A preventive maintenance scheduling method and system for the health state of a mine card

Abstract

The application provides a preventive maintenance scheduling method and system for the health state of a mine card, relates to the technical field of mine vehicle scheduling and equipment health management, and obtains a whole vehicle health index of the mine card through multi-source sensor data fusion. When the index falls into a warning interval, a task allocation weight attenuation coefficient is calculated based on an index attenuation function, and then an actual task allocation weight is determined in combination with a basic task allocation weight to finally generate a scheduling instruction. The equipment health state can be quantified as a continuously adjustable transport capacity weight, smooth coupling of maintenance decision and production scheduling is realized, and the effective transport capacity output of a mine area is maximized under the premise of ensuring the safety of the mine card.

Description

Technical Field

[0001] This application relates to the field of mine vehicle scheduling and equipment health management technology, and in particular to a preventive maintenance scheduling method and system for mine vehicle health status. Background Technology

[0002] In open-pit mining operations, mining dump trucks (hereinafter referred to as mining trucks) are responsible for transporting ore and overburden, and their operational reliability directly affects the production efficiency of the mining area. Currently, mainstream mines at home and abroad generally adopt the following two types of maintenance and scheduling schemes.

[0003] The first type is the planned maintenance model based on fixed operating time, which mandates maintenance and component replacement according to the maintenance cycles (such as working hours or mileage) specified in the equipment manual, such as the planned maintenance systems recommended by equipment manufacturers like Caterpillar and Komatsu. This model does not take into account the variability in health status of mining trucks from the same batch due to differences in actual operating conditions, and has the drawbacks of "over-maintenance" or "under-maintenance".

[0004] The second type is the state maintenance mode based on a single sensor threshold, which monitors key indicators such as engine exhaust temperature, transmission oil pressure, and brake pad wear. When any indicator exceeds a preset threshold, an alarm or shutdown is triggered. In recent years, some large mines have deployed information systems that link vehicle-mounted data acquisition terminals with dispatch centers, but the early warning logic is still centered on single-point thresholds.

[0005] The existing solutions described above have the following obvious drawbacks.

[0006] First, there is a lack of quantitative characterization of the overall vehicle health status. Existing methods cannot effectively integrate information from multiple heterogeneous sensors, such as those from the powertrain, transmission system, and actuators. The health status of each component is isolated from each other, and a unified, continuous, and comparable evaluation index cannot be formed.

[0007] Second, the maintenance triggering mechanism is completely disconnected from dynamic capacity scheduling. The equipment warning result is merely a binary switch for "whether to repair" and is not quantitatively mapped to the task allocation weight. Mining cards with declining health may still be assigned full-load tasks, accelerating their deterioration.

[0008] Third, the health deterioration process lacks early and smooth intervention. The existing hard threshold alarm only triggers a sudden shutdown when the indicator exceeds the critical value, without implementing gradual load reduction during the transition period from good to ineffective health index, resulting in a high risk of sudden downtime and drastic fluctuations in effective capacity. Summary of the Invention

[0009] The purpose of this application is to provide a preventive maintenance scheduling method and system for mining truck health status, in order to solve the problems of lack of quantitative characterization of vehicle-level health status, separation of maintenance and scheduling, and lack of early smooth intervention in the health deterioration process in the existing technology.

[0010] To address the aforementioned technical problems, in a first aspect, this application provides a preventative maintenance scheduling method based on the health status of mining trucks, comprising: Obtain the overall vehicle health index and basic task allocation weights of the mining card; In response to the vehicle health index being in the warning range, the task allocation weight decay coefficient of the mining truck is calculated based on the exponential decay function; The actual task allocation weight is determined based on the task allocation weight decay coefficient and the basic task allocation weight. Based on the weights assigned to the actual tasks, scheduling instructions are generated.

[0011] In one possible implementation, the exponential decay function is:

[0012] in, The overall vehicle health index, The attenuation coefficient is... The warning threshold; the attenuation coefficient Determined by the following formula:

[0013] in, For mining trucks, the overall vehicle health index Deteriorated to the emergency maintenance threshold Cumulative failure probability during the period .

[0014] In one possible implementation, the warning threshold In response to the determination of the design life and historical maintenance records of key components of mining trucks.

[0015] In one possible implementation, obtaining the overall vehicle health index of the mining truck includes: collecting health data through multi-source sensors deployed on key components of the mining truck; and calculating the overall vehicle health index based on the health data using DS evidence theory.

[0016] In one possible implementation, the method further includes: The mutation layer monitoring system determines that when the instantaneous monitoring value of a critical component exceeds the preset safety range, the overall health index of the mining truck is equal to the emergency maintenance threshold. Global capacity monitoring: When the global effective capacity coefficient is lower than the preset capacity threshold, compensation measures will be implemented.

[0017] In one possible implementation, the attenuation coefficient The warning threshold Hth is dynamically adjusted in response to the slope of the mining area road and the ambient temperature; when the road slope exceeds the preset slope threshold, the threshold is increased. The value of the warning threshold is adjusted upwards when the ambient temperature exceeds the preset range. .

[0018] In one possible implementation, the health data is collected at a preset collection frequency, and the vehicle edge computing unit broadcasts the vehicle health index at a preset broadcast frequency; the scheduling system updates the actual task allocation weight matrix at a preset update cycle.

[0019] The detailed calibration values, complete parameter mapping tables, and numerical calculation examples for typical scenarios of the above coefficients are given in the detailed implementation methods below.

[0020] Secondly, this application provides a preventative maintenance scheduling system for mining truck health status, including: The health index acquisition module is used to obtain the overall vehicle health index and basic task allocation weight of the mining truck. The attenuation coefficient calculation module is used to calculate the task allocation weight attenuation coefficient based on the exponential attenuation function in response to the vehicle health index being in the warning range. The weight determination module is used to determine the actual task allocation weight based on the task allocation weight decay coefficient and the basic task allocation weight. The scheduling instruction generation module is used to generate scheduling instructions based on the weights allocated to the actual tasks. The mutation monitoring module is used to directly determine that the overall health index of the mining truck is equal to the emergency maintenance threshold when the instantaneous monitoring value of a critical component exceeds the preset safety range, and to generate a shutdown dispatch instruction.

[0021] Thirdly, this application provides an electronic device, comprising: Memory, used to store computer programs; A processor, used to execute the computer program to implement the steps of the preventive maintenance scheduling method for mining truck health status as described in the first aspect above.

[0022] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, can implement the steps of the preventive maintenance scheduling method for mining truck health status as described in the first aspect above.

[0023] The preventive maintenance scheduling method for mining trucks provided in this application constructs a vehicle health index. As a unified health measurement benchmark, heterogeneous health information from multiple sources such as engines, transmissions, and hydraulic systems is fused within the DS evidence theory framework, realizing the transformation of mining truck health status from qualitative description to quantitative continuous measurement; based on this, an exponential decay function is used... Within the warning range An internal, continuous, smooth mapping is established between the health index and task allocation weights, allowing the mining truck load to gradually decrease as the health index declines. This exponential decay function is based on a physical model of fatigue cumulative damage to key components of mining trucks under constant stress, matching the nonlinear acceleration trend of failure probability in the warning interval under Miner's linear cumulative damage law, thus avoiding the sudden drop in transport capacity caused by traditional hard threshold logic. Simultaneously, through... and The dynamic correction mechanism incorporates external stress factors such as mine road slope and ambient temperature into the attenuation model, enabling the scheduling strategy to adapt to the environment. Ultimately, the actual task allocation weights are directly injected into the scheduling instruction generation process, achieving a close mathematical correlation between equipment health status and production capacity scheduling. This maximizes the effective output of the mining area while ensuring equipment safety, significantly reducing the probability of sudden downtime and extending the effective service life of mining trucks. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A schematic diagram of the overall process of the preventive maintenance scheduling method for mining truck health status provided in the embodiments of this application; Figure 3 A schematic diagram illustrating the decay of the vehicle health index over operating time and the division of warning intervals, provided in an embodiment of this application; Figure 4 A schematic diagram illustrating the change of the decay function curve with the health index as provided in the embodiments of this application; Figure 2 A schematic diagram illustrating the process of calculating the vehicle health index using multi-source sensor data fusion, as provided in an embodiment of this application. Figure 5 A flowchart illustrating the mutation layer monitoring and global capacity monitoring provided in this application embodiment; Figure 6A schematic diagram illustrating the process of dynamically correcting the attenuation coefficient of the mining area road slope and the early warning threshold of the ambient temperature, as provided in the embodiments of this application. Figure 7 This is a schematic diagram of the structure of a preventive maintenance scheduling system for mining truck health status provided in an embodiment of this application. Detailed Implementation

[0026] During the research and development process, this application discovered a long-standing core contradiction in the field: the scheduling system treats mining trucks as homogeneous transport capacity units, allocating tasks solely based on rated load and location, while ignoring the significant health dispersion of different mining trucks within the same fleet due to differences in service life, maintenance history, and component wear. The root of this contradiction lies in the fact that equipment monitoring and production scheduling are separated into two independent domains, lacking a common mathematical language to convey the crucial information of "how much transport capacity this vehicle can currently handle." To address this, this application uses the overall vehicle health index as a unified metric axis, constructing a continuous transition curve from normal transport capacity to zero capacity in the health index space through an exponential decay function, thereby mathematically mapping state perception and scheduling decisions.

[0027] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments. Obviously, the described embodiments are merely some embodiments of the present application, and not all embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0028] The core of this application is to provide a preventative maintenance scheduling method based on the health status of mining cards. A flowchart of one specific implementation is shown below. Figure 1 As shown, the method includes: S101. Obtain the overall vehicle health index and basic task allocation weight of the mining card.

[0029] This step is the data entry point for the entire scheduling method. Vehicle health index. It is a dimensionless quantitative indicator, with a value range from 0 to 1, where This indicates that the mining card is in brand new factory condition. This indicates that the mining card has completely lost its operational capability. Basic task weighting. This indicates the baseline transport capacity of the mining truck when it is in good working order, and is determined by the truck's rated load capacity and the average travel speed in the mining area.

[0030] Overall vehicle health index The specific methods for obtaining it are as follows.

[0031] like Figure 2As shown, in the first step, the vehicle-mounted edge computing unit performs noise reduction, feature extraction, and normalization processing on the signals and data collected by multi-source sensors deployed on multiple key components of the mining truck.

[0032] Specifically, multi-source sensors are first deployed on several key components of the mining truck. These key components include the engine, transmission, hydraulic system, main reducer, braking system, and tire assembly. Each multi-source sensor collects health data at a preset acquisition frequency: the vibration sensor collects vibration signals at 10 kHz, the temperature sensor collects temperature signals at 1 Hz, the pressure sensor collects pressure signals at 100 Hz, and the oil particle size sensor collects oil cleanliness data at 0.1 Hz.

[0033] The onboard edge computing unit performs local preprocessing on the collected raw data, including noise reduction, feature extraction, and normalization. For vibration signals, the root mean square velocity and peak factor are extracted as feature indicators; for temperature signals, the ratio of the operating average to the rated upper limit is extracted as the temperature ratio feature indicator, where the operating average is calculated using a moving average over the past 5 minutes; for pressure signals, the pulsation fluctuation rate is extracted as a feature indicator; and for oil particle size data, the ISO4406 cleanliness level is extracted as a feature indicator.

[0034] The second step involves converting the multiple feature indicators extracted in the first step into basic probability assignments using a pre-defined membership function.

[0035] Subsequently, the Dempster-Shafer (DS) evidence theory (a decision fusion method for handling uncertain information) was used to fuse the aforementioned multi-source feature indicators. For the first... Key components define the recognition framework. The monitoring data from each multi-source sensor are transformed into basic probability assignments (BPAs) through a pre-defined membership function.

[0036] For example, the membership function can be a triangular membership function, where the three vertices correspond to the typical values ​​of the feature index in the three states of "healthy," "sub-healthy," and "abnormal." For example, when the root mean square velocity of engine vibration is 2.3 mm / s, the basic probability assignment value of this sensor data for the "healthy" state is 0.85, for "sub-healthy" it is 0.12, and for "abnormal" it is 0.03. The basic probability assignments of multiple multi-source sensors for the same component are synthesized using the Dempster combination rule, and the synthesis formula is:

[0037] in, and The basic probability assignment function for two independent sources of evidence. and These represent the state hypotheses supported by two sources of evidence. To support the state assumptions after synthesis, Let represent the empty set. This formula multiplies the basic probability assignments supporting the same state hypothesis from two sources of evidence, sums them, and then divides by . Subtracting the degree of evidence conflict (i.e. the sum of the products of the two evidence sources supporting the mutually exclusive hypothesis) yields the synthesized basic probability assignment value.

[0038] To more clearly illustrate the fusion process of the DS evidence theory, a specific example is given below. Suppose that a vibration sensor and a temperature sensor are deployed on a component of a mining truck engine, and at a certain sampling time, the basic probability distribution values ​​of the two evidence sources are shown in Table 1 below.

[0039] Table 1

[0040] To synthesize a "healthy" state ( For example, the combinations involved in the synthesis include situations where both sources of evidence are "healthy" ( ),but:

[0041] The conflict of evidence (where two sources of evidence support the mutually exclusive hypothesis) is as follows:

[0042] Therefore, the basic probability distribution of the "healthy" state after synthesis is:

[0043] That is, the fusion posterior probability that the component belongs to a "healthy" state. Similarly, the probabilities of "sub-health" and "abnormal" states can be synthesized, and the sum of the three equals 1.

[0044] The third step is to apply evidence combination rules (such as Dempster's combination rules) to iteratively fuse the basic probability allocations of each sensor evidence source under the same key component in pairs, so as to integrate multi-source heterogeneous information and resolve conflicts and uncertainties between evidence.

[0045] For engine components, in addition to vibration and temperature sensors, pressure and oil particle size sensors may also be deployed. When there are multiple (more than two) sources of evidence, the Dempster combination rule is used for pairwise iterative synthesis: first, the BPA of the vibration sensor and the temperature sensor are synthesized into the first intermediate result; then, the first intermediate result is synthesized into the BPA of the pressure sensor into the second intermediate result; finally, the second intermediate result is synthesized into the BPA of the oil particle size sensor into the final fusion probability distribution of the component.

[0046] The fourth step is to obtain the fusion posterior probability of the key component through iterative fusion. This probability is the health probability of the component, which reflects the reliability of its current operating state.

[0047] The above fusion process is performed on each key component to obtain the fusion posterior probability that each component belongs to a "healthy" state.

[0048] The fifth step is to sum the health probabilities of each key component according to their reliability importance to obtain the overall vehicle health index.

[0049] Then, using the reliability importance of each key component as a weight, the health probabilities of each component are weighted and summed to obtain the overall vehicle health index. The calculation formula is:

[0050] in, This represents the total number of critical components. For the first The reliability importance weight of each component satisfies , This represents the fusion posterior probability that the component is in a "healthy" state. For example, the engine has a weight of 0.35, the transmission 0.25, the hydraulic system 0.15, the braking system 0.15, and other components a total of 0.10.

[0051] Let component 1 be the engine, component 2 be the gearbox, component 3 be the hydraulic system, component 4 be the braking system, and component 5 be other components. Expanding the summation symbol, we get:

[0052] in, The results of the DS fusion calculation mentioned earlier are directly quoted, namely:

[0053] The health probabilities of other components (transmission, hydraulic system, braking system, and other components) have not yet been given specific fusion results in the preceding text. To fully demonstrate the term-by-term expansion and calculation process of the weighted summation formula, a set of demonstrative assumptions is temporarily set here (in actual engineering applications, these values ​​need to be independently fused from multi-source sensors of each component using DS evidence theory):

[0054] The reliability importance weights for each critical component are given by the values ​​in the document example:

[0055] Sum all the products to get the overall vehicle health index:

[0056] In this demonstration scenario, the overall health index of the mining truck is approximately 0.736 (ranging from 0 to 1).

[0057] It should be noted that the health probabilities (0.78, 0.82, 0.75, 0.80) for the transmission, hydraulic system, braking system, and other components are merely simulated values ​​added temporarily for the purpose of demonstrating the weighted summation step, and are not the calculated results given earlier in the document. In actual engineering applications, the true posterior probabilities obtained by transforming the multi-source sensor data of these components through membership functions and fusing them using DS evidence theory must be substituted into the above formula to obtain the true overall vehicle health index of the mining truck.

[0058] Basic task weight allocation The determination method is as follows: quantification is performed in ton-kilometers per hour, calculated by multiplying the rated load capacity (tons) of the mining truck by its average travel speed (km / h). For example, the basic task allocation weight for a mining truck with a rated load capacity of 220 tons and an average travel speed of 25 km / h is... Tons per kilometer per hour.

[0059] Step 6: Output the calculated vehicle health index to the on-board display terminal or remote monitoring platform as a quantitative basis for mining truck maintenance decisions and fault warnings.

[0060] S102. In response to the vehicle health index being in the warning range, calculate the task allocation weight decay coefficient of the mining card based on the exponential decay function.

[0061] The logical connection of this step is as follows: In step S101, the overall vehicle health index of each mining truck is obtained. Next, this health state needs to be converted into a numerical value that can be injected into scheduling decisions—a decay coefficient. Since there is a gradual performance degradation transition period between a mining card's healthy state and its faulty state, a binary logic of "either full load or shutdown" should not be used within this period; instead, a continuous and smooth reduction mechanism should be employed. Therefore, this step introduces an exponential decay function, mapping the health index to a decay coefficient between 0 and 1.

[0062] The warning interval is defined as follows: ,in This is the warning threshold (the upper boundary of the warning interval). The emergency maintenance threshold (the lower boundary of the warning interval, when...) Touch or below When, the exponential decay function The output is forced to zero, that is... (Mining card exits operation). For example... Figure 3 As shown, when the mining card is in brand new condition As running time accumulates, Gradually decrease, when cross It enters the warning zone at that time.

[0063] Warning threshold The determination method is based on the design life and historical maintenance records of the key components of the mining truck.

[0064] For example, taking the design overhaul life of a mining truck engine as a benchmark, assuming the design B10 life (i.e., the number of operating hours corresponding to a 10% failure probability) of a certain engine model is 18,000 hours, and combining historical failure data statistics of the same model of mining truck in actual operation in the mining area, a safety margin coefficient of 0.85 is adopted. The safety margin coefficient of 0.85 comes from the statistical analysis results of historical failure data of the same model of mining truck, specifically the reciprocal of the 80th quantile of the distribution of the rate of decline of the health index in the historical failure data, representing a conservative estimate covering 80% of the historical samples. Therefore, for a mining truck with an operating time of 12,000 hours, the warning threshold is... .

[0065] For fleets with complete maintenance records, the Weibull distribution can be used to fit the component failure probability, and the health index value corresponding to the cumulative failure probability reaching 5% can be used as the warning threshold. In another separate example, the warning threshold can also be directly taken. The corresponding emergency maintenance threshold at this time (These values ​​are simplified examples and are not consistent with the aforementioned calculations based on B10 lifetime.) They belong to two independent parameter setting scenarios and are not related to each other.

[0066] Within the warning interval, the task allocation weight decay coefficient is calculated using an exponential decay function. For example... Figure 4 As shown, the expression for the attenuation function is:

[0067] in, As for the current overall vehicle health index, The attenuation coefficient is... This is the warning threshold.

[0068] The dimensional consistency of the formula is explained as follows: Vehicle Health Index and warning threshold Both are dimensionless normalization indices (ranging from 0 to 1), therefore the difference between the two is... It is a dimensionless number. Because... Since it is a dimensionless number, its reciprocal is also a dimensionless number. The independent variable of an exponential function is a dimensionless number, and the output of the entire function is a dimensionless decay coefficient, which satisfies the mathematical rigor requirements of exponential operations.

[0069] From the expression of the decay function, we know that: when (That is, when the mining card health index is exactly at the upper boundary of the warning range), the attenuation coefficient Mining truck capacity will not be reduced in any way; when The continuous decline is approaching At that time, attenuation coefficient Monotonically decreasing from 1. Attenuation coefficient. initial value The calibration method is as follows: Let the mining truck's overall vehicle health index be... Deteriorated to the emergency maintenance threshold During this period, based on historical failure statistics of mining trucks of the same model under the same operating conditions, the average cumulative failure probability within this interval was: Here, "sudden failure" is defined as: an event in which a mining truck completely shuts down due to unplanned reasons and cannot be restored to operation through onboard emergency measures, excluding shutdowns caused by reasons other than the equipment itself, such as driver error, external collisions, or severe weather. Requirements are in place. At that point, the proportion of remaining capacity corresponding to the initial attenuation coefficient is equal to... ,Right now Therefore, the initial calibration value is derived as follows:

[0070] in . It is a dimensionless probability value (ranging from 0 to 1), therefore It is a dimensionless number. It is a dimensionless number. It is a dimensionless number.

[0071] During online scheduling, the on-board edge computing unit calculates the vehicle health index in real time. runtime instantaneous rate of change .when After entering the warning zone, the dispatch system adjusts the current degradation rate accordingly. Real-time adaptive correction is performed, and the correction formula is as follows: .

[0072] in, This represents the average degradation rate of the same type of mining truck under historical operating conditions. Adaptive coefficients This correction mechanism grants mining cards with rapidly declining health indices a larger decay coefficient, allowing for more rapid capacity reduction and ensuring the algorithm can be implemented at any decision-making moment.

[0073] For example, take , , Based on failure statistics of the same model of mining truck during the period when the health index deteriorated from 0.70 to 0.30, the cumulative failure probability is... (i.e., 0.05), then In another scenario with high reliability requirements, take... ,but .

[0074] To explain more clearly Different The variation patterns under different preset reliability requirements are given below. The calibration results are compared in Table 2.

[0075] Table 2

[0076] As can be seen from Table 2, The smaller the value (i.e., the higher the reliability requirement), the better. The smaller the value, the flatter the decay curve, and the more gradual the reduction in mining capacity within the warning range; conversely, the larger the value, the more gradual the reduction. The larger the value, The larger the value, the steeper the decay curve, and the more rapidly the capacity reduction. Mine dispatchers can select the appropriate value based on the importance level of different work faces. Values ​​are taken Calibration.

[0077] S103. Based on the task allocation weight decay coefficient and the basic task allocation weight, determine the actual task allocation weight.

[0078] Step S102 yields an attenuation coefficient between 0 and 1. This coefficient reflects the reduction in capacity of the mining truck's current health state relative to its full health state. To apply this reduction ratio to specific scheduling decisions, this attenuation coefficient needs to be combined with the basic task allocation weights obtained in step S101. Multiply by the product to obtain the actual transport capacity quota that the mine can undertake under its current health condition.

[0079] Actual task weight allocation By assigning weights to basic tasks Multiply by the attenuation coefficient get:

[0080] For example, the basic task allocation weight of a 220-ton mining truck Ton-km / h, current vehicle health index , , ,but:

[0081] ton-kilometers per hour This means the current actual transport capacity of this mining truck is approximately 98.1% of the basic quota, a reduction of about 1.9%. If the health index of this mining truck further deteriorates to... hour:

[0082] ton-kilometers per hour The reduction is approximately 3.8%. When the situation deteriorates to the emergency maintenance threshold... hour:

[0083] The theoretical reduction rate is 5%, which corresponds to a 5% cumulative failure probability under this operating condition.

[0084] To more clearly demonstrate the overall vehicle health index The actual task allocation weight changes under different values, and the mapping relationship under continuous values ​​is given in Table 3 below.

[0085] Table 3

[0086] As can be seen from Table 3, Throughout the entire warning range, from 0.70 to 0.30, the attenuation coefficient monotonically decreases from 1.000 to 0.950, and the actual task allocation weight decreases from 5500 ton-km / h to 5225 ton-km / h. This mapping relationship ensures the continuity and smoothness of capacity reduction, avoiding the problem of a sudden drop in capacity at a certain critical point in traditional hard threshold schemes.

[0087] S104. Based on the weights assigned to the actual tasks, generate scheduling instructions.

[0088] Step S103 yielded the actual task allocation weights for all mining cards globally. These weights then constitute the input parameters for scheduling optimization. The scheduling system needs to translate these weights into executable production instructions, including the loading capacity of each mining truck, the number of transport trips, route assignment, and maintenance of the plant entry sequence.

[0089] The specific method for generating scheduling instructions is as follows: the scheduling system iteratively calculates the actual task allocation weight matrix of the global mining cards at a preset update cycle (e.g., every 5 minutes), and then assigns the weights based on the actual task allocation of each mining card. The numerical proportions are used to allocate transportation tasks. For example, assuming that there are currently 4 mining trucks operating simultaneously in the mining area, and their basic task allocation weight is 5500 tons·km / hour, the current health index and the corresponding actual task allocation weight are shown in Table 4 below.

[0090] Table 4

[0091] Based on this, the dispatching system generates differentiated task assignment schemes: the No. 01 mining truck with a higher health index undertakes 25 fully loaded transport missions, while the No. 04 mining truck with a lower health index undertakes 23 missions (corresponding to an actual weight of 5236 ton-km / hour, which can be allocated 23 fully loaded missions, totaling 5060 ton-km / hour, not exceeding its actual transport capacity limit). The No. 02 mining truck, with 25 fully loaded trips corresponding to a speed of 5500 ton-km / h, slightly exceeds its actual transport capacity limit (5445). In actual execution, the dispatching system either rounds down to 24 trips (5280 ton-km / h) or appropriately reduces the single load to approximately 217.8 tons (5445 / 25) to achieve precise matching. The specific strategy adopted depends on the actual loading equipment accuracy and production scheduling rules of the mine, which is not limited in this application.

[0092] Furthermore, the method also includes: performing mutation layer monitoring. Specifically, when the instantaneous monitoring value of a critical component exceeds a preset safety range, the overall health index of the mining truck is directly determined to be equal to the emergency maintenance threshold, and a shutdown scheduling instruction is generated.

[0093] This step is a mandatory intervention independent of the main scheduling closed loop. Its logical connection lies in: despite the overall vehicle health index... While DS evidence theory fusion calculations can comprehensively reflect the overall health level of mining cards, this fusion process suffers from computational delays and has limited sensitivity to sudden and rapid deterioration events. Therefore, a separate monitoring channel with a faster response time is needed.

[0094] In this step, when the instantaneous monitoring value of a critical component exceeds the preset safety range, a forced intervention logic is triggered, directly determining that the overall health index of the mining truck equals the emergency maintenance threshold. Regardless of the current fusion calculation What are the values? The preset safety range includes an upper threshold and a lower threshold. For example, when the engine lubricating oil pressure momentarily drops below 0.10 MPa (the preset lower safety threshold is 0.15 MPa), or the brake system air pressure momentarily drops below 0.55 MPa (the preset lower safety threshold is 0.60 MPa), or the engine exhaust temperature momentarily exceeds 850°C (the preset upper safety threshold is 800°C), the system will immediately... Set as The dispatch system then generates a shutdown and plant entry instruction. For example... Figure 5 As shown in Branch 1, this mechanism operates independently of the main scheduler in a closed loop and has a higher priority than the fused calculation value.

[0095] Furthermore, the method also includes: performing global capacity monitoring. Specifically, when the global effective capacity coefficient is lower than a preset capacity threshold, compensation measures are implemented.

[0096] The logical connection of this step lies in the fact that, in scenarios where multiple mining trucks operate simultaneously, even if each mining truck is assigned a weight according to its actual task, Even when carrying out missions, the overall transport capacity may still decrease significantly due to multiple mining trucks simultaneously being in the warning range, affecting the overall production efficiency of the mining area. Therefore, it is necessary to monitor the effective transport capacity level from a global perspective and activate the compensation mechanism when necessary.

[0097] In this step, the scheduling system calculates the global effective capacity coefficient in real time. That is, the ratio of the sum of the actual task allocation weights of all mining cards to the sum of the basic task allocation weights. When When the capacity falls below a preset capacity threshold (for example, the preset capacity threshold is 0.80), compensation measures are implemented. The specific methods for determining the compensation measures are shown in Table 5 below.

[0098] Table 5

[0099] For example, assuming there are currently 5 mining trucks operating simultaneously in the mining area, the sum of their basic task allocation weights... Tons per kilometer per hour. The sum of the actual task allocation weights for all mining trucks is calculated. The global effective transport capacity coefficient is calculated as ton-kilometers per hour. If the value is exactly equal to the preset capacity threshold, no compensation measures will be triggered, and the current scheduling strategy will be maintained. If it further decreases to 19,000 tons per hour, then If the value falls within the 0.60 to 0.70 range, the scheduling system will automatically implement a two-level compensation measure of "activating the backup mining card + adjusting the operation priority".

[0100] In practice, one vehicle with a health index will be selected from the backup mining truck queue. The spare mining trucks, with a rated load capacity of 220 tons and a basic task allocation weight of 5500 ton-km / h, were put into operation. Simultaneously, two mining trucks used for stripping operations were temporarily reassigned to the mining line to prioritize the completion of ore transportation tasks. Figure 5 As shown in branch two, the compensation measures are in Gradually exit after restoring to above 0.80.

[0101] Dynamic correction: the attenuation coefficient The warning threshold Hth is dynamically adjusted in response to the slope of the mining area road and the ambient temperature; when the road slope exceeds the preset slope threshold, the threshold is increased. The value of the warning threshold is adjusted upwards when the ambient temperature exceeds the preset range. .

[0102] Within the warning range, the attenuation coefficient λ can be dynamically corrected in response to the slope of the mining area road, and the warning threshold Hth can be dynamically corrected in response to the ambient temperature. For example... Figure 6 As shown, the specific correction method is as follows.

[0103] When the road slope greater than the preset slope threshold (For example, when the slope threshold is 8%), increase The corrected formula is:

[0104] in, For slope correction factor (for example, (per percentage). For example, when hour, .

[0105] When the ambient temperature Exceeding the preset normal working range (for example, to When this happens, the warning threshold will be raised. This allows mining cards to enter load reduction mode earlier. The correction method is as follows: when hour, ; when hour, ; in, Temperature correction factor (for example, (per degree Celsius). For example, when the ambient temperature reaches hour, .

[0106] When road slope correction and ambient temperature correction are triggered simultaneously, they are executed in the order of "temperature correction first, then slope correction," meaning that the slope correction is adjusted upwards first based on the ambient temperature. get , and then The warning range is calculated based on the road slope, and then corrected accordingly. get Finally and Substitute the values ​​into the decay function to calculate the final actual task allocation weights. This order is determined based on field test data—temperature's impact on the health index mainly affects long-term stability (therefore, the baseline threshold is adjusted first), while slope has a more direct impact on the decay rate (therefore, the decay coefficient slope is adjusted later).

[0107] In terms of hardware deployment, each multi-source sensor continuously collects health data at the aforementioned collection frequency and uploads it to the vehicle-mounted edge computing unit. The vehicle-mounted edge computing unit then broadcasts the calculated vehicle health index at a preset broadcast frequency (exemplarily, the broadcast frequency is 1 Hz). The system broadcasts the data to the dispatch center via the mining area's private wireless network. The dispatch system updates the global actual task allocation weight matrix at a preset update cycle (for example, an update cycle of 5 minutes), and regenerates the dispatch instructions for each mining card after each update.

[0108] The preventive maintenance scheduling method for mining trucks provided in this application constructs a vehicle health index. As a unified health measurement benchmark, heterogeneous health information from multiple sources such as engines, transmissions, and hydraulic systems is fused within the DS evidence theory framework, realizing the transformation of mining truck health status from qualitative description to quantitative continuous measurement; based on this, an exponential decay function is used... Within the warning range An internal, continuous, smooth mapping is established between the health index and task allocation weights, allowing the mining truck load to gradually decrease as the health index declines. This exponential decay function is based on a physical model of fatigue cumulative damage to key components of mining trucks under constant stress, matching the nonlinear acceleration trend of failure probability in the warning interval under Miner's linear cumulative damage law, thus avoiding the sudden drop in transport capacity caused by traditional hard threshold logic. Simultaneously, through... and The dynamic correction mechanism incorporates external stress factors such as mine road slope and ambient temperature into the attenuation model, enabling the scheduling strategy to adapt to the environment; ultimately, it assigns weights to actual tasks. The direct injection of scheduling instructions into the process enables deep mathematical coupling between equipment health status and production capacity scheduling, maximizing the effective output of the mining area while ensuring equipment safety.

[0109] like Figure 1 As shown, the dashed line represents the supplementary monitoring step in the preferred embodiment, which... Figure 1 The dashed boxes in the middle are used to distinguish them from the solid boxes in the main scheduling loop. For example... Figure 5 As shown, the relationship between mutation layer monitoring, global capacity monitoring, and the main scheduling closed loop (including steps S101 to S104) is as follows: the main scheduling closed loop runs continuously; mutation layer monitoring obtains the instantaneous monitoring values ​​of each key component from the branch points of the main scheduling closed loop, and immediately intervenes in the current scheduling result once the trigger condition is met; global capacity monitoring obtains the actual task allocation weight of each mining card from the output end of the main scheduling closed loop, evaluates the capacity level in a global dimension, and triggers compensation measures. Both supplement and improve the main scheduling closed loop.

[0110] Figure 7 This is a schematic diagram illustrating a specific implementation of a preventative maintenance scheduling system for mining truck health status, as provided in this application embodiment. (Refer to...) Figure 7 The system may include: The Health Index Acquisition Module (100) is used to acquire the overall vehicle health index and basic task allocation weight of the mining truck. The attenuation coefficient calculation module (200) is used to calculate the task allocation weight attenuation coefficient based on the exponential attenuation function in response to the vehicle health index being in the warning range. The weight determination module (300) is used to determine the actual task allocation weight based on the task allocation weight decay coefficient and the basic task allocation weight; The scheduling instruction generation module (400) is used to generate scheduling instructions based on the weights allocated to the actual tasks.

[0111] The preventive maintenance scheduling system for mining truck health status in this application is used to implement the aforementioned preventive maintenance scheduling method for mining truck health status. Therefore, the specific implementation of the preventive maintenance scheduling system for mining truck health status can be found in the embodiment section of the preventive maintenance scheduling method for mining truck health status above. The specific implementation can be referred to the description of the corresponding embodiments, which will not be repeated here.

[0112] This application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for executing the computer program to implement the steps of any of the above-described preventive maintenance scheduling methods for mining truck health status.

[0113] This application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of any of the above-described preventive maintenance scheduling methods for mining truck health status.

[0114] In one exemplary embodiment, the aforementioned computer-readable storage medium may include, but is not limited to, various media capable of storing computer programs, such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, or optical disks.

[0115] Embodiments of the present invention also provide a computer program product, which includes a computer program that, when executed by a processor, implements the steps in any of the above embodiments of the preventive maintenance scheduling method for mining card health status.

[0116] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0117] The above provides a detailed description of a preventative maintenance scheduling method and system for mining truck health status provided by this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only for the purpose of helping to understand the method and its core ideas. It should be noted that those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of this application.

Claims

1. A preventative maintenance scheduling method based on the health status of mining trucks, characterized in that, include: Obtain the overall vehicle health index and basic task allocation weights of the mining card; In response to the vehicle health index being in the warning range, the task allocation weight decay coefficient of the mining truck is calculated based on the exponential decay function; The actual task allocation weight is determined based on the task allocation weight decay coefficient and the basic task allocation weight. Based on the weights assigned to the actual tasks, a scheduling instruction is generated; When the instantaneous monitoring value of a critical component exceeds the preset safety range, the overall health index of the mining truck is directly determined to be equal to the emergency maintenance threshold, and a shutdown dispatch instruction is generated.

2. The method according to claim 1, characterized in that, The exponential decay function is: ,in, The overall vehicle health index, The attenuation coefficient is... As the warning threshold, the exponential decay function is only used when... Valid within the warning range. This is the threshold for emergency maintenance. When the vehicle health index At that time, the actual task allocation weight is forcibly reset to zero; the attenuation coefficient The initial value is determined by the following formula: ,in, For mining trucks, the overall vehicle health index Deteriorated to the emergency maintenance threshold Cumulative failure probability during the period The attenuation coefficient The initial value is predetermined by this formula, and the attenuation coefficient During online scheduling, the vehicle health index is dynamically updated adaptively based on its real-time change rate. The update formula is as follows: ,in This represents the average degradation rate of the same type of mining truck under historical operating conditions. For preset adaptive coefficients and .

3. The method according to claim 2, characterized in that, The warning threshold In response to the determination of the design life and historical maintenance records of key components of mining trucks.

4. The method according to claim 1, characterized in that, The process of obtaining the overall vehicle health index of the mining truck includes: collecting health data through multi-source sensors deployed on key components of the mining truck; and calculating the overall vehicle health index based on the health data using DS evidence theory.

5. The method according to claim 2, characterized in that, Also includes: When the instantaneous monitoring value of a critical component exceeds the preset safety range, the overall health index of the mining truck is directly determined to be equal to the emergency maintenance threshold. When the global effective capacity coefficient is lower than the preset capacity threshold, compensation measures will be implemented.

6. The method according to claim 2, characterized in that, The attenuation coefficient The warning threshold is dynamically adjusted in response to the slope of the mining area roads. The system dynamically adjusts based on ambient temperature; when the road slope exceeds a preset slope threshold, the value is increased; when the ambient temperature exceeds a preset range, the warning threshold is raised. When both are triggered simultaneously, the correction order is: first, ambient temperature correction, then road slope correction; the corrected warning threshold. With emergency maintenance threshold The following conditions still need to be met. ,and The value remains unchanged.

7. The method according to claim 4, characterized in that, The health data is collected at a preset collection frequency, and the vehicle edge computing unit broadcasts the vehicle health index at a preset broadcast frequency.

8. A preventative maintenance scheduling system for mining truck health status, characterized in that, include: The health index acquisition module is used to obtain the overall vehicle health index and basic task allocation weight of the mining truck. The attenuation coefficient calculation module is used to calculate the task allocation weight attenuation coefficient based on the exponential attenuation function in response to the vehicle health index being in the warning range. The weight determination module is used to determine the actual task allocation weight based on the task allocation weight decay coefficient and the basic task allocation weight. The scheduling instruction generation module is used to generate scheduling instructions based on the weights allocated to the actual tasks. The mutation monitoring module is used to directly determine that the overall health index of the mining truck is equal to the emergency maintenance threshold when the instantaneous monitoring value of a critical component exceeds the preset safety range, and to generate a shutdown dispatch instruction.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of a preventative maintenance scheduling method for mining truck health status as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, enables a preventative maintenance scheduling method for mining truck health status as described in any one of claims 1 to 7.

Images