Blanking method of refined raw materials for building

Abstract

The invention discloses a refined raw material blanking method for buildings, relates to the technical field of building construction, and has the effects that data is acquired in real time through a data acquisition and modeling module, then the size and shape of a material are processed through a path generation module, and a surface image of the material is captured in real time; a geometric digital model of a material is generated after key features are extracted, then the geometric digital model of the material is combined with data collected by a sensor in real time, an initial path is generated through an algorithm, detailed optimization is performed by adopting specific parameters, and the accuracy of the path and the material utilization rate are ensured; and finally, the material utilization rate and the blanking precision are improved through an optimization algorithm, and after an initial path is generated, the system dynamically adjusts the path according to sensor data fed back by the real-time monitoring module.

Description

Technical Field

[0001] The invention relates to the technical field of building construction, and more specifically, to a method for cutting refined raw materials for construction. Background Art

[0002] With the acceleration of urbanization and the rapid development of the construction industry, the demand for building materials is growing. In this process, the cutting of building materials is an important part of construction production. Cutting refers to the process of removing a certain shape, quantity or quality of materials from the whole or batch of materials after determining the shape, quantity or quality of the materials required to make a certain equipment or product. In the building materials industry, cutting technology is widely used in the processing of various building materials such as steel, aluminum, and wood.

[0003] However, the existing cutting method technology has a fixed path and cannot dynamically adjust the cutting path according to real-time data changes, resulting in low material utilization and a lot of leftovers. Low material utilization means that more materials are needed to meet the same construction needs, which directly increases the material procurement cost. Excessive leftover materials require additional storage space, and the transportation process will also increase energy consumption and expenses, indirectly increasing costs.

[0004] Therefore, in order to solve the above technical problems, the present application proposes a method for cutting refined raw materials for construction. Summary of the invention

[0005] In view of the shortcomings of the prior art, the object of the present invention is to provide a method for cutting refined raw materials for construction.

[0006] To achieve the above object, the present invention provides the following technical solution: a method for cutting refined raw materials for construction, comprising the following steps:

[0007] A: Build a data acquisition and modeling module to collect material data through sensors and input or build a geometric digital model of the building product;

[0008] B: Set up a path generation module to generate a feasible material cutting path plan, generate a geometric digital model of the material, and optimize the geometric digital model of the material, the data model collected by the sensor in real time, and the geometric digital model of the building product through the principal component analysis algorithm KCPA based on the kernel function; optimize the material cutting path plan based on the above-mentioned optimized geometric digital model of the material, the data model collected by the sensor in real time, and the geometric digital model of the building product;

[0009] C: Set up a real-time monitoring module, which summarizes the status of materials and machine tools monitored by position sensors, pressure sensors, current sensors, temperature sensors, and vibration sensors to form a unified data stream, perform time synchronization and spatial alignment, form a complete description of material and equipment status, and perform abnormality detection;

[0010] D: Build a verification and adjustment module, the verification module tests the optimized material cutting path plan in a computer simulation environment to obtain feasible parameters of the optimized material cutting path plan; obtain feedback information of the feasible parameters of the optimized material cutting path plan through the verification module; the real-time adjustment module adjusts the material cutting path plan according to the feedback information of the verification module.

[0011] Preferably, the data acquisition and modeling module annotates material properties, adds property labels such as density and weight to each material, and creates a record for each material in the database.

[0012] Preferably, the path generation module removes high-frequency noise in the measurement data by Fast Fourier Transform (FFT) to ensure the accuracy of material size and shape.

[0013] Preferably, the path generation module applies the Canny edge detection algorithm to perform image gradient detection, extract edges and key features, and generate a geometric digital model of the material.

[0014] Preferably, the path generation module improves the accuracy of edge detection through non-maximum suppression and dual threshold detection, and generates an accurate material geometry digital model.

[0015] Preferably, the path generation module generates an initial path through dynamic programming, and after constructing a path cost matrix, performs a recursive search through a dynamic programming algorithm (DP).

[0016] Preferably, after the initial path is generated by the path generation module, the system dynamically adjusts the path according to real-time feedback of sensor data, such as cutting pressure, material thickness, etc.

[0017] Preferably, the path generation module adopts an enhanced genetic algorithm (EGA) to evaluate the adaptability of each path scheme in combination with material utilization and molding time.

[0018] Preferably, the genetic operation in the path generation module selects excellent individuals for mating and mutation through a tournament selection mechanism to generate new path plans, and avoids falling into local optimality through iterative optimization.

[0019] Preferably, the path generation module controls the probability of accepting inferior solutions through simulated annealing (SA) operation in combination with a genetic algorithm, so as to enhance the stability of path optimization.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] In the present invention, multi-dimensional data is collected in real time and combined with geometric modeling to dynamically generate and optimize the cutting path. Different from the fixed path in the prior art, this method can dynamically adjust the cutting path according to real-time data changes, thereby greatly improving the utilization rate of materials.

[0022] In the present invention, by combining sensor feedback and intelligent algorithms, such as enhanced genetic algorithm and simulated annealing, not only can the cutting speed be adjusted according to real-time data such as pressure and thickness during the cutting process, but also the path optimization algorithm can be continuously iterated to avoid local optimal problems and achieve global optimal path planning. DETAILED DESCRIPTION

[0023] The present invention provides a method for cutting refined raw materials for construction, and the specific method is divided into the following steps:

[0024] A: Data Collection and Modeling

[0025] A1: Data collection:

[0026] A11: Sensor type: Pressure and weighing sensors are used for data collection. Pressure and weighing sensors are installed under the support frame and platform of the processing equipment. On the one hand, they are used to monitor the weight of the material to help calculate the density and storage volume of the material. On the other hand, they are used to monitor the pressure changes of the equipment during the processing process in real time.

[0027] A12: Data collection process: First, perform initialization settings and calibrate all sensors to ensure measurement accuracy. Configure the data collection frequency and set the sensor sampling interval according to the speed and accuracy requirements of the production line, and finally start real-time data collection.

[0028] A2: Data modeling: Create targets based on metric conventional models, build digital models of building products, accurately describe their geometric features and physical properties, and provide a data basis for subsequent material optimization. The specific steps are as follows:

[0029] A21: Geometric model creation: First, create a target based on the metric conventional model, set the corresponding parameters to complete 3D modeling, use computer-aided design (CAD) software to generate a three-dimensional geometric model of the material, and set the corresponding type parameters and instance parameters.

[0030] A22: Material property labeling: Attach property labels to each material, such as density and weight. In the previous steps, the weighing sensor and material library have acquired some data, calculated and labeled the density and weight of the material. Create a record for each material in the database, including the geometry model path and property information.

[0031] A3: Data integration: Integrate modeling results and data collected by sensors to ensure data integrity.

[0032] B: Path Generation

[0033] B1: Initial solution generation: Generate a feasible initial material cutting path to provide a basis for subsequent optimization. The specific steps are as follows:

[0034] B11: In the process of cutting, accurate material size and shape data are the basis for optimizing the path. Fast Fourier Transform (FFT) removes high-frequency noise in the measured data to ensure the accuracy of material size and shape, providing stable data for subsequent path generation.

[0035] B12: Capture the material surface image through a high-definition camera, apply the Canny edge detection algorithm to detect the image gradient, extract the edges and key features, and generate a digital model of the material. Evaluate the material utilization rate of the initial path plan based on the target detection evaluation index, and use the total material area as an input parameter to preliminarily evaluate the material utilization rate of the path.

[0036] Image recognition and directional gradients are then used to identify the edges and key features of the material surface. This process improves the accuracy of edge detection through non-maximum suppression and dual threshold detection, generating an accurate material geometry model. Identify the edges and key features of the material surface. This process improves the accuracy of edge detection through non-maximum suppression and dual threshold detection, generating an accurate material geometry model.

[0037]

[0038] Among them, and are the images in G x and G y Directional gradient; and It is the gradient of the image in the x and y directions, which is used to indicate the rate of change of the image in these two directions; I indicates the image intensity value; and are the partial derivatives of the image in the x and y directions, respectively, which are used to detect edge changes.

[0039] B2: Initial path generation

[0040] In the process of cutting materials, initial path generation is the basic step to optimize material utilization. This step combines the geometric model of the material with the data collected by the sensor in real time, generates the initial path through a series of intelligent algorithms, and uses specific parameters for detailed optimization to ensure the accuracy of the path and material utilization.

[0041] B21: The first step is geometric data processing and material modeling.

[0042] First, in terms of material modeling, a high-precision 3D laser scanner is used to scan the material, and the high-frequency noise is removed by Fast Fourier Transform (FFT) to ensure the accuracy of the model's geometric data. The scanning accuracy is 0.05 mm, which is suitable for the geometric feature extraction of complex materials. Subsequently, edge detection is performed. The material boundary is accurately extracted by the improved Canny edge detection algorithm, and the geometric model of the material is further generated. During edge detection, the system adopts a dual threshold mechanism, setting the high threshold to 0.7 and the low threshold to 0.4 to ensure that the noise is reduced while maintaining edge accuracy. Finally, in terms of model parameters, the generated material model should include key parameters such as length, width, and thickness. Combined with the material density information, the actual density of the material is calculated using the weight sensor data fed back by the sensor in real time (the error does not exceed 1g), and then marked in the model.

[0043] B22: The second step is to construct the path cost matrix. The cost matrix is ​​generated by combining the geometric model, density, hardness and other properties of the material. The cost function is as follows:

[0044] C(i,j)=α·D(i,j)+β·H(i,j)+γ·V(i,j)+δ·P(i,j)

[0045] in:

[0046] C(i,j): The total cost from position i to position j.

[0047] D(i,j): The distance between two points in millimeters.

[0048] H(i,j): Change in hardness of the material, measured by a hardness sensor, in megapascals (MPa).

[0049] V(i,j): speed of cutting tool in mm / s.

[0050] P(i,j): The pressure of the cutting tool, measured by a pressure sensor, in Newton (N).

[0051] α, β, γ, δ: weights of each item. This method uses default values ​​of 0.25, 0.2, 0.3, and 0.25 respectively, which can be dynamically adjusted according to the properties of different materials.

[0052] B23: The third step is to generate the initial path through dynamic programming. After constructing the path cost matrix, a recursive search is performed through the dynamic programming algorithm (DP) to find the path with the lowest cost from the starting point to the end point. The specific process is as follows:

[0053] Initial path generation: The system first generates a default cutting path as the initial solution. At this time, the default speed of the cutting tool is 50 mm / s, and the cutting depth is adjusted according to the material density and thickness.

[0054] Dynamically adjust the path: After the initial path is generated, the system dynamically adjusts the path based on real-time feedback from sensor data (such as cutting pressure, material thickness, etc.). If a deviation of more than 10% in cutting pressure or a change of ±0.05 mm in material thickness is detected, the system will immediately adjust the cutting speed or path curve.

[0055] B3: Improve material utilization and cutting accuracy through optimization algorithms. The specific steps are as follows:

[0056] B31: Enhanced Genetic Algorithm (EGA) is used to evaluate the adaptability of each path plan, combining material utilization and molding time.

[0057] EGA (genetic) evaluates the adaptability of each path plan in path optimization, and calculates the error and time data of each path by combining material utilization and molding time through fitness function. The processing logic includes generating the initial path plan, constructing multiple material cutting plans to form a population, and performing fitness evaluation. Genetic operation selects excellent individuals for mating and mutation through the tournament selection mechanism to generate new path plans. EGA continuously improves the material utilization of the path through iterative optimization to avoid falling into the local optimum.

[0058] In this method, the initial population consists of 20 paths, each of which is generated based on different material properties and geometric shapes. The fitness function of the path in this method is evaluated by material utilization, cutting time and cutting accuracy. The fitness function is as follows:

[0059]

[0060] in:

[0061] U: Material utilization rate, ranging from 0% to 100%.

[0062] T: cutting time, in seconds.

[0063] E: Cutting error, in millimeters. The default error is controlled within ±0.05 mm.

[0064] w1,w2,w3: fitness weights, 0.4, 0.35, 0.25 respectively.

[0065] Then the genetic operation is performed. This method uses the parameter settings of crossover rate 0.8 and mutation rate 0.05, and iterates through the tournament selection mechanism (selecting 4 path solutions in each round). After each round of iteration, the best path solution is selected to be retained, and the suboptimal path solution is fine-tuned to avoid local optimality.

[0066] Subsequently, the error and time data for each path plan are calculated based on the fitness function to reflect the comprehensive performance of the path and provide an evaluation criterion for path selection and genetic operations.

[0067]

[0068] Among them, is the fitness value of the path plan; is the material utilization rate of the path plan; is the molding time of the path plan.

[0069] B32: Genetic operation: First, select excellent individuals for mating based on the fitness evaluation results calculated in the previous step, and select individuals with high fitness through the tournament selection mechanism. The selection formula is as follows:

[0070]

[0071] Among them, P(i) is the probability of selecting an individual; ∑f(i) is the sum of the fitness of all individuals.

[0072] Then, crossover is performed between different paths according to the selection results of the previous step to generate new individuals. A new path plan is generated through the crossover operation. The crossover formula is as follows:

[0073] C k =αP i +(1-α)P j

[0074] Among them, C k is a new individual generated by crossover; P i and P j is the parent individual; α is the crossover coefficient.

[0075] Furthermore, some of the paths generated by crossover are mutated to improve the paths by introducing small random disturbances, increase the diversity of path solutions, and avoid local optimality. The mutation formula is as follows:

[0076] M i =P i +σN(0,1)

[0077] Among them, M i is the individual after mutation; σ is the mutation intensity; N(0,1) is the standard normal distribution.

[0078] Finally, continue to evaluate the fitness of these solutions, and perform elite retention operations based on the evaluation results to retain the path solution with the highest fitness to ensure that the optimal path is inherited. Use the simulated annealing method (SA), set the initial temperature T0, control the probability of accepting inferior solutions, and gradually reduce the temperature parameters of the system to make the path optimization process more stable.

[0079] At the same time, in order to enhance the stability of path optimization, simulated annealing (SA) operation is performed. SA is combined with genetic algorithm to control the probability of accepting inferior solutions and set the initial temperature T o , gradually reducing the temperature parameters to achieve global stability of path optimization. Its role is to improve the global optimality of the path and ensure the optimal performance of the final path solution. By gradually cooling the system temperature, SA effectively avoids falling into the local optimal solution and improves the global optimization effect of the path.

[0080]

[0081] P (accept) is the probability of accepting an inferior solution; △E is the energy difference between the current solution and the candidate solution; k is the Boltzmann constant; T is the current temperature.

[0082] In this process, a dynamic temperature update method is further adopted for system annealing. The temperature is updated according to the set cooling rate, and the system temperature is slowly reduced to make the path stable.

[0083] T new =α·T old

[0084] Among them, T new is the updated temperature; T old is the current temperature; α is the cooling coefficient.

[0085] When the temperature is lower than the set threshold or the path is no longer improved, the step is terminated. In order to further improve the global optimization effect, this method combines the SA algorithm and gradually cools the system through the initial temperature T_0 = 1000K to reduce the probability of falling into the local optimum during the iteration process. The cooling rate is 0.95, and the optimization is terminated when the system temperature drops to 10K.

[0086] C: Real-time monitoring

[0087] Use position sensors and pressure sensors to monitor the status of materials and machine tools.

[0088] Perform monitoring data processing. Aggregate data from different sensors to form a unified data stream. Integrate multi-source data, synchronize time and align space to form a complete description of material and equipment status. Perform anomaly detection and analyze abnormal changes in data streams in real time. By analyzing sensor data, identify and record abnormal events and detect abnormal conditions (such as excessive equipment pressure, excessive current, high temperature, equipment vibration).

[0089] Detect anomalies using moving average and standard deviation:

[0090]

[0091] in, is the moving average, σ is the standard deviation, and k is the threshold constant.

[0092] Feedback and Adjustment:

[0093] Collect data such as material feeding speed, equipment pressure, temperature, etc. in real time, analyze the real-time data, and identify anomalies and deviations.

[0094] E(t)=w s S(t)+w p P(t)+w t T(t)

[0095] Where E(t) is the error; w s ,w p ,w t is the weight coefficient.

[0096] D: Path verification and adjustment

[0097] D1: Virtual simulation: First, test the optimized cutting path in a computer simulation environment, evaluate its feasibility, conduct accuracy assessment, and analyze the material utilization of the path.

[0098] D2: Real-time adjustment: During the actual material cutting process, the path execution is monitored in real time, deviations and errors are recorded, and the material cutting path and parameters are adjusted dynamically.

[0099] New path = current path + feedback deviation

[0100] The actual processing steps in the production process are: in wood processing, optimize the path to produce building wood materials, monitor in real time and make dynamic adjustments to improve material utilization; in metal accessories processing, in the production process of metal products, combine genetic algorithms and simulated annealing technology to improve cutting accuracy, reduce material waste, and ensure maximum material utilization. When the cutting pressure changes by more than ±10 Newtons or the position deviation exceeds ±0.1 mm, the system will immediately adjust the cutting parameters to ensure maximum material utilization and cutting accuracy.

[0101] Example 1

[0102] A method for cutting refined raw materials of building wood:

[0103] Through step A, the sampling interval of the sensor for wood is set according to the speed and accuracy requirements of the production line, real-time data collection is performed, and then a geometric digital model of the target product is established.

[0104] Through step B, the size and shape of the wood are processed, and then the wood surface image is captured in real time, and the key features are extracted to generate a geometric digital model of the current wood. The geometric digital model of the wood is then combined with the data collected by the sensor in real time, and an initial path is generated through an algorithm. At the same time, the geometric digital model of the current wood and the data collected by the sensor in real time are optimized in detail to ensure the accuracy of the path and the utilization rate of the wood. Finally, the optimization algorithm is used to improve the utilization rate of the wood and the cutting accuracy.

[0105] Through step C, position sensors and pressure sensors are used to monitor the status of wood and machine tools, and data from different sensors are aggregated to form a unified data stream. Multi-source data is integrated, time synchronization and spatial alignment are performed to form a complete description of the status of wood and equipment, and anomaly detection is performed to analyze abnormal changes in the data stream in real time. By analyzing sensor data, excessive equipment pressure, excessive current, high temperature, and equipment vibration are detected, and abnormal events are identified and recorded.

[0106] Through step D, the optimized cutting path is tested in a computer simulation environment to evaluate its feasibility and accuracy, analyze the utilization of path materials, track the processing status of materials through the real-time monitoring module, monitor the path execution in real time, record deviations and errors, and dynamically adjust the cutting path.

[0107] Compared with traditional cutting, this method collects data in real time and combines it with geometric modeling to dynamically generate and optimize the cutting path. Different from the fixed path of the existing technology, this method can dynamically adjust the cutting path according to real-time data changes, greatly improving the utilization rate of materials.

[0108] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Any ordinary technician in the industry can smoothly implement the present invention as shown in the specification and described above. However, any equivalent changes made by technicians familiar with the profession without departing from the scope of the technical solution of the present invention using the technical content disclosed above are all equivalent embodiments of the present invention. At the same time, any equivalent changes, modifications and evolutions made to the above embodiments based on the essential technology of the present invention are within the protection scope of the technical solution of the present invention.

Claims

1. A method for cutting fine raw materials for construction, characterized in that: The steps include: A: Build a data acquisition and modeling module to collect material data through sensors and input or build a geometric digital model of the building product; B: Set up a path generation module to generate a feasible material cutting path plan, generate a geometric digital model of the material, and optimize the geometric digital model of the material, the data model collected by the sensor in real time, and the geometric digital model of the building product through the principal component analysis algorithm KCPA based on the kernel function; optimize the material cutting path plan based on the above-mentioned optimized geometric digital model of the material, the data model collected by the sensor in real time, and the geometric digital model of the building product; C: Set up a real-time monitoring module, which summarizes the status of materials and machine tools monitored by position sensors, pressure sensors, current sensors, temperature sensors, and vibration sensors to form a unified data stream, perform time synchronization and spatial alignment, form a complete description of material and equipment status, and perform abnormality detection; D: Build a verification and adjustment module, the verification module tests the optimized material cutting path plan in a computer simulation environment to obtain feasible parameters of the optimized material cutting path plan; obtain feedback information of the feasible parameters of the optimized material cutting path plan through the verification module; the real-time adjustment module adjusts the material cutting path plan according to the feedback information of the verification module.

2. The method for cutting refined raw materials for construction according to claim 1, characterized in that: The data acquisition and modeling module annotates material properties, adds density and weight attribute labels to each material, and creates a record for each material in the database.

3. The method for cutting refined raw materials for construction according to claim 2, characterized in that: The path generation module removes high frequency noise in the measurement data by fast Fourier transform (FFT).

4. The method for cutting refined raw materials for construction according to claim 3, characterized in that: The path generation module uses the Canny edge detection algorithm to perform image gradient detection, extract edges and key features, and generate a geometric digital model of the material.

5. The method for cutting refined raw materials for construction according to claim 4, characterized in that: The path generation module improves the accuracy of edge detection through non-maximum suppression and dual threshold detection, and generates an accurate material geometry digital model.

6. The method for cutting refined raw materials for construction according to claim 5, characterized in that: The path generation module generates an initial path through dynamic programming, and after constructing a path cost matrix, performs a recursive search through a dynamic programming algorithm DP.

7. The method for cutting refined raw materials for construction according to claim 6, characterized in that: The path generation module adopts enhanced genetic algorithm (EGA) and combines material utilization and molding time to evaluate the adaptability of each path solution.

8. The method for cutting refined raw materials for construction according to claim 7, characterized in that: The genetic operation in the path generation module selects excellent individuals for mating and mutation through a tournament selection mechanism to generate new path plans, and avoids falling into local optimality through iterative optimization.

9. The method for cutting refined raw materials for construction according to claim 8, characterized in that: The path generation module controls the probability of accepting inferior solutions through simulated annealing SA operation combined with genetic algorithm, so as to enhance the stability of path optimization.

10. The method for cutting refined raw materials for construction according to claim 9, characterized in that: The simulated annealing SA operation in the path generation module further adopts a dynamic temperature update method.

Images