Method and System for Weighting of Operational Quality Indicators for Self-Propelled Overhead Contact Line Maintenance Equipment

Abstract

This invention discloses a method and system for weighting operational quality indicators of catenary self-propelled maintenance equipment, comprising: determining the target maintenance operation type; extracting key quantitative indicators from the catenary acceptance standards as evaluation factors; and determining the deviation levels of each evaluation factor; using an orthogonal experimental method with the pantograph-catenary dynamic operation performance index as the response variable to calculate the basic weight coefficients of each evaluation factor; in the actual operating environment, collecting and calculating actual operation data of catenary maintenance operations in real time, establishing a dynamic weight compensation method based on model-free adaptive control, updating the pseudo-gradient online according to the quality deviation, and calculating the environmental adaptive weight coefficients; the method of this invention solves the problem of scientific allocation of operational quality indicator weights and environmental adaptation for catenary self-propelled maintenance equipment.

Description

Technical Field

[0001] This invention belongs to the field of rail transit catenary operation and maintenance assessment technology, and particularly relates to a method and system for weighting the operation quality indicators of catenary self-propelled operation and maintenance equipment. Background Technology

[0002] The overhead contact line is a core component of the traction power supply system for electrified railways, and its operational status directly affects train safety and operational efficiency. In recent years, intelligent self-propelled maintenance equipment for the overhead contact line based on robotic operations has become a key development direction in the industry. In existing technologies, the operational quality assessment of self-propelled maintenance equipment for the overhead contact line typically relies on acceptance standards to independently determine the pass / fail status of each indicator. This method has the following technical problems:

[0003] (1) Lack of scientific method for multi-index weight allocation: The existing acceptance standard only specifies the qualified range of each index, without giving a method for weight allocation among multiple indicators. When multiple quality indicators are all within the qualified range but each has its own advantages and disadvantages, it is impossible to perform quantitative scoring and ranking, resulting in strong subjectivity and lack of comparability in the evaluation results; (2) Weight allocation does not consider environmental adaptability: The existing evaluation method usually adopts fixed weights. The same operation quality deviation has different degrees of influence on the dynamic performance of the pantograph under different environmental conditions. The fixed weight evaluation model is difficult to accurately reflect the actual operation effect; (3) Lack of complete technical solution of "offline calibration + online adaptation": Although the orthogonal test method can be used for multi-index weight analysis, the weights obtained are static; although the adaptive method has environmental adaptability, it lacks a benchmark weight as a reference.

[0004] Therefore, there is an urgent need for a method and system for allocating the weights of operation quality indicators for self-propelled overhead contact line maintenance equipment that can scientifically allocate the weights of operation quality indicators and has environmental adaptability. Summary of the Invention

[0005] To address the problems existing in the background art, the present invention aims to provide a method for weighting operational quality indicators of self-propelled overhead contact line maintenance equipment, the method comprising the following steps:

[0006] Determine the type of maintenance work. Based on the determined maintenance work type, extract K quantitative indicators affecting the work quality from the overhead contact line acceptance standards corresponding to the work type as key evaluation factors, denoted as Kk. ,in For each evaluation factor Within the acceptable range specified in the overhead contact line acceptance standards, select Each of the following is a set of different deviation levels and is assigned a value: ;

[0007] Based on the number of evaluation factors K and the deviation level corresponding to each evaluation factor, an appropriate orthogonal array is selected, and an orthogonal experiment is carried out based on the orthogonal array. The overhead contact line maintenance operation is performed respectively, and the average value of the response variable Y reflecting the dynamic operating performance of the pantograph-catenary system is obtained for the overhead contact line system during each operation.

[0008] For each evaluation factor Calculate the mean of the measured average values ​​of the response variables at different deviation levels. And calculate the evaluation factors. range ; Calculate the evaluation factors based on the range values ​​of each evaluation factor. Basic weighting coefficient ,in, Representative evaluation factors The mean of the measured average of the response variable at the first deviation level. Representative evaluation factors The mean of the measured average of the response variable at the second deviation level. Represents the Kth evaluation factor The extreme range;

[0009] In a real-world work environment, for each evaluation factor Real-time acquisition and calculation of actual operation data of overhead contact line maintenance at sampling time t. Establish a dynamic weight compensation method based on adaptive optimization:

[0010]

[0011] in, This represents the increment of the weighting coefficient; For the i-th evaluation factor The target data of the acceptance criteria at sampling time t; This represents the pseudo gradient of the i-th evaluation factor at sampling time t;

[0012] Calculate the evaluation factors at sampling time t Adaptive weight coefficients Thus, an adaptive weight set is obtained. , The Kth evaluation factor represents the sampling time t. Adaptive weighting coefficients.

[0013] Furthermore, the method also includes verifying and updating the basic weight coefficients: performing overhead contact line maintenance operations using verification samples, comparing the weighted comprehensive score prediction results with the normalized maintenance operation response variable values, and if the comparison results exceed the response variable value deviation threshold, then redetermining the number of evaluation factors and / or the number of deviation levels corresponding to each evaluation factor, and redesigning the orthogonal experiment until the accuracy of the adaptive weight set meets the requirements.

[0014] Furthermore, the formula for calculating the weighted comprehensive score prediction result is as follows: ,in, Normalized evaluation factors when performing overhead contact line maintenance work using validation samples The deviation level value.

[0015] Furthermore, the pseudo gradient of the i-th evaluation factor at sampling time t. The algorithm is updated online using the following adaptive estimation algorithm:

[0016] ;

[0017] in, and These are the preset learning parameters.

[0018] Furthermore, the deviation levels include at least: a first deviation level representing excellent work quality, a second deviation level representing acceptable work quality, and a third deviation level representing critically acceptable work quality.

[0019] Furthermore, the maintenance operation type is one of the following: dropper installation operation, dropper length adjustment operation, pull-out value adjustment operation, bolt tightening operation, or insulator cleaning operation.

[0020] Furthermore, when the maintenance operation type is dropper installation or dropper length adjustment, the response variable Y is the standard deviation of the pantograph-catenary contact force; when the maintenance operation type is pull-out value adjustment, the response variable Y is the extreme value of the contact force; when the maintenance operation type is bolt tightening, the response variable Y is the contact resistance value; and when the maintenance operation type is insulator cleaning, the response variable Y is the leakage current value.

[0021] A weight allocation system for the operation quality indicators of self-propelled catenary maintenance equipment is provided to implement the weight allocation method for the operation quality indicators of self-propelled catenary maintenance equipment as described above. The system includes:

[0022] The evaluation factor and deviation level determination module is used to determine the maintenance operation type. Based on the determined maintenance operation type, it extracts K quantitative indicators affecting the operation quality from the overhead contact line acceptance standard corresponding to the operation type as key evaluation factors, denoted as... ,in For each evaluation factor Within the acceptable range specified in the overhead contact line acceptance standards, select Each of the following is a set of different deviation levels and is assigned a value: ;

[0023] The overhead contact line maintenance module based on orthogonal experiments is used to select an appropriate orthogonal array based on the number of evaluation factors K and the deviation level of each evaluation factor, conduct orthogonal experiments based on the orthogonal array, perform overhead contact line maintenance operations, and obtain the average measurement value of the response variable Y reflecting the dynamic operating performance of the pantograph-catenary system during each operation.

[0024] The module for obtaining the basic weights of evaluation factors is used to calculate the weights of each evaluation factor. Calculate the mean of the measured average values ​​of the response variables at different deviation levels. And calculate the evaluation factors. range ; Calculate the evaluation factors based on the range values ​​of each evaluation factor. Basic weighting coefficient ,in, Representative evaluation factors The mean of the measured average of the response variable at the first deviation level. Representative evaluation factors The mean of the measured average of the response variable at the second deviation level. Represents the Kth evaluation factor The extreme range;

[0025] The dynamic weight compensation module is used to adjust the weights for each evaluation factor in a real-world working environment. Real-time acquisition and calculation of actual operation data of overhead contact line maintenance at sampling time t. A dynamic weight compensation method based on adaptive optimization is established to obtain the weight coefficient increment. ;

[0026] The adaptive weight acquisition module is used to calculate the evaluation factors at sampling time t. Adaptive weight coefficients .

[0027] A non-transitory computer-readable storage medium storing a computer program that, when executed by a processor, implements the method for weighting operational quality indicators of self-propelled overhead contact line maintenance equipment as described above.

[0028] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the weight allocation method for the operation quality indicators of the overhead contact line self-propelled maintenance equipment as described above.

[0029] The beneficial technical effects of this invention are as follows:

[0030] (1) The present invention uses the dynamic operation performance index of the pantograph-catenary system as the response variable of the orthogonal test, and quantifies the actual impact of the deviation of each quality index on the current collection quality of the pantograph-catenary system through range analysis. Based on this, the basic weight coefficient is calculated, thus avoiding the subjectivity of the traditional expert scoring method.

[0031] (2) The present invention introduces the idea of ​​adaptive control algorithm, which collects the measured values ​​of quality parameters in real time during actual operation. It can estimate the pseudo gradient and dynamically compensate the weights online using only input and output data, which is particularly suitable for operation and maintenance scenarios in unstructured outdoor environments.

[0032] (3) This invention organically integrates the offline calibration of orthogonal experiments with the online compensation of the adaptive mechanism. The orthogonal experiment provides a benchmark weight to ensure stability, and the adaptive mechanism performs dynamic fine-tuning on this basis to ensure environmental adaptability. The two complement each other and work together to solve the inherent limitations of using either method independently.

[0033] (4) The online parameter update of the adaptive mechanism algorithm involves only a small number of multiplication and addition operations, with low computational complexity. It is suitable for deployment in the vehicle-mounted edge computing unit of self-propelled maintenance equipment, realizing millisecond-level dynamic weight response, and has good real-time performance and engineering practicality. Attached Figure Description

[0034] Figure 1 This is a flowchart illustrating the method for allocating the weights of operational quality indicators for self-propelled overhead contact line maintenance equipment, as provided in an embodiment of the present invention. Detailed Implementation

[0035] This invention provides a method and system for weighting operational quality indicators of catenary self-propelled maintenance equipment, comprising: determining the target maintenance operation type; extracting key quantitative indicators from the catenary acceptance standards as evaluation factors; employing an orthogonal experimental method, using the pantograph-catenary dynamic operation performance index as the response variable, and calculating the basic weight coefficients of each factor; in the actual operating environment, collecting and calculating actual operation data in real time, establishing a dynamic weight compensation method based on model-free adaptive control, updating the pseudo-gradient online according to the quality deviation, and calculating the environmental adaptive weight coefficients; the method of this invention solves the problem of scientific allocation and environmental adaptation of operational quality indicator weights for catenary self-propelled maintenance equipment.

[0036] The following, in conjunction with the accompanying drawings, provides a clearer and more complete description of the method and system for allocating the weights of the operation quality indicators for the self-propelled overhead contact line maintenance equipment provided by this invention:

[0037] Example 1

[0038] Figure 1 This is a flowchart illustrating the method for allocating the weights of the operation quality indicators for self-propelled overhead contact line maintenance equipment provided in this embodiment. The method includes the following steps:

[0039] Step (1): Determine the target maintenance operation type; In the actual operation of the self-propelled maintenance equipment of the overhead contact line, select the corresponding maintenance operation type based on the on-site defects. Generally speaking, the maintenance operation type is one of the following: dropper installation operation, dropper length adjustment operation, pull-out value adjustment operation, bolt tightening operation or insulator cleaning operation.

[0040] Step (2): Based on the target maintenance operation type, extract K quantitative indicators affecting the operation quality from the overhead contact line acceptance standard corresponding to that operation type as key evaluation factors, denoted as... ,in ;

[0041] Specifically, from all quality indicators involved in the overhead contact line acceptance standards, through expert screening, statistical analysis of historical operation and maintenance data, or sensitivity analysis, the K key evaluation factors that have the greatest impact on pantograph-catenary operation performance are selected; to balance evaluation accuracy and testing costs, preferably... For each evaluation factor (i=1, 2...K), within the acceptable range specified in the overhead contact line acceptance standard, select... Each of the following is a set of different deviation levels, and values ​​are assigned to them: ;

[0042] As an example, for the dropper installation operation, five quantitative indicators affecting the quality of the dropper installation operation were selected as key evaluation factors, namely:

[0043] Contact wire height deviation (unit: mm);

[0044] Installation position deviation along the line direction (unit: mm);

[0045] : String length deviation (unit: mm);

[0046] : Height difference between the contact lines of two adjacent dropper points (unit: mm);

[0047] : Deviation of suspension tension value (unit: %);

[0048] For each evaluation factor Within the acceptable range specified in the acceptance criteria, select There are 10 different deviation levels, among which The deviation levels include at least: a first deviation level representing excellent operational quality, a second deviation level representing qualified operational quality, and a third deviation level representing critically qualified operational quality. The deviation levels characterize the degree of variation of the evaluation factor within the allowable range; by comparing the magnitude of change in the response variable under different deviation levels, the sensitivity of the evaluation factor to the overhead contact line operation performance is reflected; the greater the magnitude of change in the response variable, the higher the importance of the corresponding evaluation factor, and the larger its weight coefficient; simultaneously, the more deviation levels there are, the more refined the characterization of the influence law of the evaluation factor, but correspondingly, the experimental scale will also increase; therefore, after comprehensive consideration, 3 to 5 deviation levels can be selected.

[0049] In this embodiment, for the five evaluation factors of the dropper installation operation, the deviation level can be three: a first deviation level representing excellent operation quality, a second deviation level representing qualified operation quality, and a third deviation level representing critical qualified operation quality. Values ​​are assigned to these three deviation levels based on expert experience, as shown in Table 1.

[0050] Table 1

[0051]

[0052] Furthermore, to improve the accuracy of weight calculation and to more precisely characterize the influence of evaluation factors, it can be considered that, for the five evaluation factors of the dropper installation operation, there are five deviation levels: the first deviation level representing excellent work quality, the second deviation level representing good work quality, the third deviation level representing medium work quality, the fourth deviation level representing qualified work quality, and the fifth deviation level representing critical qualified work quality, as shown in Table 2.

[0053] Table 2

[0054]

[0055] For other maintenance operation types, such as dropper length adjustment, pull-out value adjustment, bolt tightening, or insulator cleaning, quantitative indicators affecting operation quality are extracted from the contact network acceptance standards corresponding to the operation type as key evaluation factors. For each evaluation factor, within the qualified range specified in the acceptance standards, at least three different deviation levels are selected and assigned values. It should be noted that the deviation levels include at least: a first deviation level representing excellent operation quality, a second deviation level representing qualified operation quality, and a third deviation level representing critically qualified operation quality.

[0056] Step (3): Based on the number of evaluation factors K and the deviation level corresponding to each evaluation factor, select the appropriate orthogonal table and conduct an orthogonal experiment with N sets of experimental combinations; it should be noted that when each evaluation factor has the same number of deviation levels, a standard orthogonal table is used; when the number of deviation levels corresponding to each evaluation factor is different, a mixed-level orthogonal table is used; parameter combinations that fall outside the orthogonal experiment are used as verification samples.

[0057] For example, regarding the number of evaluation factors For maintenance work types where the deviation level number corresponding to each evaluation factor is 3, select... The standard orthogonal array (i.e., 18 trials, 5 evaluation factors, and 3 bias levels);

[0058] Step (4): Perform overhead contact line maintenance operations according to the parameter combinations set by the orthogonal test scheme; for the overhead contact line system during each operation, obtain the average value of the response variable Y reflecting the dynamic operating performance of the pantograph-catenary system; in fact, during each operation, the response variable values ​​reflecting the dynamic operating performance of the pantograph-catenary system can be measured at preset time intervals by using the contact force detection device, acceleration sensor and data acquisition system installed on the train pantograph, and by averaging the multiple sets of parallel measurement values ​​of the obtained response variable Y, the average value of the response variable Y reflecting the dynamic operating performance of the pantograph-catenary system can be obtained.

[0059] Specifically, according to the overhead contact line acceptance standard "TB 10421-2018 Railway Electric Traction Power Supply Engineering Construction Quality Acceptance Standard" corresponding to the type of operation, when the target maintenance operation type is dropper installation or dropper length adjustment, the response variable Y is selected as the standard deviation of pantograph-catenary contact force. This is because the dropper installation position and dropper length directly affect the uniformity of force on the overhead contact line elastic chain suspension structure, which in turn affects the dynamic contact stability between the pantograph and the contact wire. The standard deviation of pantograph-catenary contact force can effectively characterize the degree of fluctuation in pantograph-catenary contact force. The smaller the value, the more uniform the contact force and the more stable the dynamic performance of the pantograph-catenary system. Therefore, the standard deviation of pantograph-catenary contact force is used as the response variable for this type of maintenance operation.

[0060] When the target maintenance operation type is pull-out value adjustment operation, the response variable Y is selected as the contact force extreme value; because pull-out value adjustment will directly affect the lateral contact state between the pantograph slide and the contact wire. When the pull-out value deviation is too large, it is easy to cause local contact impact to be enhanced, thereby generating abnormal contact force peak value; the contact force extreme value can sensitively reflect the degree of pantograph-catenary transient impact, so the contact force extreme value is used as the response variable for this type of maintenance operation.

[0061] When the target maintenance operation type is bolt tightening, the response variable Y is selected as the contact resistance value. This is because the bolt tightening quality directly affects the conductivity of the electrical connection. When the bolt is not tightened enough or there is poor contact, the contact resistance of the connection will increase, thereby affecting the stability of the power supply of the contact network. Contact resistance can directly characterize the quality of electrical connection, so the contact resistance value is used as the response variable for this type of maintenance operation.

[0062] When the target maintenance operation type is insulator cleaning, the response variable Y is selected as the leakage current value. This is because the accumulation of dirt on the surface of the insulator will increase the leakage current and affect the insulation performance of the contact network. After the insulator is cleaned, the degree of dirt on its surface decreases, and the leakage current can directly reflect the insulation restoration effect. Therefore, the leakage current value is used as the response variable for this type of maintenance operation.

[0063] Step (5): For each evaluation factor Calculate the mean of the measured average values ​​of the response variable at different deviation levels. And calculate the evaluation factors. range ; Calculate the evaluation factors based on the range values ​​of each evaluation factor. Basic weighting coefficient ,in, Representative evaluation factors The mean of the measured average of the response variable at the first deviation level. Representative evaluation factors The mean of the measured average of the response variable at the second deviation level. Represents the Kth evaluation factor The extreme range;

[0064] For example, regarding the number of evaluation factors Table 3 shows a partial orthogonal experimental combination table for maintenance operation types where the deviation level number corresponding to each evaluation factor is 3. In Table 3, 1, 2, and 3 represent the first, second, and third deviation levels, respectively. Regarding the evaluation factors... Calculate its performance in all group trials The mean of the average values ​​of the response variables at the first level of deviation. Calculate its performance in all group trials The mean of the average of the response variables measured at the second level of deviation. Calculate its performance in all group trials The mean of the average values ​​of the response variables at the third level of deviation. Then, calculate the evaluation factors. range Based on this range value, calculate the evaluation factors. Basic weighting coefficient ;in, Represents the Kth evaluation factor Extremely poor.

[0065] Table 3

[0066]

[0067] Step (6): In the actual working environment, for each evaluation factor Real-time acquisition and calculation of actual operation data of overhead contact line maintenance at sampling time t. Establish a dynamic weight compensation method based on adaptive optimization:

[0068]

[0069] in, This represents the increment of the weighting coefficient; For the i-th evaluation factor The target data of the acceptance criteria at sampling time t; The pseudo gradient of the i-th evaluation factor at sampling time t is updated online using the following adaptive estimation algorithm:

[0070]

[0071] in, and These are the preset learning parameters;

[0072] Calculate the evaluation factors at sampling time t Adaptive weight coefficients ;

[0073] Thus, an adaptive weight set is obtained. , The Kth evaluation factor represents the sampling time t. Adaptive weighting coefficients;

[0074] It should be noted that, for each evaluation factor Real-time acquisition and calculation of actual operation data of overhead contact line maintenance at sampling time t. The method involves collecting and calculating evaluation factors through manual or automated equipment. Corresponding data that reflects the actual work quality; for example, for dropper installation work, actual work data... The deviation can be the contact wire height deviation, the installation position deviation along the track direction, the dropper length deviation, the difference in contact wire height between two adjacent dropper points, or the dropper tension deviation. It can be measured using one or more of the following: a contact wire inspection trolley, a track inspection trolley, a laser measuring device, a machine vision inspection device, and a tension detection device. Among these, the contact wire height deviation... The actual conductor height of the contact wire can be measured using a laser ranging device mounted on an overhead contact line inspection trolley or track inspection trolley, and compared with the designed conductor height to obtain the conductor height deviation value; the installation position deviation along the track direction. The actual position of the dropper installation point can be obtained through machine vision positioning devices, coded positioning devices, or laser ranging devices, and compared with the designed installation position to obtain the installation position deviation value along the line direction; dropper length deviation. The actual length of the dropper can be measured using a laser length measuring device, electronic length measuring instrument, or robot end-effector measurement module, and compared with the designed length to obtain the dropper length deviation value; the height difference between the contact lines of two adjacent dropper points. The contact wire height at adjacent dropper positions can be continuously measured by the contact wire inspection trolley, and the height difference between the two measuring points can be calculated to obtain the contact wire height difference value between two adjacent dropper points; the dropper tension value deviation. The actual tension value of the dropper can be measured by a tension sensor, electronic force gauge or dropper tension detection device, and compared with the design tension value to obtain the deviation value of the dropper tension value.

[0075] Step (7): Use validation samples to validate and update the basic weight coefficients: Compare the weighted comprehensive score prediction results with the normalized response variable values. If the comparison results exceed the response variable value deviation threshold, return to step (2) to redetermine the number of evaluation factors K and / or the number of deviation levels corresponding to each evaluation factor, and design an orthogonal experiment until the accuracy of the adaptive weight set meets the requirements. When redetermining the number of evaluation factors K and / or the number of deviation levels corresponding to each evaluation factor, when the number of evaluation factors increases, select an orthogonal table with a larger capacity; when the number of redesigned evaluation factors decreases, select an orthogonal table with a matching capacity.

[0076] It should be noted that when using validation samples to update the basic weight coefficients, the method for obtaining the weighted comprehensive score prediction result is as follows:

[0077] For each evaluation factor in the validation sample Given the corresponding deviation level values, the evaluation factors are first normalized. The deviation level value is obtained Then, based on the adaptive weight coefficients obtained in step (6) Calculate the weighted comprehensive score prediction results The adaptive weight system obtained in step (6) is used here because the adaptive weight coefficient has taken into account the dynamic compensation under the actual working environment, and can more accurately reflect the actual impact of each evaluation factor on the response variable. The prediction result calculated by the adaptive weight coefficient is compared with the true value of the response variable, which is a comprehensive verification of the overall performance of this method.

[0078] In addition, in the normalized evaluation factors Obtaining the deviation level numerical value When normalizing the response variable values, the selection of the upper and lower limits of normalization should be based on the "TB 10421-2018 Railway Electric Traction Power Supply Engineering Construction Quality Acceptance Standard".

[0079] Based on different target maintenance operation types, practical work experience, and engineering operation specifications, corresponding deviation thresholds for response variable values ​​can be set. For example, for the response variable being the standard deviation of pantograph-catenary contact force, the deviation threshold is set at 10%–15%; for the response variable being the extreme value of contact force, the deviation threshold is set at 8%–12%; for the response variable being the contact resistance value, the deviation threshold is set at 5%–10%; and for the response variable being the leakage current value, the deviation threshold is set at 10%–20%.

[0080] In addition, when determining the key evaluation factors for other maintenance operation types, from all quality indicators involved in the overhead contact line acceptance standards, through expert screening, historical operation and maintenance data statistical analysis, or sensitivity analysis, the K key evaluation factors with the greatest impact on the performance of the maintenance operation type are selected. For example, when the target maintenance operation type is dropper installation, the key evaluation factors include: contact wire height deviation, installation position deviation along the line direction, dropper length deviation, contact wire height difference between two adjacent dropper points, and dropper tension value deviation; when the target maintenance operation type is pull-out value adjustment, the key evaluation factors include: pull-out value deviation, contact wire height deviation, positioner slope deviation, mid-span offset value deviation, and adjacent dropper height difference; when the target maintenance operation type is bolt tightening, the key evaluation factors include: tightening torque deviation, bolt tightening angle deviation, displacement deviation of the fastened part, and spring washer compression state parameters; when the target maintenance operation type is insulator cleaning, the key evaluation factor includes: insulator surface salt density.

[0081] It should be noted that the operation of comparing the weighted comprehensive score prediction result with the normalized response variable value in step (7) is carried out in the normalized space on both sides, and is comparable.

[0082] Specifically:

[0083] On the right, the response variable value Y is normalized according to the upper and lower limits specified in the acceptance criteria to obtain the normalized response variable value;

[0084] Left side: For each evaluation factor in the verification sample, the deviation level value is normalized according to the acceptable range specified in the acceptance criteria, resulting in... Then multiply and sum the results by the adaptive weighting coefficients to obtain the weighted comprehensive score prediction result. This result is also within the normalization space and is a predictive estimate of the normalized response variable;

[0085] Both sides are dimensionless normalized quantities, and the difference between them is the prediction error. If the error exceeds the preset deviation threshold of the response variable value, it indicates that the current weight system is not good enough to predict the performance of the pantograph-catenary operation, and it is necessary to return to step (2) to redesign the evaluation factors or deviation level.

[0086] Example 2

[0087] This embodiment provides a weight allocation system for the operation quality indicators of self-propelled catenary maintenance equipment, used to implement the weight allocation method for the operation quality indicators of self-propelled catenary maintenance equipment described in Embodiment 1. The system includes:

[0088] The evaluation factor and deviation level determination module is used to determine the maintenance operation type. Based on the determined maintenance operation type, it extracts K quantitative indicators affecting the operation quality from the overhead contact line acceptance standard corresponding to the operation type as key evaluation factors, denoted as... ,in For each evaluation factor Within the acceptable range specified in the overhead contact line acceptance standards, select Each of the following is a set of different deviation levels and is assigned a value: ;

[0089] The overhead contact line maintenance module based on orthogonal experiments is used to select an appropriate orthogonal array based on the number of evaluation factors K and the deviation level of each evaluation factor, conduct orthogonal experiments based on the orthogonal array, perform overhead contact line maintenance operations, and obtain the average measurement value of the response variable Y reflecting the dynamic operating performance of the pantograph-catenary system during each operation.

[0090] The module for obtaining the basic weights of evaluation factors is used to calculate the weights of each evaluation factor. Calculate the mean of the measured average values ​​of the response variables at different deviation levels. And calculate the evaluation factors. range ; Calculate the evaluation factors based on the range values ​​of each evaluation factor. Basic weighting coefficient ,in, Representative evaluation factors The mean of the measured average of the response variable at the first deviation level. Representative evaluation factors The mean of the measured average of the response variable at the second deviation level. Represents the Kth evaluation factor The extreme range;

[0091] The dynamic weight compensation module is used to adjust the weights for each evaluation factor in a real-world working environment. Real-time acquisition and calculation of actual operation data of overhead contact line maintenance at sampling time t. A dynamic weight compensation method based on adaptive optimization is established to obtain the weight coefficient increment. ;

[0092] The adaptive weight acquisition module is used to calculate the evaluation factors at sampling time t. Adaptive weight coefficients .

[0093] Furthermore, the present invention adopts the following technical solution:

[0094] A non-transitory computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for weighting operational quality indicators of self-propelled overhead contact line maintenance equipment.

[0095] Furthermore, the present invention adopts the following technical solution:

[0096] An electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the above-described method for allocating the weights of the operation quality indicators of the self-propelled overhead contact line maintenance equipment.

[0097] From the above description of the embodiments, those skilled in the art will clearly understand that the facilities of the present invention can be implemented using software plus necessary general-purpose hardware platforms. Embodiments of the present invention can be implemented using existing processors, or by dedicated processors used for this or other purposes for suitable systems, or by hardwired systems. Embodiments of the present invention also include non-transitory computer-readable storage media, comprising machine-readable media for carrying or having machine-executable instructions or data structures stored thereon; such machine-readable media can be any available medium accessible by a general-purpose or special-purpose computer or other machine with a processor. For example, such machine-readable media can include RAM, ROM, EPROM, EEPROM, CD-ROM or other optical disc storage, disk storage or other magnetic storage devices, or any other medium that can be used to carry or store the required program code in the form of machine-executable instructions or data structures and is accessible by a general-purpose or special-purpose computer or other machine with a processor. When information is transmitted or provided to a machine via a network or other communication connection (hardwired, wireless, or a combination of hardwired and wireless), that connection is also considered a machine-readable medium.

[0098] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A method for allocating weights of operational quality indicators for self-propelled overhead contact line maintenance equipment, characterized in that: The method includes the following steps: Determine the type of maintenance work. Based on the determined maintenance work type, extract K quantitative indicators affecting the work quality from the overhead contact line acceptance standards corresponding to the work type as key evaluation factors, denoted as Kk. ,in For each evaluation factor Within the acceptable range specified in the overhead contact line acceptance standards, select Each of the following is a set of different deviation levels and is assigned a value: ; Based on the number of evaluation factors K and the deviation level corresponding to each evaluation factor, an appropriate orthogonal array is selected, and an orthogonal experiment is carried out based on the orthogonal array. The overhead contact line maintenance work is carried out respectively, and the average value of the response variable Y reflecting the dynamic operating performance of the pantograph-catenary system is obtained for the overhead contact line system during each operation. For each evaluation factor Calculate the mean of the measured average values ​​of the response variables at different deviation levels. And calculate the evaluation factors. range ; Calculate the evaluation factors based on the range values ​​of each evaluation factor. Basic weighting coefficient ,in, Representative evaluation factors The mean of the measured average of the response variable at the first deviation level. Representative evaluation factors The mean of the measured average of the response variable at the second deviation level. Represents the Kth evaluation factor The extreme range; In a real-world work environment, for each evaluation factor Real-time acquisition and calculation of actual operation data of overhead contact line maintenance at sampling time t. Establish a dynamic weight compensation method based on adaptive optimization: ； in, This represents the increment of the weighting coefficient; For the i-th evaluation factor The target data of the acceptance criteria at sampling time t; This represents the pseudo gradient of the i-th evaluation factor at sampling time t; Calculate the evaluation factors at sampling time t Adaptive weight coefficients Thus, an adaptive weight set is obtained. , The Kth evaluation factor represents the sampling time t. Adaptive weighting coefficients.

2. The method for weighting the operation quality indicators of self-propelled overhead contact line maintenance equipment according to claim 1, characterized in that, The method also includes verifying and updating the basic weight coefficients: performing overhead contact line maintenance work using verification samples, comparing the weighted comprehensive score prediction results with the normalized maintenance work response variable values, and if the comparison results exceed the response variable value deviation threshold, then redetermining the number of evaluation factors and / or the number of deviation levels corresponding to each evaluation factor, and redesigning the orthogonal experiment until the accuracy of the adaptive weight set meets the requirements.

3. The method for allocating the weights of the operation quality indicators for self-propelled overhead contact line maintenance equipment according to claim 2, characterized in that, The formula for calculating the weighted comprehensive score prediction result is as follows: ,in, Normalized evaluation factors when performing overhead contact line maintenance work using validation samples The deviation level value.

4. The method for weighting the operation quality indicators of self-propelled overhead contact line maintenance equipment according to claim 1, characterized in that, The pseudo gradient of the i-th evaluation factor at sampling time t The algorithm is updated online using the following adaptive estimation algorithm: ； in, and These are the preset learning parameters.

5. The method for weighting the operation quality indicators of self-propelled overhead contact line maintenance equipment according to claim 1, characterized in that, The deviation levels include at least: a first deviation level representing excellent work quality, a second deviation level representing acceptable work quality, and a third deviation level representing critically acceptable work quality.

6. The method for weighting the operation quality indicators of self-propelled overhead contact line maintenance equipment according to claim 1, characterized in that, The maintenance operation type is one of the following: dropper installation operation, dropper length adjustment operation, pull-out value adjustment operation, bolt tightening operation, or insulator cleaning operation.

7. The method for weighting the operation quality indicators of self-propelled overhead contact line maintenance equipment according to claim 6, characterized in that, When the maintenance operation type is dropper installation or dropper length adjustment, the response variable Y is the standard deviation of the pantograph-catenary contact force; when the maintenance operation type is pull-out value adjustment, the response variable Y is the extreme value of the contact force; when the maintenance operation type is bolt tightening, the response variable Y is the contact resistance value; when the maintenance operation type is insulator cleaning, the response variable Y is the leakage current value.

8. A weighted allocation system for the operational quality indicators of self-propelled overhead contact line maintenance equipment, used to implement the weighted allocation method for operational quality indicators of self-propelled overhead contact line maintenance equipment as described in any one of claims 1-7, characterized in that, The system includes: The evaluation factor and deviation level determination module is used to determine the maintenance operation type. Based on the determined maintenance operation type, it extracts K quantitative indicators affecting the operation quality from the overhead contact line acceptance standard corresponding to the operation type as key evaluation factors, denoted as... ,in For each evaluation factor Within the acceptable range specified in the overhead contact line acceptance standards, select Each of the following is a set of different deviation levels and is assigned a value: ; The overhead contact line maintenance module based on orthogonal experiments is used to select an appropriate orthogonal array based on the number of evaluation factors K and the deviation level of each evaluation factor, conduct orthogonal experiments based on the orthogonal array, perform overhead contact line maintenance operations, and obtain the average measurement value of the response variable Y reflecting the dynamic operating performance of the pantograph-catenary system during each operation. The module for obtaining the basic weights of evaluation factors is used to calculate the weights of each evaluation factor. Calculate the mean of the measured average values ​​of the response variables at different deviation levels. And calculate the evaluation factors. range ; Calculate the evaluation factors based on the range values ​​of each evaluation factor. Basic weighting coefficient ,in, Representative evaluation factors The mean of the measured average of the response variable at the first deviation level. Representative evaluation factors The mean of the measured average of the response variable at the second deviation level. Represents the Kth evaluation factor The extreme range; The dynamic weight compensation module is used to adjust the weights for each evaluation factor in a real-world working environment. Real-time acquisition and calculation of actual operation data of overhead contact line maintenance at sampling time t. A dynamic weight compensation method based on adaptive optimization is established to obtain the weight coefficient increment. ; The adaptive weight acquisition module is used to calculate the evaluation factors at sampling time t. Adaptive weight coefficients .

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for weighting operational quality indicators of self-propelled overhead contact line maintenance equipment as described in any one of claims 1 to 6.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method for weighting the operation quality indicators of the contact network self-propelled maintenance equipment as described in any one of claims 1 to 6.

Images