Method and device for automatically recommending maintenance guidance, equipment and storage medium
Through the fault graph vector model based on neural network, automatic comparison and recommendation of maintenance guidelines is solved, and the problem of manual maintenance guidance in the existing technology is low efficiency, achieving a more efficient maintenance process.
Patent Information
- Application Number
- CN202411445861.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-16
- Publication Date
- 2025-02-18
AI Technical Summary
In the prior art, engineers need to manually determine the type of fault and find corresponding maintenance guidelines, resulting in low efficiency in obtaining maintenance guidelines and affecting the efficiency of the entire maintenance process.
By collecting the vehicle data of the vehicle to be repaired, based on the fault map vector model trained by the preset neural network, the similarity between the fault map vector and the preset fault map is automatically compared and obtained, the target fault map is determined and the maintenance guidance is automatically recommended.
It improves the efficiency of obtaining maintenance guidance, reduces the dependence on manpower and engineer experience, shortens maintenance time, and improves the efficiency of the entire maintenance process.
Smart Images
Figure CN119477256A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of vehicle fault detection and maintenance, and in particular to a method, device, equipment and storage medium for automatically recommending maintenance instructions. Background Art
[0002] As the country develops further, the number of cars in use increases, and the pressure on car repair factories increases dramatically. In related technologies, engineers need to manually determine the type of fault, and then look up the fault map based on the fault type to obtain the maintenance instructions corresponding to the fault type to assist in car fault repairs. However, this process requires a lot of manpower and engineers' work experience, which consumes a lot of time and energy. For new fault types, new fault maps and corresponding maintenance instructions need to be recommended, and the new fault maps and corresponding maintenance instructions rely on manual transmission at an extremely slow speed. Due to the above problems, the efficiency of obtaining maintenance instructions is greatly affected, and thus the efficiency of the entire maintenance process is affected. In summary, there is an urgent need for a method of automatically and personally recommending maintenance instructions to improve the efficiency of obtaining maintenance instructions, thereby improving the efficiency of the entire maintenance process. Summary of the invention
[0003] The present application provides a method, device, equipment and computer-readable storage medium for automatically recommending maintenance instructions, which can solve the technical problem in the prior art that manual acquisition of maintenance instructions requires a lot of reliance on manpower and engineers' work experience, consumes a lot of time and energy, affects the efficiency of obtaining maintenance instructions, and further affects the efficiency of the entire maintenance process.
[0004] In a first aspect, an embodiment of the present application provides a method for automatically recommending maintenance instructions, the method for automatically recommending maintenance instructions comprising:
[0005] Collect vehicle data of vehicles to be repaired;
[0006] Based on the fault spectrum vector model, a fault spectrum vector is obtained according to the vehicle data of the vehicle to be repaired, wherein the fault spectrum vector model is obtained by training a preset neural network with training data, and the training data includes static data, dynamic data and annotated fault spectrum vectors;
[0007] According to the fault map vector and the preset fault map, maintenance instructions are automatically recommended.
[0008] In combination with the first aspect, in one implementation, automatically recommending maintenance instructions according to the fault map vector and the preset fault map includes:
[0009] Comparing the fault map vector with a plurality of preset fault maps;
[0010] Obtaining the similarity between the fault map vector and each of the preset fault maps to determine a target preset fault map;
[0011] According to the target preset fault map, maintenance instructions are automatically recommended.
[0012] In combination with the first aspect, in one implementation, obtaining the similarity between the fault map vector and each of the preset fault maps to determine a target preset fault map includes:
[0013] A target fault map with the highest similarity is determined from the similarities between the plurality of fault map vectors and the respective preset fault maps.
[0014] In combination with the first aspect, in one implementation, after automatically recommending maintenance instructions according to the fault map vector and the preset fault map, the method further includes:
[0015] Obtaining the similarity between the fault spectrum vector and each of the preset fault spectra;
[0016] The maintenance guide is input into a target server so that the target server generates a maintenance strategy according to the maintenance guide, and the maintenance strategy is used to schedule an idle robot to perform maintenance on the vehicle to be maintained.
[0017] In combination with the first aspect, in one implementation, before obtaining the fault spectrum vector based on the fault spectrum vector and according to the vehicle data of the vehicle to be repaired, the method further includes:
[0018] Importing the acquired data to be trained into the preset neural network for training, the data to be trained includes static data, dynamic data and annotated fault map vectors, wherein the static data includes: tire pressure, wear degree, vehicle appearance, coolant level, fuel filter, fuel pump, fuel injector, oil leakage status of the oil circuit, and the dynamic data includes: maximum speed of the wheel, working conditions of the cooling fan, water tank, water pump, working conditions of the headlights, turn signals, brake lights, and reverse lights;
[0019] Determining whether the preset neural network after training is in a convergence state;
[0020] If the trained preset neural network is in a convergence state, the fault spectrum vector model is obtained, and the fault spectrum vector model includes:
[0021] Where X t is the quantized value corresponding to the vehicle data, I t For X t The corresponding weight coefficient, f() is the activation function, b t is the bias term, Lt In combination with the first aspect, in one implementation, determining whether the preset neural network is in a convergence state includes:
[0022] Obtaining a loss value of the preset neural network after training;
[0023] If the loss value is less than or equal to a preset standard loss value, it is determined that the preset neural network is in a convergence state;
[0024] If the loss value is greater than a preset standard loss value, it is determined that the preset neural network is not in a convergence state, and the preset neural network is trained again until the preset neural network is in a convergence state.
[0025] In combination with the first aspect, in one implementation, determining whether the preset neural network is in a convergence state includes:
[0026] Obtaining the training times of the preset neural network after training;
[0027] If the number of training times is less than a preset standard number of training times, it is determined that the preset neural network is not in a convergence state, and the preset neural network is trained again until the preset neural network is in a convergence state;
[0028] If the number of training times is greater than or equal to a preset standard number of times, it is determined that the preset neural network is in a convergence state.
[0029] In a second aspect, an embodiment of the present application provides a device for automatically recommending maintenance instructions, the device for automatically recommending maintenance instructions comprising:
[0030] A collection module, used to collect the whole vehicle data of the vehicle to be repaired;
[0031] An acquisition module, used for acquiring a fault spectrum vector model based on a preset neural network and the whole vehicle data of the vehicle to be repaired, wherein the fault spectrum vector model is obtained by training the preset neural network with training data, and the whole vehicle data includes static data, dynamic data and annotated fault spectrum vectors;
[0032] The recommendation module is used to automatically recommend maintenance instructions based on the fault map vector model and the preset fault map.
[0033] In a third aspect, an embodiment of the present application provides a device for automatically recommending maintenance instructions, characterized in that the device for automatically recommending maintenance instructions comprises a processor, a memory, and a program for automatically recommending maintenance instructions stored in the memory and executable by the processor, wherein when the program for automatically recommending maintenance instructions is executed by the processor, the steps of the method for automatically recommending maintenance instructions as described above are implemented.
[0034] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, characterized in that a program for automatically recommending maintenance instructions is stored on the computer-readable storage medium, wherein when the program for automatically recommending maintenance instructions is executed by a processor, the steps of the method for automatically recommending maintenance instructions as described above are implemented.
[0035] The beneficial effects brought by the technical solution provided in the embodiments of the present application include:
[0036] By collecting the whole vehicle data of the vehicle to be repaired;
[0037] Based on a preset neural network, according to the whole vehicle data of the vehicle to be repaired, a fault spectrum vector model is obtained, wherein the fault spectrum vector model is obtained by training the preset neural network through training data, and the whole vehicle data includes static data, dynamic data and annotated fault spectrum vectors;
[0038] According to the fault map vector model and the preset fault map, maintenance instructions are automatically recommended, which solves the technical problem in the related technology that manual acquisition of maintenance instructions requires a lot of reliance on manpower and engineers' on-the-job experience, consumes a lot of time and energy, affects the efficiency of obtaining maintenance instructions, and further affects the efficiency of the entire maintenance process. The present application provides a method for automatically and personalizedly recommending maintenance instructions to improve the efficiency of obtaining maintenance instructions, thereby improving the efficiency of the entire maintenance process. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] Figure 1 A flowchart of an embodiment of a method for automatically recommending maintenance instructions for the present application;
[0040] Figure 2 For this application Figure 1 Detailed flow chart of step S30;
[0041] Figure 3 A schematic diagram of functional modules of an embodiment of a device for automatically recommending maintenance instructions according to the present application;
[0042] Figure 4 This is a schematic diagram of the hardware structure of the device for automatically recommending maintenance instructions involved in the embodiment of the present application. DETAILED DESCRIPTION
[0043] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0044] First, some technical terms in the present application are explained to facilitate those skilled in the art to understand the present application.
[0045] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below in conjunction with the accompanying drawings.
[0046] In a first aspect, an embodiment of the present application provides a method for automatically recommending maintenance instructions.
[0047] In one embodiment, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the method for automatically recommending maintenance instructions for this application. Figure 1 As shown, the method for automatically recommending maintenance guidelines includes:
[0048] Step S10: Collecting vehicle data of the vehicle to be repaired;
[0049] Exemplarily, the whole vehicle data is obtained through preset sensors, and the whole vehicle data includes static data and dynamic data. The static data includes: tire pressure, degree of wear, appearance of the whole vehicle, coolant level, fuel filter, fuel pump, fuel injector, and oil leakage status of the oil circuit; the dynamic data includes: maximum speed of the wheel, working condition of the cooling fan, water tank, water pump, working condition of the headlights, turn signals, brake lights, and reverse lights.
[0050] Step S20: Based on the fault spectrum vector model, according to the vehicle data of the vehicle to be repaired, a fault spectrum vector is obtained, wherein the fault spectrum vector model is obtained by training a preset neural network through training data, and the vehicle data includes static data, dynamic data and annotated fault spectrum vectors;
[0051] Exemplarily, the acquired multiple groups of whole vehicle data are input as the data to be trained into the preset neural network for training until the preset neural network is in a convergence state, and then the fault spectrum vector model is output according to the neural network in the convergence state; the whole vehicle data of the vehicle to be repaired is input into the fault spectrum vector model to obtain the fault spectrum vector.
[0052] Specifically, the fault spectrum vector model based on the vehicle data of the vehicle to be repaired, before obtaining the fault spectrum vector, also includes: importing the obtained data to be trained into the preset neural network for training, the data to be trained includes static data, dynamic data and annotated fault spectrum vectors, wherein the static data includes: tire pressure, wear degree of tires, vehicle appearance, coolant level, fuel filter, fuel pump, fuel injector, oil leakage status of the oil circuit, the dynamic data includes: maximum speed of wheels, working conditions of cooling fans, water tanks, water pumps, working conditions of headlights, turn signals, brake lights, and reverse lights; determining whether the preset neural network after training is in a convergence state; if the preset neural network after training is in a convergence state, obtaining the fault spectrum vector model, the fault spectrum vector model includes:
[0053] Where X t is the quantized value corresponding to the vehicle data, I t For X t The corresponding weight coefficient, f() is the activation function, b t is the bias term, L t is the fault spectrum vector.
[0054] For example, the vehicle data is quantified to ensure the degree of wear of the tire. When the tread depth of the tire is between 6 and 8 mm, it is the first-level wear. At this time, the tire performance is good and suitable for daily driving. The quantified value of the tire wear is 1; when the tread depth of the tire is between 4 and 6 mm, it is the second-level wear. At this time, the wet grip begins to be affected and attention should be paid to driving conditions. The quantified value of the tire wear is 2; when the tread depth of the tire is between 1.6 and 4 mm, it is the third-level wear. At this time, the tire performance is significantly reduced and the risk of driving in rainy days increases. The quantified value of the tire wear is 3; when the tread depth of the tire is less than 1.6 mm, it is the fourth-level wear. At this time, the tire must be replaced immediately, otherwise there will be safety hazards. The quantified value of the tire wear is 4; the quantified values of the specific vehicle data are based on the opinions and experience of each maintenance manufacturer, and the quantified values of other vehicle data are similar and will not be repeated one by one.
[0055] The weight coefficient is set according to the focus of each repair shop. For example, repair shop A focuses on repairing engines, and shop B focuses on repairing tires. Then the weight coefficient of repair shop A for engines will be greater than that of repair shop B. Similarly, the weight coefficient of repair shop B for tires will be greater than that of repair shop A.
[0056] The bias item is used to adjust the quantitative values corresponding to each vehicle data according to the customer's maintenance intention. The initial value of the bias item is 0. When the owner has his own special maintenance intention, the value of the bias item can be changed. For example, the quantitative value corresponding to the slight scratch on the vehicle appearance is 3. The customer feels that this fault has little impact on his driving experience and does not want to repair it for the time being. The bias item can be set to -3 to offset the quantitative value corresponding to the vehicle appearance. For another example, the tire wear is level 2 wear, and the corresponding quantitative value is 2. At this time, it begins to affect wet grip and requires attention to driving conditions. The corresponding tire does not need to be replaced, but the owner wants to replace the tire directly. At this time, according to the owner's maintenance intention, the corresponding bias item can be set to 2, and the quantitative value of the object can be adjusted to 4, and the corresponding maintenance strategy becomes replacing the tire.
[0057] Specifically, the determining whether the preset neural network is in a convergence state includes: obtaining the training times of the preset neural network after training; if the training times are less than the preset standard training times, determining that the preset neural network is not in a convergence state, and training the preset neural network again until the preset neural network is in a convergence state; if the training times are greater than or equal to the preset standard times, determining that the preset neural network is in a convergence state;
[0058] Exemplarily, if the number of training times of the preset neural network is 9000, which is less than the preset standard training coefficient of 10000 times, it is determined that the preset neural network is not in a convergence state, and the training is repeated again until it reaches 10000 times so that the preset neural network is in a convergence state; if the number of training times of the preset neural network is 10032, which is greater than the preset standard training coefficient of 10000 times, it is determined that the preset neural network is in a convergence state.
[0059] Specifically, determining whether the preset neural network is in a convergence state includes: obtaining the training times of the preset neural network after training; if the training times are less than the preset standard training times, determining that the preset neural network is not in a convergence state, and training the preset neural network again until the preset neural network is in a convergence state; if the training times are greater than or equal to the preset standard times, determining that the preset neural network is in a convergence state.
[0060] Exemplarily, when the loss value is 0.1, the loss value is less than the preset standard loss value of 0.5, and it is determined that the preset neural network is in a convergence state; when the loss value is 0.6, the loss value is greater than the preset standard loss value of 0.6, and it is determined that the preset neural network is not in a convergence state, and the preset neural network is trained again until the preset neural network is in a convergence state; when the loss value is 0.5, the loss value is equal to the standard loss value of 0.5, and it is determined that the preset neural network is in a convergence state.
[0061] Step S30: automatically recommending maintenance instructions based on the fault map vector model and the preset fault map.
[0062] Exemplarily, multiple fault map vectors are compared with each of the preset fault maps to obtain the similarity between the fault map vector and each preset fault map, and the highest similarity is determined from the similarities between the multiple fault map vectors and each preset fault map. If the highest similarity is higher than the preset standard similarity, maintenance instructions are automatically recommended based on the target fault map. If the highest similarity is lower than the preset standard similarity, the fault map and maintenance instructions are manually supplemented or modified until the highest similarity meets the preset standard, and then maintenance instructions are automatically recommended based on the target fault map.
[0063] In the present embodiment, the whole vehicle data of the vehicle to be repaired is collected; based on a preset neural network, a fault map vector model is obtained according to the whole vehicle data of the vehicle to be repaired, wherein the fault map vector model is obtained by training the preset neural network with training data, and the whole vehicle data includes static data, dynamic data and labeled fault map vectors; according to the fault map vector model and the preset fault map, maintenance instructions are automatically recommended, which solves the technical problem in the related technology that manual acquisition of maintenance instructions requires a lot of reliance on manpower and engineers' on-the-job experience, consumes a lot of time and energy, affects the efficiency of obtaining maintenance instructions, and further affects the efficiency of the entire maintenance process, and provides a method for automatically and personalizedly recommending maintenance instructions to improve the efficiency of obtaining maintenance instructions, thereby improving the efficiency of the entire maintenance process.
[0064] Further, in one embodiment, referring to Figure 2 , Figure 2 For this application Figure 1 Detailed flow chart of step S30 in FIG. Figure 2 As shown, the automatic recommendation of maintenance instructions based on the fault map vector and the preset fault map includes:
[0065] Step S31: comparing the fault map vector with a plurality of preset fault maps;
[0066] Exemplarily, the subscripts corresponding to each data item in the fault map vector are compared with the subscripts corresponding to each data item in the preset fault map.
[0067] Step S32: obtaining the similarity between the fault map vector and each of the preset fault maps to determine a target preset fault map;
[0068] Exemplarily, if the subscript in the fault map vector is the same as the subscript values in the preset fault map, then the two are determined to be similar and given a specific similarity. For example, if the tire wear is level two wear, its corresponding subscript value in the fault map vector is 003, and data with a subscript value of 003 is also found in the preset fault map, then the two are determined to be similar and a specific fitting value of 0.02 is given. The fitting values of multiple vehicle data are added together to obtain the similarity, and the fault map vector is compared with the similarities of each preset fault map to obtain the maximum similarity, and the fault map corresponding to the maximum similarity is set as the target fault map.
[0069] Specifically, the obtaining of the similarity between the fault spectrum vector and each of the preset fault spectra to determine the target preset fault spectrum includes: obtaining the similarity between the fault spectrum vector and each of the preset fault spectra; and determining the target fault spectrum with the highest similarity from the similarities between multiple fault spectrum vectors and each of the preset fault spectra.
[0070] Exemplarily, the similarity between the fault map vector and each of the preset fault maps is obtained, and each similarity is compared. If the highest similarity is determined to be 0.98, which is higher than the preset standard similarity of 0.95, it means that the preset fault map records this type of fault, and the target fault map is determined according to the highest similarity; if the highest similarity is determined to be 0.93, which is lower than the preset standard similarity, it means that the preset fault map does not record this type of fault and it is a new fault type. In this case, it is necessary to manually judge the fault type, give a corresponding maintenance strategy, and record the new fault type in the preset fault map, and the maintenance strategy in the maintenance instructions. Then, the similarity between the fault map vector and each of the preset fault maps is obtained again, and each similarity is compared. If the highest similarity is determined, and the highest similarity is higher than the preset standard similarity, the target fault map is determined according to the highest similarity.
[0071] Step S33: automatically recommending maintenance instructions based on the target preset fault map.
[0072] Exemplarily, for example, the direction of the fault map vector output by the fault map model is the tire repair direction, and the value is 26. The 26th page fault map of the tire repair category in the fault map corresponds to the 26th maintenance instruction, and then the 26th maintenance instruction is output.
[0073] In this embodiment, the fault spectrum vector is compared with a plurality of preset fault spectra; the similarity between the fault spectrum vector and each preset fault spectrum is obtained to determine a target preset fault spectrum; and maintenance instructions are automatically recommended based on the target preset fault spectrum. This solves the technical problem in related technologies that maintenance instructions need to be obtained by relying on manual labor and engineers' on-the-job experience, which requires a lot of manpower and time. A method for automatically and personalizedly recommending maintenance instructions is provided to improve the efficiency of obtaining maintenance instructions, thereby improving the efficiency of the entire maintenance process.
[0074] In a second aspect, an embodiment of the present application also provides a device for automatically recommending maintenance instructions.
[0075] In one embodiment, referring to Figure 3 , Figure 3 This is a functional module diagram of an embodiment of a device for automatically recommending maintenance instructions in this application. Figure 3 As shown, the device for automatically recommending maintenance instructions includes:
[0076] Collection module 01, used to collect vehicle data of the vehicle to be repaired;
[0077] An acquisition module 02 is used to acquire a fault spectrum vector model based on a preset neural network and the whole vehicle data of the vehicle to be repaired, wherein the fault spectrum vector model is obtained by training the preset neural network with training data, and the whole vehicle data includes static data, dynamic data and annotated fault spectrum vectors;
[0078] The recommendation module 03 is used to automatically recommend maintenance instructions based on the fault map vector model and the preset fault map.
[0079] Furthermore, in one embodiment, the acquisition module 02 is further used to:
[0080] Import multiple groups of vehicle data as training sets into the preset neural network for training, wherein the vehicle data includes static data and dynamic data, wherein the static data includes: tire pressure, wear degree, vehicle appearance, coolant level, fuel filter, fuel pump, fuel injector, oil leakage status of the oil circuit, and the dynamic data includes: maximum speed of the wheel, working conditions of the cooling fan, water tank, water pump, working conditions of the headlights, turn signals, brake lights, and reverse lights. The fault map vector model includes:
[0081] Where X t is the quantized value corresponding to the vehicle data, I t For X t The corresponding weight coefficient, f() is the activation function, b t is the bias term, Lt is the fault spectrum vector;
[0082] Determining whether the preset neural network is in a convergence state;
[0083] If the preset neural network is in a convergence state, the fault spectrum vector model is obtained.
[0084] Furthermore, in one embodiment, the recommendation module 03 is also used for:
[0085] Outputting a fault spectrum vector through the fault spectrum model;
[0086] Comparing the fault map vector with a plurality of preset fault maps;
[0087] Obtaining the similarity between the fault map vector and each of the preset fault maps to determine a target preset fault map;
[0088] According to the target preset fault map, maintenance instructions are automatically recommended.
[0089] Among them, the functional implementation of each module in the above-mentioned device for automatically recommending maintenance guidelines corresponds to each step in the above-mentioned method embodiment for automatically recommending maintenance guidelines, and its functions and implementation processes are no longer repeated here one by one.
[0090] In a third aspect, an embodiment of the present application provides a device for automatically recommending maintenance instructions. The device for automatically recommending maintenance instructions may be a personal computer (PC), a laptop computer, a server, or other device with data processing capabilities.
[0091] Reference Figure 4 , Figure 4 Schematic diagram of the hardware structure of the device for automatically recommending maintenance instructions involved in the embodiment of the present application. In the embodiment of the present application, the device for automatically recommending maintenance instructions may include a processor, a memory, a communication interface, and a communication bus.
[0092] The communication bus may be of any type and is used to interconnect the processor, the memory, and the communication interface.
[0093] The communication interface includes an input / output (I / O) interface, a physical interface, and a logical interface, etc., which are used to interconnect the internal components of the device for implementing the automatic recommendation of maintenance guidelines, and an interface for interconnecting the device for implementing the automatic recommendation of maintenance guidelines with other devices (such as other computing devices or user devices). The physical interface may be an Ethernet interface, an optical fiber interface, an ATM interface, etc.; the user device may be a display, a keyboard, etc.
[0094] The memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.
[0095] The processor may be a general-purpose processor, which may call the program for automatically recommending maintenance instructions stored in the memory and execute the method for automatically recommending maintenance instructions provided in the embodiment of the present application. For example, the general-purpose processor may be a central processing unit (CPU). The method executed when the program for automatically recommending maintenance instructions is called may refer to the various embodiments of the method for automatically recommending maintenance instructions in the present application, which will not be described in detail here.
[0096] Those skilled in the art will understand that Figure 4 The hardware structure shown in the figure does not constitute a limitation on the present application, and may include more or less components than shown in the figure, or combine certain components, or arrange the components differently.
[0097] In a fourth aspect, an embodiment of the present application also provides a computer-readable storage medium.
[0098] The computer-readable storage medium of the present application stores a program for automatically recommending maintenance guidelines, wherein when the program for automatically recommending maintenance guidelines is executed by a processor, the steps of the method for automatically recommending maintenance guidelines as described above are implemented.
[0099] Among them, the method implemented when the program for automatically recommending maintenance instructions is executed can refer to the various embodiments of the method for automatically recommending maintenance instructions in this application, and will not be repeated here.
[0100] It should be noted that the serial numbers of the above-mentioned embodiments of the present application are only for description and do not represent the advantages or disadvantages of the embodiments.
[0101] The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices. The terms "first", "second" and "third" are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit "first", "second" and "third" to different types.
[0102] In the description of the embodiments of the present application, "exemplary", "for example" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary", "for example" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary", "for example" or "for example" is intended to present related concepts in a specific way.
[0103] In the description of the embodiments of the present application, unless otherwise specified, “ / ” means or. For example, A / B can mean A or B. The “and / or” in the text is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, “multiple” refers to two or more than two.
[0104] In some processes described in the embodiments of the present application, multiple operations or steps that appear in a specific order are included, but it should be understood that these operations or steps may not be executed in the order in which they appear in the embodiments of the present application or in parallel, and the sequence number of the operation is only used to distinguish the different operations, and the sequence number itself does not represent any execution order. In addition, these processes may include more or fewer operations, and these operations or steps may be executed in sequence or in parallel, and these operations or steps may be combined.
[0105] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, disk, CD) as described above, and includes a number of instructions for a terminal device to execute the methods described in each embodiment of the present application.
[0106] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.
Claims
1. A method for automatically recommending maintenance instructions, characterized in that: The method for automatically recommending maintenance instructions comprises: Collect vehicle data of vehicles to be repaired; Based on the fault spectrum vector model, a fault spectrum vector is obtained according to the vehicle data of the vehicle to be repaired, wherein the fault spectrum vector model is obtained by training a preset neural network with training data, and the training data includes static data, dynamic data and annotated fault spectrum vectors; According to the fault map vector and the preset fault map, maintenance instructions are automatically recommended.
2. The method for automatically recommending maintenance instructions according to claim 1, characterized in that: The automatically recommending maintenance instructions according to the fault map vector and the preset fault map includes: Comparing the fault map vector with a plurality of preset fault maps; Obtaining the similarity between the fault map vector and each of the preset fault maps to determine a target preset fault map; According to the target preset fault map, maintenance instructions are automatically recommended.
3. The method for automatically recommending maintenance instructions according to claim 2, characterized in that: The obtaining of the similarity between the fault map vector and each of the preset fault maps to determine a target preset fault map includes: Obtaining the similarity between the fault spectrum vector and each of the preset fault spectra; A target fault map with the highest similarity is determined from the similarities between the plurality of fault map vectors and the respective preset fault maps.
4. The method for automatically recommending maintenance instructions according to claim 1, characterized in that: After automatically recommending maintenance instructions according to the fault map vector and the preset fault map, the method further includes: The maintenance guide is input into a target server so that the target server generates a maintenance strategy according to the maintenance guide, and the maintenance strategy is used to schedule an idle robot to perform maintenance on the vehicle to be maintained.
5. The method for automatically recommending maintenance instructions according to claim 1, characterized in that: Before obtaining the fault spectrum vector based on the fault spectrum vector model according to the vehicle data of the vehicle to be repaired, the method further includes: Importing the acquired data to be trained into the preset neural network for training, the data to be trained includes static data, dynamic data and annotated fault map vectors, wherein the static data includes: tire pressure, wear degree, vehicle appearance, coolant level, fuel filter, fuel pump, fuel injector, oil leakage status of the oil circuit, and the dynamic data includes: maximum speed of the wheel, working conditions of the cooling fan, water tank, water pump, working conditions of the headlights, turn signals, brake lights, and reverse lights; Determining whether the preset neural network after training is in a convergence state; If the trained preset neural network is in a convergence state, the fault spectrum vector model is obtained, and the fault spectrum vector model includes: Where X t is the quantized value corresponding to the vehicle data, I t For X t The corresponding weight coefficient, f() is the activation function, b t is the bias term, L t is the fault spectrum vector.
6. The method for automatically recommending maintenance instructions as claimed in claim 5, characterized in that: Determining whether the preset neural network is in a convergence state includes: Obtaining a loss value of the preset neural network after training; If the loss value is less than or equal to a preset standard loss value, it is determined that the preset neural network is in a convergence state; If the loss value is greater than a preset standard loss value, it is determined that the preset neural network is not in a convergence state, and the preset neural network is trained again until the preset neural network is in a convergence state.
7. The method for automatically recommending maintenance instructions according to claim 5, characterized in that: Determining whether the preset neural network is in a convergence state includes: Obtaining the training times of the preset neural network after training; If the number of training times is less than a preset standard number of training times, it is determined that the preset neural network is not in a convergence state, and the preset neural network is trained again until the preset neural network is in a convergence state; If the number of training times is greater than or equal to a preset standard number of times, it is determined that the preset neural network is in a convergence state.
8. A device for automatically recommending maintenance instructions, characterized in that: The device for automatically recommending maintenance instructions comprises: A collection module is used to collect the whole vehicle data of the vehicle to be repaired; An acquisition module, used for acquiring a fault spectrum vector model based on a preset neural network and the whole vehicle data of the vehicle to be repaired, wherein the fault spectrum vector model is obtained by training the preset neural network with training data, and the whole vehicle data includes static data, dynamic data and annotated fault spectrum vectors; The recommendation module is used to automatically recommend maintenance instructions based on the fault map vector model and the preset fault map.
9. A device for automatically recommending maintenance instructions, characterized in that: The device for automatically recommending maintenance instructions includes a processor, a memory, and a program for automatically recommending maintenance instructions stored in the memory and executable by the processor, wherein when the program for automatically recommending maintenance instructions is executed by the processor, the steps of the method for automatically recommending maintenance instructions as described in any one of claims 1 to 7 are implemented.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a program for automatically recommending maintenance guidelines, wherein when the program for automatically recommending maintenance guidelines is executed by a processor, the steps of the method for automatically recommending maintenance guidelines according to any one of claims 1 to 7 are implemented.