of the fault feature, i and j represent any pair of device numbers, and k is the subscript index of the fault feature dimension.

[0038] As a further solution of the present invention, the path sorting module includes:

[0039] The time limit extraction submodule obtains the path information of the vehicle to the maintenance equipment in the superimposed time limit list, combines the path segment distance value and the passing speed value to determine the estimated arrival time of the vehicle, and generates an estimated time value;

[0040] The urgency comparison submodule identifies whether the current state meets the response requirements based on the estimated time value and the remaining available time information of the maintenance equipment, determines the urgency of the path in the time dimension, and obtains a response urgency value;

[0041] The priority adjustment submodule identifies the response priority of the path according to the response urgency value, rearranges the path sorting order and establishes a sorting output result, and obtains an urgent response sorting value.

[0042] Compared with the prior art, the advantages and positive effects of the present invention are:

[0043] In the present invention, the linear correlation parameters of speed, load, temperature and number of faults are screened to improve the accuracy of fault feature mapping, and the adaptability of migration decisions is enhanced through numerical distance conversion and interval dynamic judgment. The temperature difference and load correction coefficient are integrated to optimize the working condition compatibility and anti-interference ability of load calculation. The equipment location, fault degree and correlation strength are coordinated to generate the optimal maintenance combination under time and space constraints, enhance the efficiency of task aggregation, combine traffic data and the urgency of remaining time classification, and dynamically adjust the priority. The data-driven feedback mechanism runs through the entire chain from feature extraction to decision execution, forming a closed-loop optimization, and systematically improving the efficiency of fault handling and the accuracy of resource allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 is a system flow chart of the present invention;

[0045] Figure 2 It is a submodule flow chart of the present invention. DETAILED DESCRIPTION

[0046] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0047] In the description of the present invention, it should be understood that the terms "length," "width," "up," "down," "front," "back," "left," "right," "vertical," "horizontal," "top," "bottom," "inside," "outside," and the like, indicating positions or relationships, are based on the positions or relationships shown in the accompanying drawings and are intended only to facilitate the description of the present invention and simplify the description. They do not indicate or imply that the devices or elements referred to must have a specific orientation, be constructed, or operate in a specific orientation. Therefore, they should not be construed as limiting the present invention. Furthermore, in the description of the present invention, "plurality" means two or more, unless otherwise expressly and specifically defined.

[0048] See also Figure 1 The mobile machinery equipment fault maintenance service system based on big data includes:

[0049] The state recognition module collects the speed, load, and temperature data of the equipment, calculates the speed change amplitude, load fluctuation amplitude, and temperature extremes, compares the number of faults in the same time period, and calculates the linear correlation between the value change and the number of faults. It retains the values ​​whose correlation exceeds the set judgment standard and generates a set of highly correlated values.

[0050] The migration judgment module extracts the value range in the migration record based on the set of highly correlated values, calculates the distance between the current value and the middle value, performs conversion, and determines whether the conversion results are within the set range. If they are, they are marked as executable and a migration start flag set is generated;

[0051] The offset correction module compares the current temperature difference with the standard temperature based on the migration start flag set, converts it into an adjustment amplitude, and applies it to the load fluctuation calculation process to generate a corrected matching value;

[0052] The task aggregation module obtains the location and fault severity of the maintenance equipment and surrounding equipment based on the corrected matching value, calculates the distance and connection strength, selects equipment combinations that meet the distance and connection strength requirements, performs time adjustment on each group of equipment, and generates a superposition time limit list;

[0053] The path sorting module calls the superimposed time limit list, calculates the estimated arrival time of the vehicle to the maintenance equipment, compares it with the remaining available time, determines whether it is in an urgent state, adjusts the priority order, and generates an urgent response sorting value.

[0054] The highly correlated numerical value set includes the speed change amplitude correlation value, the load fluctuation amplitude correlation value and the temperature extreme difference correlation value. The migration start identification set includes the executable state mark and the numerical range judgment result. The correction matching value includes the temperature difference adjustment amplitude and the load fluctuation correction value. The superposition time limit list includes the equipment combination screening results and the time adjustment parameters. The urgent response sorting value includes the priority order adjustment result and the urgent state judgment result.

[0055] The state recognition module includes:

[0056] The data calculation submodule collects equipment speed, load, and temperature data, calculates the speed change amplitude at adjacent time points, the load fluctuation amplitude at adjacent time points, and the temperature extreme difference, and integrates the calculation results to generate a change feature set;

[0057] After collecting equipment speed, load, and temperature data, the data calculation submodule places a speed sensor near the equipment's rotating shaft, a load sensor in the force transmission path, and a temperature sensor in a location sensitive to temperature rise. Continuous data sampling is performed, with one data point collected per second. The speed fluctuation amplitude is calculated by recording the speed readings at two consecutive sampling points, then calculating the difference between these two values ​​and taking the absolute value to form a continuous series of fluctuation amplitudes. For example, if the speeds recorded at four sampling points are 1420, 1435, 1390, and 1400 rpm, the fluctuation amplitudes are 15, 45, and 10, respectively. Load fluctuation amplitude is processed in the same way as speed fluctuation amplitude: load values ​​at two adjacent time points are obtained, and the difference is taken as the positive value to form a fluctuation series. For example, when the load values ​​are 220, 240, 200, and 210 Newtons, the corresponding fluctuation amplitudes are 20, 40, and 10. Temperature data is processed by finding the highest and lowest values ​​in the temperature record within a certain time period (e.g., 60 seconds) and subtracting them to obtain the temperature range. For example, if the temperature rises from 35.4°C to 42.6°C, the temperature range for that period is 7.2°C. The speed variation sequence, load fluctuation sequence, and temperature range obtained above are integrated to generate a variation feature set.

[0058] The correlation analysis submodule calls the change feature set to obtain the number of equipment failures in the same time period, calculates the linear correlation between the speed change amplitude, load fluctuation amplitude, temperature extreme difference and the number of failures, and generates a correlation strength set;

[0059] After invoking the change feature set, the correlation analysis submodule aligns the various feature data along the time dimension to ensure comparability at the same moment. It then extracts the number of equipment fault records within the corresponding time window by time period and maps each feature to the number of faults within that time period, constructing a matching table. Next, the correlation between the speed variation, load fluctuation, and temperature extremes and the number of faults is calculated. By comparing the trends between the feature values ​​and the corresponding number of faults, it observes whether the values ​​show a consistent upward or downward trend, or whether the number of faults increases during periods of significant feature fluctuation. For example, within a 10-second time window, the speed variation recorded is 10, 12, 8, 9, 11, 7, 13, 14, 6, and 9, respectively, and the corresponding number of faults is 1, 1, 0, 0, 1, 0, 1, 1, 0, 0. If the overall observation shows that a greater speed variation is associated with a greater number of faults, a positive correlation is considered. Finally, correlation indices are derived for speed and faults, load and faults, and temperature extremes and faults, and these are integrated into a set of correlation strength values.

[0060] The screening processing submodule calls the correlation strength set, compares the speed change amplitude correlation value, load fluctuation amplitude correlation value, and temperature extreme difference correlation value with the preset judgment criteria, retains the original features corresponding to the correlation values ​​that exceed the judgment criteria, and integrates the retained items to generate a high correlation value set;

[0061] After calling the correlation strength value generated in the previous step, the screening processing submodule needs to compare the correlation value of each feature item with a pre-set judgment benchmark. This benchmark is usually derived from the equipment's historical data patterns or industry experience, such as 0.6. The correlation value of each feature is compared with the benchmark value one by one. If the value is higher than the benchmark value, the feature is judged to be correlated and retained; otherwise, it is considered that the feature has no significant indicative significance for the fault and is discarded. Taking a certain processing process as an example, if the correlation value between speed and fault is 0.68, the load is 0.41, and the temperature range is 0.73, under this benchmark, only the characteristic values ​​corresponding to speed and temperature are retained, and the load-related features are eliminated. The feature set containing highly correlated items is integrated to obtain a data source for subsequent fault prediction, modeling, or other purposes.

[0062] The migration judgment module includes:

[0063] The value extraction submodule extracts the value range in the migration record based on the highly correlated value set, obtains the upper and lower limits of the corresponding fields in the record, calculates the middle value interval according to the upper and lower limits, forms a median sequence, and generates the median sequence value;

[0064] The specific calculation formula for calculating the middle value interval according to the upper and lower limits is:

[0065]

[0066] Among them, U i Represents the upper limit of the value of the field in the i-th migration record, L i Represents the lower limit of the value of the field in the i-th migration record, S i represents the fluctuation weight coefficient of the field in the i-th migration record before and after migration, D ij Represents the value of this field at the jth sampling in the i-th migration record. represents the average value of all sampled values ​​of the field in the i-th migration record, n represents the number of samples of the field in the i-th migration record, N i The normalized number of segments representing the difference between the upper and lower limits associated with the field in the i-th migration record;

[0067] Parameter description and acquisition method:

[0068] U i By monitoring the historical data of this field, the maximum value is taken as the upper limit.

[0069] L i By monitoring the historical data of this field, the minimum value is taken as the lower limit.

[0070] D ij Actual observation data obtained through regular sampling.

[0071] The calculation formula is:

[0072]

[0073] n is the actual number of samplings.

[0074] S i It is obtained by calculating the coefficient of variation (the ratio of the standard deviation to the mean) of the field value and is used to measure the volatility of the data.

[0075] N i According to business needs, the upper and lower limit differences are divided into several segments for easy analysis.

[0076] Specific numerical settings:

[0077] Assume that the following data is obtained through monitoring and collection:

[0078] U i =150: upper limit of value.

[0079] L i =100: lower limit of value.

[0080] Sampling value D ij (5 samples in total): 110, 115, 120, 125, 130.

[0081] Calculate the average

[0082]

[0083] Calculate the standard deviation σ i :

[0084]

[0085] Calculate the coefficient of variation S i :

[0086]

[0087] According to the upper and lower limit difference U i -L i =50, set the normalized segment number N i =5.

[0088] Substitute the values ​​into the formula to calculate

[0089]

[0090] in,

[0091]

[0092] Continue calculation:

[0093]

[0094] The result shows that the calculated median interval index is 131.75, which reflects the fluctuation of the field value in the sampled data.

[0095] The distance conversion submodule calls the median sequence value of the value, calculates the difference with the current value, and performs proportional conversion based on the difference and the median value to generate the numerical offset proportional value;

[0096] The distance conversion submodule receives the median sequence value generated in the previous stage and reads the field value of the currently processed record. The difference between the two values ​​is used to determine the current record's deviation from the median. For example, in the case of traffic route optimization, if the median value of a field is 18 kilometers and the corresponding field value in the current record is 21 kilometers, the resulting offset is 3 kilometers. By comparing the maximum and minimum values ​​of the field across all samples, assuming the overall range is 10 to 26 kilometers, or a total width of 16 kilometers, the 3-kilometer offset is normalized proportionally to produce a set of scaling factors representing the current value's deviation from the median. To account for fluctuations across different migration scenarios, correction parameters can be set for these scaling factors. These correction parameters can be based on the variability of historical data or the recommended range of a specific model. For example, when data is highly discrete, the scaling influence can be appropriately amplified. This correction yields a final offset ratio value, which serves as the basis for further decision-making.

[0097] The interval determination submodule determines whether the numerical offset ratio value is between the set upper and lower limits respectively. If the interval boundary conditions are met, the current numerical state is marked as executable and the migration start flag set is obtained;

[0098] The interval determination submodule obtains the offset ratio value output by the preceding module and, based on pre-set upper and lower threshold intervals, determines whether the current record meets the migration execution conditions. Taking the scheduling control logic as an example, if the offset ratio value is set between 0.2 and 0.5 as the upper limit determination interval and between -0.5 and -0.2 as the lower limit determination interval, then a record with an offset ratio value of 0.225 is clearly within the upper limit interval and is therefore considered eligible for execution. This determination result can be identified using a Boolean flag and added to the migration identification set. Assume there are five records with offset ratio values ​​of -0.3, 0.4, 0.15, -0.55, and 0.25, respectively. Based on the above interval conditions, only -0.3, 0.4, and 0.25 meet the execution requirements. Their corresponding states can be represented as a binary array [1, 1, 0, 0, 1], which is used for screening and scheduling migration tasks in subsequent processes. The setting of interval thresholds can refer to the distribution of offset values ​​in historical data sets, and the interval boundaries can be finely set in combination with the business's tolerance for fluctuations, so that the judgment criteria are more in line with the actual data status.

[0099] The offset correction module includes:

[0100] The identification reading submodule extracts the current temperature value and the standard temperature value in the corresponding state based on the migration start identification set, obtains the numerical difference and records it as the initial difference data, and generates a temperature difference ratio sequence;

[0101] The reading process first uses the system initialization signal to identify the device's unique identifier. This identifier includes the temperature sensor's number, installation location code, and sensor type information. After identification, a verification operation is performed to verify that the identifier information is consistent with the registered content in the system database. After identity confirmation, the historical startup data set is extracted based on the identifier information. The set records the device's reference temperature standards under different operating conditions. The real-time temperature acquisition module synchronously accesses the instantaneous temperature readings of the current device state, compares the current reading with the reference temperature for the corresponding state, and calculates the difference between the two as difference data. The difference data is sequentially added to the initial difference set in the reading order. Each difference value is then normalized and compared with the corresponding reference temperature, ultimately forming a set of temperature difference ratios. For example, for a sensor numbered TX001, the current temperature is 78.6 degrees Celsius and the historical reference temperature is 72.3 degrees Celsius. The calculated difference is 6.3 degrees Celsius. The normalized ratio is the difference divided by the reference value, resulting in a ratio of approximately 0.0872. This value is recorded as a difference ratio in the sequence. The temperature difference values ​​of all sensors are processed sequentially to form a complete sequence.

[0102] The amplitude calculation submodule obtains the difference amplitude between the difference item and the reference value based on the temperature difference ratio sequence and compares it with the adjustment amplitude reference value, and converts the difference amplitude proportionally in combination with the adjustment amplitude trimming ratio to obtain the adjustment applicable amplitude value;

[0103] Based on the existing temperature difference ratio sequence, processing is performed. First, an amplitude reference value is set based on the historical temperature fluctuation range. This value is usually obtained by comprehensively analyzing the average and median of all ratios and is used as a reference item for each ratio. Then, the difference between each ratio and the reference value is calculated. The obtained difference represents the degree of deviation from the reference standard. Subsequently, an adjustment proportional coefficient is introduced and set as a fixed constant based on the device response characteristics and the empirical value of the adjustment sensitivity. This coefficient acts on the difference, amplifying or reducing it to reflect the response speed of different devices to temperature deviations. Finally, a new set of adaptive amplitude value sequences is obtained. For example, an item with a ratio of 0.0872 has a deviation of 0.0048 compared with the set amplitude reference value of 0.0920. If the adjustment coefficient is 1.2, the converted adaptive amplitude value is 0.0058. All ratio items are processed in the same way to form a complete applicable amplitude data sequence.

[0104] The load matching submodule calls the adjustment applicable amplitude value and converts the segment ratio with the current load fluctuation value. The grouping result is formed according to the load interval and the adjustment amplitude. The matching segment position is selected based on the ratio grouping order to obtain the corrected matching value.

[0105] The specific calculation formula for grouping results based on load range and adjustment range is:

[0106]

[0107] Among them, Δ i,j represents the segment adjustment characteristic value corresponding to the combination of load interval i and adjustment amplitude j, α i,k Represents the amplitude adjustment factor of the kth sample data in the i-th load segment, δ k,j represents the load deviation value of the kth sampling data under the current adjustment range j, β k Represents the original load value of the kth sampling data, μ j represents the load average of all sampling points under the current adjustment amplitude j, γ j Represents the standard deviation of the load data of the sampling points in the section corresponding to the current adjustment amplitude j, i,j represents the expected matching value corresponding to the historical combination of the i-th load interval and the j-th adjustment amplitude, and z represents the total number of sampling points in the i-th load segment;

[0108] Detailed explanation of the formula and calculation process:

[0109] Parameter definition and acquisition method:

[0110] α i,k The factor is determined by performing regression analysis on historical data to evaluate the impact of each sampling point on the overall load.

[0111] δ k,j The load value of the kth sampling point under adjustment amplitude j is monitored in real time and compared with the reference load value.

[0112] β k Raw data obtained by sensors or monitoring equipment.

[0113] μ j The calculation method is:

[0114]

[0115] γ j The calculation method is:

[0116]

[0117] i,j By conducting statistical analysis on historical data, the ideal matching value under this combination is determined.

[0118] z is determined according to the actual sampling settings.

[0119] Parameter value setting and basis:

[0120] α i,k =1.2: Based on historical data analysis, it is found that the k-th sampling data has a higher impact on the overall load, so it is set to 1.2.

[0121] δ k,j =0.05: Through real-time monitoring, the load deviation of the k-th sampling data under the adjustment amplitude j is 0.05.

[0122] β k =100W: The original load value of the kth sampling data directly measured by the sensor is 100W.

[0123] μ j =95W: Under adjustment amplitude j, the average load value of all sampling points is calculated to be 95W.

[0124] γ j =5W: The calculated standard deviation indicates that the dispersion of the load data is 5W.

[0125] i,j =0.1: The expected matching value obtained from historical statistical analysis is 0.1.

[0126] z=50: In actual sampling, there are 50 sampling points in the i-th load section.

[0127] Formula calculation process:

[0128] calculate

[0129]

[0130] calculate

[0131]

[0132] calculate

[0133] 1.2·(0.2236+0.8333)=1.2·1.0569≈1.2683;

[0134] calculate

[0135]

[0136] calculate

[0137] |1.2683-0.002|=1.2663;

[0138] calculate

[0139]

[0140] Final calculation Δ i,j :

[0141] Δ i,j =1.2663;

[0142] The results show that for the i-th load segment and j-th adjustment range combination, the segment adjustment characteristic value is 1.2663. This value is used to measure the matching degree of the current load adjustment. The closer the value is to 0, the better the adjustment effect.

[0143] The task aggregation module includes:

[0144] The position extraction submodule obtains the corrected matching value, collects the spatial position information and corresponding fault severity values ​​of the maintenance equipment and surrounding equipment, calculates the spatial distance value between the maintenance equipment and surrounding equipment based on the position coordinates, and uses the fault severity value to construct a device association information table to generate a device spatial distance set;

[0145] When the position extraction submodule obtains the spatial position information of the maintenance equipment and surrounding equipment, a laser radar, a three-dimensional visual camera or an ultra-wideband positioning device can be deployed on site to collect the coordinate data of each device in the working area. For example, a maintenance device is located at point (2.3, 4.5, 1.2) in the coordinate system, and its adjacent equipment is located at point (5.1, 7.3, 1.2). At the same time, the fault degree value of each device is synchronously collected through the device terminal report or sensor feedback system. Usually, such fault values ​​are set between 0 and 100. The higher the value, the worse the device status. If the fault degree value of the maintenance equipment is 75 and that of the surrounding equipment is 60, the system calculates the straight-line distance between the two points. Confirm the actual spatial interval between the two devices, continue processing all device pairs, and record the spatial distance and the fault value of the corresponding device as associated information items. For example, if the spatial distance corresponding to a certain combination is approximately 3.96 meters, and the devices have fault values ​​of 75 and 60 respectively, the combination will construct entries based on the combination label, spatial distance, and the degree of fault of the two devices. The information is summarized piece by piece to form a complete device association information table. In addition, the system extracts and summarizes the spatial distance data obtained from all device combinations to form an independent spatial distance set. Combined with the maintenance site example, the set content such as {3.96 meters, 5.21 meters, 6.84 meters, ...} can be obtained, and the information table and set generation are finally completed.

[0146] The strength screening submodule performs numerical mapping based on the device spatial distance set and the distance value and fault severity value in the device association information table. It calculates the correlation strength between the distance value and the fault severity value of each device group, and performs conditional judgment on the correlation strength and distance value of all device combinations. It selects device combinations that meet the strength threshold and distance limit range to obtain a target device combination list.

[0147] The specific calculation formula for calculating the correlation strength between the distance value and the fault severity value between each group of devices is:

[0148]

[0149] Among them, R ij Represents the strength of the association between device i and device j, Q ij Represents the spatial distance between device i and device j, F ik represents the fault severity value of device i under fault feature k dimension, F jk represents the fault severity value of device j under fault feature k dimension, W k represents the weight coefficient of fault feature k, m represents the total number of fault feature dimensions, i and j represent any pair of device numbers, and k is the subscript index of the fault feature dimension (the value range is 1 to m);

[0150] Parameter description and acquisition method:

[0151] Q ij : The actual physical distance between device i and device j is directly measured by a rangefinder or laser ranging device, in meters (m).

[0152] F ik : Sensors are used to monitor the performance parameters of equipment i on feature k, such as vibration amplitude, temperature, etc., and these measurements are converted into dimensionless fault severity values ​​through standardized methods, which usually range from 0 to 1.

[0153] F jk : As above, perform the same monitoring and standardization processing on device j to obtain the corresponding fault degree value.

[0154] W k : Using methods such as the analytic hierarchy process (AHP) or entropy weight method, weights are allocated according to the degree of influence of each fault feature on the overall fault to ensure the objectivity and accuracy of the weight.

[0155] m: Determine the number of fault characteristics that need to be monitored by analyzing the possible failure modes of the equipment.

[0156] Specific numerical settings and basis:

[0157] Q ij : The distance between device i and device j measured by the laser rangefinder is 5.0 meters.

[0158] m: Based on the main failure modes of the equipment, three key features are selected for monitoring.

[0159] F ik and F jk : Through sensor measurement and standardization, the following values ​​are obtained:

[0160] Feature 1: F i1 =0.6, F j1 =0.4;

[0161] Feature 2: F i2 =0.7, F j2 =0.5;

[0162] Feature 3: F i3 =0.5, F j3 =0.3;

[0163] Fault feature weight coefficient W k : Using the analytic hierarchy process, we determine the weights of each feature based on expert evaluation and historical data:

[0164] Feature 1: W1 = 0.5;

[0165] Feature 2: W2 = 0.3;

[0166] Feature 3: W3 = 0.2;

[0167] Calculation process:

[0168] Compute the square root of the sum of the squares of the weighted fault severity:

[0169]

[0170] Calculate the sum of the absolute values ​​of the differences between the fault severity values ​​of each characteristic:

[0171]

[0172] Substitute the above results into the original formula for calculation:

[0173]

[0174] Result analysis:

[0175] The calculated correlation strength value R ij ≈0.7000. This result indicates that the correlation strength between device i and device j is 0.7000, reflecting the comprehensive correlation between the two devices in terms of spatial distance and fault characteristics. Based on the preset threshold and distance limit, we can further determine whether the device combination meets the screening criteria and whether it should be included in the target device combination list.

[0176] The time limit construction submodule calls the target equipment combination list, extracts the original maintenance time values ​​of the equipment and compares them in sequence, reorganizes the time for overlapping or adjacent time periods in all combinations, and superimposes the task execution time limits of the constructed combinations based on the time reorganization results to obtain a superimposed time limit list;

[0177] The time limit construction submodule extracts the original maintenance time information of each device in the combination according to the screened target device combination. The time information is recorded as the specific start and end time. If there are multiple devices in the combination and their maintenance time overlaps, for example, the time of one device is from 10 am to 12 noon, and the time of the other device is from 11:30 am to 1 pm, the system will sort out the time intervals and integrate the time periods of these two devices into 10 am to 1 pm, which is the time period for task execution after the combination. If the time periods of some combined devices do not completely overlap but the time interval is within the allowable range, such as the interval does not exceed 60 minutes, the system will also include them. Merge into continuous time periods. For example, if the time of one device is 13:30 to 14:30 and the time of another device is 14:15 to 15:00, then merge them into 13:30 to 15:00. During the time reorganization process, calculations are performed in minutes as the smallest unit to facilitate subsequent task scheduling and resource scheduling control. In this way, the task execution time limits of all target combinations are constructed. After completion, a superimposed time limit list is formed, which lists the final executable time period corresponding to each group of equipment, such as {combination 1: 10:00-13:00, combination 2: 13:30-15:00}, etc., to meet the normative requirements for time limits in the subsequent maintenance plan construction.

[0178] The path sorting module includes:

[0179] The time limit extraction submodule obtains the path information of the vehicle to the maintenance equipment in the superimposed time limit list, combines the path segment distance value and the passing speed value to determine the vehicle's estimated arrival time, and generates an estimated time value;

[0180] During execution, the time limit extraction submodule first retrieves the maintenance vehicle's task record from the superimposed time limit list. This information includes the task's start time, maximum allowed response time, and the current remaining time limit. For example, a vehicle task begins at 08:00, has a maximum allowed response time of 120 minutes, and currently has 80 minutes remaining. The system retrieves the vehicle's route from its current location to the designated maintenance facility through structured data access methods, such as database queries or API calls. This route typically consists of multiple consecutive road segments, each with a distance and speed limit. This route segment information can be provided by a geographic information system or a pre-built traffic map library. For example, a route consists of three segments: the first is 1.5 kilometers long with a speed limit of 40 kilometers per hour; the second is 2.2 kilometers long with a speed limit of 30 kilometers per hour; and the third is 1.3 kilometers long with a speed limit of 50 kilometers per hour. For each segment, the system derives an estimated travel time based on the distance and speed limit. This travel time is calculated by calculating the relationship between distance and speed. For example, if the first segment takes nearly two minutes, the second about four minutes, and the third less than two minutes, the total estimated time from the current location to the maintenance facility is calculated by adding up the times for these three segments. This time data is then integrated and output to the subsequent evaluation module. This data is then considered in conjunction with additional information such as the road restrictions, traffic control, and vehicle type restrictions, completing the route estimation process.

[0181] The urgency comparison submodule identifies whether the current state meets the response requirements based on the estimated time value and the remaining available time information of the maintenance equipment, determines the urgency of the path in the time dimension, and obtains the response urgency value;

[0182] During execution, the urgency comparison submodule receives the estimated time data provided by the previous module and compares it with the remaining available time of the maintenance equipment. First, the remaining available time of the equipment task is extracted from the system scheduling record. For example, if the task is created at 08:00 and the maximum response time of the equipment is set to 15 minutes, when the current system time is 08:07, the remaining available time is 8 minutes. The system then compares the extracted estimated time of the path. If the estimated time is approximately 8 minutes and 20 seconds, the system determines whether the path meets the time response requirement by comparing the two values. If the estimated time exceeds the remaining time, it is considered to have timed out; otherwise, it is considered to have met the requirement. Based on this, the system establishes a path urgency classification strategy, setting the urgency level as slightly higher if the estimated time exceeds the remaining time, and as moderate or high if it exceeds a certain percentage. The urgency level is classified using thresholds: an excess of no more than 10% is considered mild, 20%-50% is considered moderate, and over 50% is considered high. For example, if the estimated time is 8.2 minutes and the remaining time is 8 minutes, and the excess is less than 5%, it is classified as slightly urgent and assigned the corresponding urgency level. This level is then used as one of the references for path prioritization and enters the priority adjustment module to perform path reordering.

[0183] The priority adjustment submodule identifies the response priority of the path according to the response urgency value, rearranges the path sorting order and establishes the sorting output result to obtain the urgency response sorting value;

[0184] The priority adjustment submodule assigns a priority level to each path based on the identified response urgency value and reorders the paths. The system first establishes a mapping rule between urgency and priority levels. For example, no urgency corresponds to the lowest priority, mild urgency corresponds to the next highest, and so on, up to high urgency, which corresponds to the highest priority. After the priority levels are assigned, the system ranks all candidate paths. This ranking process begins with a primary ranking based on priority level. If the priority levels are the same, a secondary ranking is performed based on the total estimated duration of the paths, with the path with the shorter duration being prioritized. Furthermore, auxiliary ranking factors such as vehicle resource status, current task allocation, and predicted congestion on each path can be incorporated to further fine-tune the ranking results. For example, if paths A and B both have a mild urgency level, but path A has an estimated duration of 8 minutes and path B has a duration of 9 minutes, path A will be ranked ahead of path B. After the ranking results are generated, key information such as path ID, urgency level, estimated duration, and vehicle code is recorded in a ranking output structure table. This table serves as the basis for subsequent scheduling decisions, ensuring that task paths are appropriately allocated according to their current urgency.

[0185] The above are merely preferred embodiments of the present invention and do not limit the present invention in any other form. Any technician familiar with the profession may use the technical content disclosed above to change or modify it into an equivalent embodiment with equivalent changes and apply it to other fields. However, any simple modification, equivalent change and modification made to the above embodiment based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A mobile machinery equipment fault repair service system based on big data, characterized by: The system comprises: The state recognition module collects equipment speed, load, and temperature data, calculates the speed change amplitude, load fluctuation amplitude, and temperature extremes, analyzes the linear correlation between the value changes and the number of faults, selects the correlation values ​​that exceed the set standard, and generates a set of highly correlated values; The migration judgment module extracts the value range in the migration record based on the highly correlated value set, calculates the difference between the current value and the median value, performs conversion, determines whether the result is within the set range, and generates a migration start identifier set; The offset correction module compares the current temperature difference with the standard temperature based on the migration start identifier set, converts the adjustment amplitude, and applies it to the load fluctuation calculation process to generate a corrected matching value; The task aggregation module extracts the location and fault degree of the maintenance equipment and peripheral equipment based on the modified matching value, calculates the distance and connection strength, screens the combinations that meet the conditions, and adjusts the time data to generate a superposition time limit list; The path sorting module calculates the estimated arrival time of the vehicle to the maintenance equipment according to the superimposed time limit list, compares the remaining available time, determines whether it is close to the time limit, adjusts the departure order, and generates an urgent response sorting value.

2. The mobile machinery equipment fault repair service system based on big data according to claim 1 is characterized by: The highly correlated numerical value set includes speed change amplitude correlation values, load fluctuation amplitude correlation values ​​and temperature extreme difference correlation values; the migration start identification set includes executable status mark and numerical interval judgment result; the corrected matching value includes temperature difference adjustment amplitude and load fluctuation correction value; the superposition time limit list includes equipment combination screening results and time adjustment parameters; the urgent response ranking value includes priority order adjustment results and urgent status judgment results.

3. The mobile machinery equipment fault repair service system based on big data according to claim 1, characterized in that: The state recognition module includes: The data calculation submodule collects equipment speed, load, and temperature data, calculates the speed change amplitude at adjacent time points, the load fluctuation amplitude at adjacent time points, and the temperature extreme difference, and integrates the calculation results to generate a change feature set; The correlation analysis submodule calls the change feature set to obtain the number of equipment failures in the same time period, calculates the linear correlation between the speed change amplitude, load fluctuation amplitude, temperature extreme difference and the number of failures, and generates a correlation strength set; The screening processing submodule calls the correlation strength set, compares the speed change amplitude correlation value, load fluctuation amplitude correlation value, and temperature extreme difference correlation value with the preset judgment criteria, retains the original features corresponding to the correlation values ​​that exceed the judgment criteria, and integrates the retained items to generate a high correlation value set.

4. The mobile machinery equipment fault repair service system based on big data according to claim 3 is characterized by: The migration judgment module includes: The value extraction submodule extracts the value range in the migration record based on the highly correlated value set, obtains the upper and lower limits of the values ​​of the corresponding fields in the record, calculates the middle value interval according to the upper and lower limits, forms a value median sequence, and generates a value median sequence value; The distance conversion submodule calls the median sequence value of the numerical value, calculates the difference with the current numerical value, performs proportional conversion based on the difference and the median value, and generates a numerical offset proportional value; The interval determination submodule determines whether the numerical value is within the set upper limit interval and the lower limit interval according to the numerical value offset ratio value. If the interval boundary conditions are met, the current numerical state is marked as executable and a migration start identifier set is obtained.

5. The mobile machinery equipment fault repair service system based on big data according to claim 4 is characterized by: The specific calculation formula for calculating the intermediate value interval according to the upper and lower limits is: Among them, U i Represents the upper limit of the value of the field in the i-th migration record, L i Represents the lower limit of the value of the field in the i-th migration record, S i represents the fluctuation weight coefficient of the field in the i-th migration record before and after migration, D ij Represents the value of this field at the jth sampling in the i-th migration record. represents the average value of all sampled values ​​of the field in the i-th migration record, n represents the number of samples of the field in the i-th migration record, N i The normalized number of segments representing the difference between the upper and lower bounds associated with the field in the i-th migration record.

6. The mobile machinery equipment fault repair service system based on big data according to claim 4, characterized in that: The offset correction module includes: The identification reading submodule extracts the current temperature value and the standard temperature value in the corresponding state based on the migration start identification set, obtains the numerical difference and records it as initial difference data, and generates a temperature difference ratio sequence; The amplitude calculation submodule obtains the difference amplitude between the difference item and the reference value according to the temperature difference ratio sequence and compares it with the adjustment amplitude reference value, and converts the difference amplitude proportionally in combination with the adjustment amplitude trimming ratio to obtain the adjustment applicable amplitude value; The load matching submodule calls the adjustment applicable amplitude value and converts the segment ratio with the current load fluctuation value, forms a grouping result according to the load interval and the adjustment amplitude, filters the matching segment position based on the ratio grouping order, and obtains the corrected matching value.

7. The mobile machinery equipment fault repair service system based on big data according to claim 6, characterized in that: The specific calculation formula for forming the grouping result based on the load interval and the adjustment range is: Among them, Δ i,j represents the segment adjustment characteristic value corresponding to the combination of load interval i and adjustment amplitude j, α i,k Represents the amplitude adjustment factor of the kth sample data in the i-th load segment, δ k,j Represents the load deviation value of the kth sampling data under the current adjustment range j, β k Represents the original load value of the kth sampling data, μ j represents the load average of all sampling points under the current adjustment amplitude j, γ j Represents the standard deviation of the load data of the sampling points in the section corresponding to the current adjustment amplitude j, i,j represents the expected matching value corresponding to the historical combination of the i-th load interval and the j-th adjustment amplitude, and z represents the total number of sampling points in the i-th load segment.

8. The mobile machinery equipment fault repair service system based on big data according to claim 6, characterized in that: The task aggregation module includes: The position extraction submodule obtains the corrected matching value, collects the spatial position information and corresponding fault severity values ​​of the maintenance equipment and surrounding equipment, calculates the spatial distance value between the maintenance equipment and surrounding equipment based on the position coordinates, and uses the fault severity value to construct a device association information table to generate a device spatial distance set; The strength screening submodule performs numerical mapping based on the device spatial distance set and the distance value and fault severity value in the device association information table, calculates the correlation strength between the distance value and the fault severity value of each device group, and performs conditional judgment on the correlation strength and distance value of all device combinations, screening device combinations that meet the strength threshold and distance limit range to obtain a target device combination list; The time limit construction submodule calls the target equipment combination list, extracts the original maintenance time value of the equipment and compares it in sequence, reorganizes the time for overlapping or adjacent time periods in all combinations, and superimposes the task execution time limit of the construction combination based on the time reorganization result to obtain the superimposed time limit list.

9. The mobile machinery equipment fault repair service system based on big data according to claim 8, characterized in that: The specific calculation formula for calculating the correlation strength between the distance value and the fault severity value between each group of devices is: Among them, R ij Represents the strength of the association between device i and device j, Q ij Represents the spatial distance between device i and device j, F ik represents the fault severity value of device i under fault feature k dimension, F jk represents the fault severity value of device j under fault feature k dimension, W k represents the weight coefficient of fault feature k, m represents the total number of dimensions of the fault feature, i and j represent any pair of device numbers, and k is the subscript index of the fault feature dimension.

10. The mobile machinery equipment fault repair service system based on big data according to claim 8, characterized in that: The path sorting module includes: The time limit extraction submodule obtains the path information of the vehicle to the maintenance equipment in the superimposed time limit list, combines the path segment distance value and the passing speed value to determine the estimated arrival time of the vehicle, and generates an estimated time value; The urgency comparison submodule identifies whether the current state meets the response requirements based on the estimated time value and the remaining available time information of the maintenance equipment, determines the urgency of the path in the time dimension, and obtains a response urgency value; The priority adjustment submodule identifies the response priority of the path according to the response urgency value, rearranges the path sorting order and establishes a sorting output result, and obtains an urgent response sorting value.

Images