Building material loss prediction method and system
Through deep learning models and image and ultrasonic detection technology, the transportation loss of building materials is predicted, and the problem of unpredictable transportation loss in the existing technology is solved, and the accuracy of material supply and construction progress are guaranteed.
Patent Information
- Application Number
- CN202410981124.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-12
AI Technical Summary
The existing building materials loss prediction scheme fails to effectively predict losses during transportation, resulting in insufficient materials on the construction site, increasing costs and delaying construction periods.
The deep learning training model is used to combine image recognition and ultrasonic detection technology to obtain the types, transportation methods, distances and quality inspection information of building materials, calculate the material loss coefficient, accurately predict transportation losses, and ensure material supply and construction progress.
By accurately predicting transportation losses, optimizing material supply, saving costs, ensuring smooth progress of construction progress, reducing detection accuracy and work burden, and reducing hardware investment.
Smart Images

Figure CN118941108A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to big data processing technology, and in particular to a method and system for predicting building material loss based on a training model. Background Art
[0002] The estimation and prediction of building material loss is of great significance in the field of construction, for example: 1. Cost control. Through the estimation and prediction of the loss, the use of materials can be reasonably arranged during the procurement and construction process to avoid unnecessary cost expenditure; 2. Ensure the smooth progress of the project. Through the estimation and prediction of the loss, the delay of the construction period due to the lack of materials can be avoided. Therefore, the estimation and prediction of building material loss is a very important link in the field of construction. At present, the building material loss prediction scheme is mainly aimed at predicting the loss generated during the construction process, but it does not predict the loss generated during the transportation of building materials. Therefore, it often happens that the number of available building materials delivered to the construction site does not meet the construction requirements due to the large transportation loss, which will cause the need to rearrange resources for material transportation (i.e., cost increase) and slow down the construction progress. It can be seen that an estimation and budgeting scheme for the transportation loss of building materials is one of the problems that urgently need to be solved. Summary of the invention
[0003] In view of this, the present application aims to solve at least one of the technical problems existing in the prior art. To this end, the present application proposes a loss prediction method and system for building materials to accurately estimate and predict the transportation loss of building materials.
[0004] In a first aspect, an embodiment of the present application provides a method for predicting loss of building materials, the method comprising:
[0005] Acquire a first information input matrix, wherein the information in the first information input matrix includes types of building materials, transportation methods, and transportation distances;
[0006] After inputting the first information input matrix into the deep learning training model for data processing, a material loss coefficient is obtained, wherein the material loss coefficient is used to represent the ratio between the amount of material loss and the amount of initial material.
[0007] In some embodiments, the transportation method includes a shock absorption coefficient of the transportation tool, and the step of determining the shock absorption coefficient of the transportation tool includes:
[0008] Determine the corresponding reference shock absorption coefficient according to the type of transportation vehicle;
[0009] When it is determined that a shock-absorbing structure is provided in the transport vehicle and / or the packaging of the building materials is anti-collision packaging, the reference shock-absorbing coefficient is increased.
[0010] In some embodiments, the information in the first information input matrix also includes material loading and unloading methods and / or weather forecast information.
[0011] In some embodiments, the material loading and unloading method includes the type of loading and unloading tools and the shock absorption coefficient of the loading and unloading tools.
[0012] In some embodiments, the method further includes the step of training a deep learning training model, wherein the step of training the deep learning training model includes:
[0013] Acquire a second information input matrix, wherein the information in the second information input matrix includes types of historical building materials, historical transportation methods, and historical transportation distances;
[0014] Obtaining a historical material loss coefficient, wherein the historical material loss coefficient is determined based on a historical initial material quantity and a historical material loss quantity;
[0015] After training the deep learning training model using the second information input matrix and the historical material loss coefficient, a trained deep learning training model is obtained.
[0016] In some embodiments, the method further comprises the step of obtaining a historical material loss amount, wherein the step of obtaining a historical material loss amount comprises:
[0017] Obtain quality inspection information of construction materials transported to the construction site;
[0018] Determine the building materials that do not meet the quality requirements based on the quality inspection information;
[0019] The quantity of building materials that do not meet quality requirements shall be treated as the historical material loss quantity.
[0020] In some embodiments, the quality inspection information is information obtained by conducting quality inspection on the construction materials transported to the construction site using image recognition technology and / or ultrasonic detection technology.
[0021] In a second aspect, an embodiment of the present application provides a system for predicting loss of building materials, the system comprising:
[0022] Testing devices for testing the quality of building materials;
[0023] A processing device, comprising at least one processor, for executing the steps of the above-mentioned method for predicting the loss of building materials;
[0024] The detection device is communicatively connected with the processing device.
[0025] In some embodiments, the detection device includes an image recognition device and / or an ultrasonic detection device.
[0026] In a third aspect, an embodiment of the present application provides a system for predicting loss of building materials, the system comprising:
[0027] A first acquisition module is used to acquire a first information input matrix, wherein the information in the first information input matrix includes the type, transportation method and transportation distance of the building materials;
[0028] The first processing module is used to input the first information input matrix into the deep learning training model for data processing to obtain a material loss coefficient, wherein the material loss coefficient is used to represent the ratio between the amount of material loss and the amount of initial material.
[0029] The present application can achieve the following technical effects: the present application scheme obtains variable parameters that are strongly correlated with the transportation loss of building materials, wherein the type of the variable parameters at least includes the type of building materials, the mode of transportation, and the transportation distance, and then inputs the variable parameters into the trained deep learning training model for processing, and then obtains the corresponding material loss coefficient, wherein the material loss coefficient is used to represent the proportional relationship between the amount of material loss and the amount of initial material. It can be seen that by estimating the predicted material loss coefficient, the amount of initial materials required for transportation can be determined based on the actual amount of materials required at the construction site and the material loss coefficient, which is conducive to the rational arrangement and utilization of transportation resources and materials, avoids waste, saves costs, and can ensure the smooth progress of the project. In addition, the present scheme uses a deep learning training model for prediction and processing, which has high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0030] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments.
[0031] Figure 1 A structural block diagram of a building material loss prediction system provided in an embodiment of the present application;
[0032] Figure 2 A schematic diagram of the steps of a method for predicting loss of building materials provided in an embodiment of the present application;
[0033] Figure 3 Another structural block diagram of a building material loss prediction system provided in an embodiment of the present application. DETAILED DESCRIPTION
[0034] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described through the implementation method with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are 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 the present application.
[0035] In order to ensure that the cost of the construction project is within an acceptable range and to ensure the smooth progress of the construction project, it is very necessary to estimate the loss of building materials. As for the loss of building materials, in addition to the loss of building materials during the construction process, there will actually be loss of building materials during the transportation process. For example, the bumps and vibrations of the transport vehicle during the transportation will damage some building materials, resulting in the existence of building materials that do not meet the quality requirements after the building materials are transported to the construction site. The current number of available building materials cannot meet the needs of the current construction progress, which requires the re-arrangement of transportation resources and slows down the construction progress. To this end, the embodiment of the present application provides a solution for predicting the loss of building materials during the transportation process.
[0036] Reference Figure 1 In order to achieve accurate prediction of building material loss, the present application mainly adopts a deep learning training model to achieve this. Therefore, the implementation of this solution mainly includes two parts. One part is the training content for the deep learning training model, and the other part is to use the trained deep learning training model to process data to obtain the required loss parameters. Therefore, a building material loss prediction system provided in an embodiment of the present application specifically includes: a detection device and a processing device, and the detection device and the processing device are communicatively connected. Their specific description is as follows.
[0037] The detection device is used to perform quality detection on building materials.
[0038] Specifically, in order to train the deep learning training model, it is necessary to obtain the loss of building materials during transportation, so it is necessary to perform quality inspection on the building materials transported to the construction site, that is, this detection device is mainly a detection device placed at the construction site. At present, the detection device mainly includes some tools for assisting staff in building material quality inspection, such as laser level, ruler and other tools to detect whether the steel is bent and deformed due to transportation, or a camera with a simple shooting function to shoot the surface of the steel. The staff can manually detect and confirm whether there are too many cracks on the steel through the captured pictures / videos, which makes it unusable, and then input the detection information obtained after the quality inspection of the building materials into the information recording device in the detection device. The information recording device can send the quality inspection information of the building materials to the processing device, so that it can determine the amount of material loss based on the quality inspection information. It can be seen that this currently commonly used detection method still involves too much manual operation, which not only increases the workload of the staff, but also requires additional inspection personnel, increases investment costs, and has low detection accuracy.
[0039] Based on this, in some embodiments, the detection device includes an image recognition device and / or an ultrasonic detection device.
[0040] Specifically, the image recognition device may be a three-dimensional laser image camera, i.e., a three-dimensional laser scanner. Since the working principle of the three-dimensional laser scanner is to perform laser scanning on the target by utilizing the principle of laser ranging, a three-dimensional model of the target can be obtained by recording the three-dimensional coordinates, reflectivity, texture and other information of a large number of dense points on the surface of the target, and the cracks / cracks existing on the surface of the steel can be identified based on the three-dimensional coordinates, reflectivity, texture and other information of the points on the surface of the target. For example, since the cracks / cracks will cause the laser signal emitted by the reflection device to fail to be reflected, if the laser intensity corresponding to the three-dimensional coordinates of the points in an area is lower than the first preset threshold, the area can be determined to be a crack / crack. It can be seen that in this way, the surface cracks of the building materials can be automatically identified, and when the number of cracks exceeds the quality specification requirements, it is determined that the building materials do not meet the quality requirements. In addition, it should be noted that, in order to better manage resources at construction sites, three-dimensional laser imaging devices are generally used to perform laser scanning on construction sites to obtain three-dimensional models, and then the model is optimized and matched with the management and control data corresponding to each node in combination with BIM technology, so as to conduct all-round management and control of construction projects at construction sites in real time. It can be seen that for this embodiment, a three-dimensional laser scanner is used to realize the quality inspection of building materials, so there is no need to purchase additional equipment, which reduces the investment cost of hardware equipment.
[0041] In addition, the above-mentioned image recognition device is mainly used to identify and detect surface defects of building materials, but for defects inside building materials, or problems with the building materials themselves, such as material deterioration, local corrosion / rust problems on the materials, or changes in the toughness / rigidity of the local position of the building materials due to collision or stretching, it is impossible to identify them through the above-mentioned image recognition device. For this reason, this embodiment also combines an ultrasonic detection device to perform quality detection on building materials. Specifically, an ultrasonic detection device is used to transmit an ultrasonic signal to the target to be measured at a specific position, and then the ultrasonic detection device identifies the sound wave feature information based on the ultrasonic signal reflected back by the target to be measured. When the difference between it and the original sound wave feature information is very large, for example, the similarity of the change trend between the spectral characteristics of the sound wave is lower than the second threshold, it is judged that after being transported to the construction site, it has suffered relatively large damage, so it does not meet the quality requirements, so it should be treated as an unusable building material. Similarly, since ultrasonic detection devices are needed in work nodes such as cave detection and soil moisture content detection, it can be seen that reusing them in the quality inspection of building materials does not require excessive additional hardware purchase costs.
[0042] It can be seen that by adopting the above-mentioned detection device to realize the quality inspection of building materials, not only the workload of the staff is reduced and the work efficiency is improved, but also the quality inspection is carried out by combining the detection technology of three-dimensional images and ultrasonic signals, which has high accuracy and does not require too much investment in the purchase cost of hardware equipment.
[0043] The processing device includes at least one processor for executing the following steps of the method for predicting and estimating the loss of building materials. The specific description of the steps of the method is as follows.
[0044] Reference Figure 2 , a method for predicting and estimating building material loss, the method comprises the following steps.
[0045] A. Training the deep learning training model
[0046] A1. Obtain a second information input matrix, wherein the information in the second information input matrix includes types of historical building materials, historical transportation methods, and historical transportation distances.
[0047] Specifically, for the loss of building materials during transportation, it is found that the types of variable parameters that are strongly correlated with them are: 1. Types of building materials. For example, when the building materials are glass, ceramics and other easily broken materials, the corresponding transportation loss rate will be relatively high. Similarly, if the building materials are heavy materials such as stone and steel, they will produce different degrees of wear and / or deformation due to the bumps and vibrations of the transportation tools. In other words, they also have a certain transportation loss rate, but it will be lower than the transportation loss rate of the above-mentioned easily broken materials. Similarly, for different types of building materials, they will have corresponding different transportation loss rates; 2. Transportation distance. The bumps and vibrations on the transportation route will cause a certain degree of collision between building materials or between building materials and the storage space of the transportation vehicle, which will cause the loss of building materials. The longer the transportation distance, the higher the transportation loss rate; 3. Transportation mode. Due to different transportation modes and different transportation tools, the corresponding shock absorption coefficients will also be different. The higher the shock absorption coefficient, the higher the shock absorption capacity, and the lower the bumps and vibrations to which the building materials will be subjected during transportation, that is, the lower the damage. Therefore, in this embodiment, the types of historical building materials, historical transportation methods, and historical transportation distances are used as training input data, which can more accurately predict and estimate the corresponding transportation loss rate, that is, the loss coefficient of building materials.
[0048] In some embodiments, the type of building materials not only refers to the property type of the building materials themselves, such as glass, ceramics, steel, stone, etc., but may also include the merchants described in the building materials. For example, the type of building materials belongs to steel from supplier a, stone from supplier b, etc. In other words, the type of building materials includes the property type and the supplier type. In this way, when the predicted estimated material loss coefficient is relatively large, in addition to considering other solutions, you can also consider changing the supplier of the building materials with this property (because the damage resistance of building materials with the same property provided by different suppliers will be different) to reduce the predicted estimated material loss coefficient.
[0049] Therefore, the second information input matrix is represented as [C0, C1, ..., Cn, D0, D1, ..., Dn, E0, E1, ..., En]. The values of C0 ... Cn are used to represent different types of building materials, among which the values of C0 ... Ci are used to represent the suppliers of building materials, and the values of Ci+1 ... Cn are used to represent the property types of building materials; the values of D0 ... Dn are used to represent the transportation distance; and the values of E0 ... En are used to represent the mode of transportation. n is a positive integer, and its specific value can be set according to actual needs.
[0050] In addition, it should be explained that, since the above-mentioned second information input matrix is composed of existing historical information, the data used to constitute the second information input matrix - the type of building materials, transportation method, and transportation distance, all refer to historically known data.
[0051] In some embodiments, the transportation mode includes the shock absorption coefficient of the transportation tool, that is, the value of E0...En is used to represent the shock absorption coefficient of the transportation tool. The step of determining the shock absorption coefficient of the transportation tool specifically includes the following steps.
[0052] A11. Determine the corresponding reference shock absorption coefficient according to the type of transportation.
[0053] Specifically, if the transport vehicle is a vehicle, the shock absorption coefficients of different types of vehicles are different. Generally, the shock absorption coefficient may refer to the damping coefficient of the shock absorber of the car, which ranges from 650Ns / m to 2013Ns / m, and is used to indicate the shockproof performance of the vehicle. The better the performance, the less vibration the interior of the vehicle can receive. Therefore, by obtaining the specific model of the transport vehicle currently used to transport construction materials, the shock absorption coefficient corresponding to the transport vehicle can be accurately obtained. At this time, the shock absorption coefficient is the benchmark shock absorption coefficient.
[0054] A12. When it is determined that the transport vehicle is provided with a shock-absorbing structure and / or the packaging of the building materials is anti-collision packaging, the base shock-absorbing coefficient is increased.
[0055] Specifically, in addition to the shock-absorbing capability of the transport vehicle itself, in order to improve the shock-absorbing effect of the building materials, a shock-absorbing structure can be set in the transport vehicle and the building materials can be packaged in an anti-collision packaging method to further reduce the vibration damage caused by the transport vehicle to the building materials during transportation, which is equivalent to further improving the shock-absorbing performance of the transport vehicle. Therefore, when the above-mentioned shock-absorbing structure and / or anti-collision packaging are present, the value of the reference shock-absorbing coefficient can be increased.
[0056] In some embodiments, for the above-mentioned increase in the baseline shock absorption coefficient, the specific calculation formula used is: K1 = (1 + k2) * (1 + k3) * K0. Among them, K1 represents the final shock absorption coefficient obtained, k2 is the first adjustment coefficient corresponding to the shock absorption performance of the shock absorption structure, k3 is the second adjustment coefficient corresponding to the shock absorption performance of the anti-collision package, and K0 is represented as the baseline shock absorption coefficient. It can be seen that when there is a shock absorption structure and / or an anti-collision package, the value of K0 will be increased to obtain the final shock absorption coefficient. In addition, for the first adjustment coefficient k2, its value is proportional to the shock absorption performance of the shock absorption structure. The better the shock absorption performance, the larger the value of k2. Similarly, the better the shock absorption performance of the anti-collision package, the larger the value of k3.
[0057] In some embodiments, the second information input matrix may also include historical material loading and unloading patterns and / or historical weather forecast information.
[0058] It is further found that regarding the estimation and prediction of transportation loss of construction materials, it also has the following two types of variable parameters that are strongly correlated with it.
[0059] 1. Material loading and unloading methods. Incorrect loading / unloading methods will cause materials to be easily damaged by collision. In particular, if the loading / unloading method does not have a shockproof protection structure, it is more likely to be damaged by collision. Therefore, the type of loading and unloading method and its shockproof protection performance are strongly correlated with the material loss during transportation. The type of loading and unloading method can be specifically represented by the type of loading and unloading tools, and the shockproof protection performance of the method can be represented by the shockproof coefficient of the loading and unloading tools. Specifically, for the second information input matrix, it also includes F0...Fn, and the values of F0...Fi are used to correspond to the types of loading and unloading tools, and the values of Fi+1...Fn are used to correspond to the shockproof coefficients of the loading and unloading tools.
[0060] 2. Weather forecast information. The worse the weather, the more severe the bumps on the transportation route, which increases the wear and tear of materials during transportation. Therefore, there is a strong correlation between weather conditions and material loss during transportation. Generally, the weather information is the forecast information for the future transportation time, which can be obtained from various public official information. Specifically, for the second information input matrix, it also includes T0...Tn, which is mainly used to correspond to the weather type information predicted in the future transportation time period.
[0061] It can be seen that by supplementing and improving the types of training input data, it is possible to further adapt to complex transportation conditions, thereby further improving the accuracy of the predicted estimated material loss coefficient.
[0062] A2. Obtain a historical material loss coefficient, wherein the historical material loss coefficient is determined based on the historical initial material quantity and the historical material loss quantity.
[0063] Specifically, the initial material quantity refers to the quantity of construction materials that are initially scheduled for transportation, that is, the quantity of construction materials to be transported, while the material loss quantity refers to the quantity of construction materials that are damaged and cannot be used after being transported to the destination construction site, and the material loss coefficient refers to the ratio between the material loss quantity and the initial material quantity. The initial material quantity can be obtained directly from the project management form.
[0064] Regarding the material loss quantity, in some embodiments, the specific steps for obtaining it are as follows.
[0065] A21. Obtain quality inspection information of construction materials transported to the construction site.
[0066] Specifically, for the quality inspection information, the above-mentioned image recognition device and / or ultrasonic detection device can be used to perform quality inspection on the construction materials transported to the construction site to obtain the corresponding quality inspection information.
[0067] A22. Determine the construction materials that do not meet the quality requirements based on the quality inspection information.
[0068] Specifically, step A22 includes the following steps.
[0069] A221. Surface defects of building materials are identified through image recognition technology using an image recognition device.
[0070] A222. After transmitting ultrasonic signals to building materials through the ultrasonic detection technology of the ultrasonic detection device, the ultrasonic signals reflected by the building materials are received, and the acoustic signal characteristics (such as frequency, amplitude, duration, etc.) of the received ultrasonic signals are identified, and then the difference between the identified acoustic signal characteristics and the initial acoustic signal characteristics is calculated. The initial acoustic signal refers to the ultrasonic signal obtained after ultrasonic detection of building materials that have not been transported and meet the quality requirements; and for the difference, the larger the value, the greater the difference between the identified acoustic signal characteristics and the initial acoustic signal characteristics, and the greater the damage to the building materials transported to the construction site.
[0071] A223. When the number of identified surface defects is greater than or equal to a first quantity threshold and / or the difference is greater than or equal to a first preset value, the building material is determined to be a building material that does not meet quality requirements.
[0072] A224. When the number of identified surface defects is greater than or equal to a second number threshold, and the difference is greater than or equal to a second preset value, the building material is determined to be a building material that does not meet quality requirements. The second numerical threshold is less than the second numerical threshold, and the second preset value is less than the first preset value.
[0073] For the above step A223, when any damage of the building material exceeds the acceptable range, the building material is directly judged as not meeting the quality requirements. However, the same building material may have both surface defects and internal defects (not visible to the naked eye), so an overall judgment is made based on the existence of both. Therefore, when the number of surface defects is greater than or equal to the second number threshold and less than the first number threshold, and the difference is greater than or equal to the second preset value and less than the first preset value, the overall judgment is that the building material still does not meet the quality requirements.
[0074] It can be seen that by adopting the above method to judge whether the building materials meet the quality requirements, not only the processing efficiency is high, but also the accuracy is further improved.
[0075] A23. The quantity of construction materials that do not meet quality requirements shall be treated as historical material loss quantity.
[0076] A3. After training the deep learning training model using the second information input matrix and the historical material loss coefficient, a trained deep learning training model is obtained.
[0077] Since the above is a training step, the training input data and training output data mentioned above all contain historical data.
[0078] B. Use the trained deep learning training model to run the application online
[0079] B1. Obtain a first information input matrix, wherein the information in the first information input matrix includes types of building materials, transportation methods, and transportation distances.
[0080] Based on the above training steps, it can be known that in one embodiment, the transportation mode includes a shock absorption coefficient of the transportation tool, and the step of determining the shock absorption coefficient of the transportation tool includes:
[0081] B11. Determine the corresponding reference shock absorption coefficient according to the type of transportation vehicle;
[0082] B12, when it is determined that the transport vehicle is provided with a shock-absorbing structure and / or the building material package is a collision-proof package, the base shock-absorbing coefficient is increased. The increase of the base shock-absorbing coefficient described here is the same as the above step A12, and will not be repeated here.
[0083] In one embodiment, the information in the first information input matrix also includes material loading and unloading methods and / or weather forecast information.
[0084] In one embodiment, the material loading and unloading method includes the type of loading and unloading tools and the shock absorption coefficient of the loading and unloading tools.
[0085] Since the above information mentioned in the training step and the information mentioned in this step have the same meaning and function, except that the information in the training step is historical information, while the information in this step is the information actually obtained when the training model is run, the above information will not be elaborated here.
[0086] B2. After inputting the first information input matrix into the deep learning training model for data processing, a material loss coefficient is obtained, wherein the material loss coefficient is used to represent the ratio between the amount of material loss and the amount of initial material.
[0087] Specifically, after obtaining the material loss coefficient, the initial amount of building materials, that is, the amount of building materials to be transported, can be calculated based on the actual amount of building materials required at the construction site and the loss coefficient. It can be seen that by using the above scheme, it is conducive to the reasonable allocation and management of building material resources, and it can also improve processing efficiency and the accuracy of prediction and estimation.
[0088] Reference Figure 3 The present application also provides a system for predicting the loss of building materials, the system comprising:
[0089] A first acquisition module is used to acquire a first information input matrix, wherein the information in the first information input matrix includes the type, transportation method and transportation distance of the building materials;
[0090] The first processing module is used to input the first information input matrix into the deep learning training model for data processing to obtain a material loss coefficient, wherein the material loss coefficient is used to represent the ratio between the amount of material loss and the amount of initial material.
[0091] In one embodiment, the mode of transportation includes a shock absorption coefficient of the transportation tool, and the system further includes a first determination module, the first determination module is used to determine the shock absorption coefficient of the transportation tool, and the first determination module specifically includes:
[0092] A first determination submodule, used to determine a corresponding reference shock absorption coefficient according to the type of transportation vehicle;
[0093] The first processing submodule is configured to increase the reference shock absorption coefficient when it is determined that the transport vehicle is provided with a shock absorption structure and / or the packaging of the building materials is an anti-collision packaging.
[0094] In one embodiment, the information in the first information input matrix also includes material loading and unloading methods and / or weather forecast information.
[0095] In one embodiment, the material loading and unloading method includes the type of loading and unloading tools and the shock absorption coefficient of the loading and unloading tools.
[0096] In one embodiment, the system further includes a training system, and the training system is used to train the deep learning training model. The training system specifically includes:
[0097] A first acquisition submodule is used to acquire a second information input matrix, wherein the information in the second information input matrix includes types of historical building materials, historical transportation methods, and historical transportation distances;
[0098] A second acquisition submodule is used to acquire a historical material loss coefficient, wherein the historical material loss coefficient is determined based on the historical initial material quantity and the historical material loss quantity;
[0099] The first training module is used to train the deep learning training model using the second information input matrix and the historical material loss coefficient to obtain a trained deep learning training model.
[0100] In one embodiment, a third acquisition submodule is further included, and the third acquisition submodule is used to obtain the historical material loss amount. The third acquisition submodule also includes:
[0101] A first acquisition unit is used to acquire quality inspection information of construction materials transported to the construction site;
[0102] A first determination unit, used to determine the building materials that do not meet the quality requirements according to the quality detection information;
[0103] The first determining unit is used to take the quantity of the building materials that do not meet the quality requirements as the historical material loss quantity.
[0104] In one embodiment, the quality inspection information is information obtained by performing quality inspection on the construction materials transported to the construction site using image recognition technology and / or ultrasonic detection technology.
[0105] Since the processing device of the system embodiment includes at least one processor for executing the steps of the above method embodiment, the beneficial effects of the system of the present embodiment are the same as the beneficial effects of the above method embodiment, and will not be elaborated in detail here.
[0106] In addition, the present application also provides a computer-readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the steps of the above method embodiment.
[0107] For the processors mentioned in the above storage medium embodiment and the system embodiment, the number can be at least 1, and at least any step in the above method embodiment can be executed. When the number is at least two, at least two processors can be connected to each other by communication, not limited to wired or wireless communication connection, and the at least one processor can be connected to various intelligent terminal devices by communication. In addition, the processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.
[0108] Finally, it should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0109] Note that the above are only preferred embodiments of the present application and the technical principles used. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the appended claims.
Claims
1. A method for predicting the loss of building materials, It is characterized in that The method includes: Acquire a first information input matrix, wherein the information in the first information input matrix includes types of building materials, transportation methods, and transportation distances; After inputting the first information input matrix into the deep learning training model for data processing, a material loss coefficient is obtained, wherein the material loss coefficient is used to represent the ratio between the amount of material loss and the amount of initial material.
2. The method according to claim 1, It is characterized in that The transportation mode includes a shock absorption coefficient of the transportation tool, and the steps for determining the shock absorption coefficient of the transportation tool include: Determine the corresponding reference shock absorption coefficient according to the type of transportation vehicle; When it is determined that a shock-absorbing structure is provided in the transport vehicle and / or the packaging of the building materials is anti-collision packaging, the reference shock-absorbing coefficient is increased.
3. The method according to claim 1, It is characterized in that The information in the first information input matrix also includes material loading and unloading methods and / or weather forecast information.
4. The method according to claim 3, It is characterized in that The material loading and unloading method includes the type of loading and unloading tools and the shock absorption coefficient of the loading and unloading tools.
5. The method according to any one of claims 1 to 4, It is characterized in that The method also includes the step of training the deep learning training model, wherein the step of training the deep learning training model includes: Acquire a second information input matrix, wherein the information in the second information input matrix includes types of historical building materials, historical transportation methods, and historical transportation distances; Obtaining a historical material loss coefficient, wherein the historical material loss coefficient is determined based on a historical initial material quantity and a historical material loss quantity; After training the deep learning training model using the second information input matrix and the historical material loss coefficient, a trained deep learning training model is obtained.
6. The method according to claim 5, It is characterized in that The method further includes the step of obtaining the historical material loss amount, wherein the step of obtaining the historical material loss amount includes: Obtain quality inspection information of construction materials transported to the construction site; Determine the building materials that do not meet the quality requirements based on the quality inspection information; The quantity of building materials that do not meet quality requirements shall be treated as the historical material loss quantity.
7. The method according to claim 6, It is characterized in that The quality inspection information is information obtained by inspecting the quality of the construction materials transported to the construction site using image recognition technology and / or ultrasonic detection technology.
8. A loss prediction system for building materials, It is characterized in that The system includes: A first acquisition module is used to acquire a first information input matrix, wherein the information in the first information input matrix includes the type, transportation method and transportation distance of the building materials; The first processing module is used to input the first information input matrix into the deep learning training model for data processing to obtain a material loss coefficient, wherein the material loss coefficient is used to represent the ratio between the amount of material loss and the amount of initial material.
9. A loss prediction system for building materials, It is characterized in that The system includes: Testing devices for testing the quality of building materials; A processing device, comprising at least one processor, for executing the method according to any one of claims 1 to 7; The detection device is communicatively connected with the processing device.
10. The system of claim 9, It is characterized in that The detection device includes an image recognition device and / or an ultrasonic detection device.
Citation Information
Cited By
Flood prevention block stone loss rate estimation method and device, medium and computer program
CN122492798A
A flood prevention block stone loss rate estimation method, device, medium and computer program
CN122492798B
