Method and equipment for regenerating foundry waste sand

Abstract

The invention discloses a regeneration method and equipment for foundry waste sand. The method comprises the following steps: filling the foundry waste sand into a sealed container; a steam valve of the sealed container is opened, a pressure release valve of the sealed container is closed, water vapor is added into the sealed container for pressurization treatment, and when the pressure intensity in the sealed container reaches 0.5-5 MPa and the temperature reaches 150-260 DEG C, adding of the water vapor is stopped; closing a steam valve of the sealed container, and keeping constant temperature and constant pressure in the sealed container for 5-20 minutes; opening a pressure release valve of the sealed container to carry out pressure reduction treatment, and collecting sand in the sealed container when the pressure intensity in the sealed container is reduced to normal pressure; and carrying out drying separation and winnowing treatment on the sand to obtain regenerated sand. Therefore, the foundry waste sand can be regenerated and recycled, so that resource waste and environmental pollution are avoided.

Description

Technical Field

[0001] The present application relates to the field of waste sand regeneration, and more specifically, to a method and device for regenerating foundry waste sand. Background Art

[0002] A large amount of waste sand is generated in foundry enterprises, and most of the waste sand will be discarded, which will cause environmental pollution on the one hand and waste of resources on the other hand. For the sustainable development of the foundry industry, it is necessary to recycle the foundry waste sand to avoid waste of resources and environmental pollution.

[0003] At present, the public regeneration methods of foundry waste sand mainly include dry mechanical regeneration, thermal regeneration and wet mechanical regeneration. The dry mechanical regeneration method mainly uses centrifugal and grinding methods to make the material impact and rub to remove the impurity film on the surface of the waste sand to achieve the purpose of regeneration. However, the removal rate of the impurity film by dry mechanical regeneration does not exceed 80%, which will greatly reduce the recovery rate of regenerated sand. The surface of the regenerated sand will have cracks, and the acid consumption value of the regenerated sand is high, which cannot meet the technical quality requirements of foundry sand. In addition, the regeneration process causes great wear on the sand and produces a large amount of fine powder, which puts great technical pressure on secondary processing and utilization.

[0004] The thermal regeneration method is to roast the waste sand at 700-800℃ to burn out the impurities on the surface of the silica sand to achieve the purpose of film removal and regeneration. Although the thermal regeneration method has a good removal effect on flammable organic impurity films such as phenolic resin and cold box resin, it is not effective in removing inorganic impurity films such as clay and water glass, and the tail gas after roasting needs to be purified.

[0005] The wet mechanical regeneration method uses water as a medium to soak and scrub the waste sand, thereby removing the impurity film on the surface of the waste sand to achieve the purpose of film removal and regeneration. The wet mechanical regeneration method has a good removal effect on water-soluble inorganic impurity films such as clay and water glass, but it is not effective in removing organic impurity films such as phenolic resin and cold box resin. In addition, the subsequent sand, water and impurities need to be further separated, and even the wastewater generated in the production process needs to be treated.

[0006] Therefore, an optimized recycling solution for used foundry sand is desired. Summary of the invention

[0007] In order to solve the above technical problems, the present application is proposed. The embodiments of the present application provide a method and device for regenerating foundry waste sand, which can regenerate and recycle foundry waste sand to avoid resource waste and environmental pollution.

[0008] According to one aspect of the present application, a method for regenerating waste foundry sand is provided, comprising:

[0009] Put the waste foundry sand into a sealed container;

[0010] Open the steam valve of the sealed container and close the pressure relief valve of the sealed container, and add water vapor to the sealed container for pressurization treatment until the pressure in the sealed container reaches 0.5-5MPa and the temperature reaches 150-260°C, and then stop adding water vapor;

[0011] Close the steam valve of the sealed container and maintain the temperature and pressure in the sealed container for 5-20 minutes;

[0012] Opening the pressure relief valve of the sealed container to reduce the pressure, and collecting the sand in the sealed container when the pressure in the sealed container drops to normal pressure; and

[0013] The sand is dried, separated and air-selected to obtain regenerated sand.

[0014] According to another aspect of the present application, there is provided a regeneration device for waste foundry sand, comprising:

[0015] A foundry waste sand loading module, used for loading the foundry waste sand into a sealed container;

[0016] A water vapor adding module, used to open the steam valve of the sealed container and close the pressure relief valve of the sealed container, and add water vapor to the sealed container for pressurization treatment until the pressure in the sealed container reaches 0.5-5MPa and the temperature reaches 150-260°C, and then stop adding water vapor;

[0017] A constant temperature and pressure maintaining module, used to close the steam valve of the sealed container and maintain a constant temperature and pressure in the sealed container for 5-20 minutes;

[0018] a depressurization module, used for opening the pressure relief valve of the sealed container to perform depressurization, and collecting the sand in the sealed container when the pressure in the sealed container drops to normal pressure; and

[0019] The drying and air separation module is used for drying, separating and air separation of the sand to obtain regenerated sand.

[0020] Compared with the prior art, the present application provides a method and device for regenerating foundry waste sand, which comprises: loading foundry waste sand into a sealed container; opening the steam valve of the sealed container and closing the pressure relief valve of the sealed container, and adding water vapor to the sealed container for pressurization treatment until the pressure in the sealed container reaches 0.5-5MPa and the temperature reaches 150-260°C, then stopping the addition of water vapor; closing the steam valve of the sealed container, and maintaining the constant temperature and pressure in the sealed container for 5-20 minutes; opening the pressure relief valve of the sealed container for pressure reduction treatment, and collecting the sand in the sealed container when the pressure in the sealed container drops to normal pressure; and drying, separating and air separation the sand to obtain regenerated sand. In this way, foundry waste sand can be recycled to avoid waste of resources and environmental pollution. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. The following drawings are not intentionally scaled to the actual size, and the focus is on illustrating the main purpose of the present application.

[0022] Figure 1 The present invention is a flowchart of a method for regenerating waste foundry sand according to an embodiment of the present application.

[0023] Figure 2 This is a flow chart of sub-step S150 of the method for regenerating waste foundry sand according to an embodiment of the present application.

[0024] Figure 3 Schematic diagram of the architecture of sub-step S150 of the method for regenerating waste foundry sand according to an embodiment of the present application.

[0025] Figure 4 This is a flowchart of sub-step S152 of the method for regenerating waste foundry sand according to an embodiment of the present application.

[0026] Figure 5 This is a flowchart of sub-step S1522 of the method for regenerating waste foundry sand according to an embodiment of the present application.

[0027] Figure 6 It is a block diagram of a regeneration device for waste foundry sand according to an embodiment of the present application.

[0028] Figure 7 This is an application scenario diagram of the method for regenerating waste foundry sand according to an embodiment of the present application. DETAILED DESCRIPTION

[0029] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0030] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0031] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0032] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0033] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0034] In order to achieve sustainable development of the foundry industry, it is necessary to recycle and reuse foundry waste sand to avoid resource waste and environmental pollution. At present, the public foundry waste sand regeneration methods mainly include dry mechanical regeneration method, thermal regeneration method and wet mechanical regeneration method. Although there are many types of waste sand regeneration methods nowadays, there are many problems in these methods. Therefore, an optimized regeneration scheme for foundry waste sand is desired.

[0035] Specifically, in the technical solution of the present application, a method for regenerating foundry waste sand is proposed. Figure 1 Flow chart of the method for regenerating waste foundry sand according to an embodiment of the present application, as shown in Figure 1As shown, the process includes the following steps: S110, loading foundry waste sand into a sealed container; S120, opening the steam valve of the sealed container and closing the pressure relief valve of the sealed container, and adding water vapor to the sealed container for pressurization treatment, until the pressure in the sealed container reaches 0.5-5MPa and the temperature reaches 150-260°C, and then stopping adding water vapor; S130, closing the steam valve of the sealed container, and maintaining a constant temperature and pressure in the sealed container for 5-20min; S140, opening the pressure relief valve of the sealed container for pressure reduction treatment, and collecting the sand in the sealed container when the pressure in the sealed container drops to normal pressure; and, S150, drying, separating and air separation the sand to obtain regenerated sand.

[0036] Further, in step S110, the sealed container into which the foundry waste sand is loaded needs to be able to withstand high pressure and high temperature, and before loading the foundry waste sand into the container, it should be ensured that the inside of the container is clean and tidy without any impurities. The amount of foundry waste sand can be determined according to the size of the container and the amount of regenerated sand required. After loading the sand, the sealed container is tightly closed to ensure that the subsequent pressurization and depressurization processes can proceed normally.

[0037] Further, in step S120, the steam valve of the sealed container is opened, and water vapor is added to the container for pressurization treatment. During the pressurization process, it is necessary to monitor the pressure and temperature in the container to ensure that the pressure and temperature in the container that need to be monitored reach the required range. The purpose of pressurization is to allow water vapor to penetrate into the interior of the foundry waste sand so that it can fully absorb water vapor. This process generally takes from tens of minutes to several hours, depending on the size of the container and the desired treatment effect. In a specific example of the present application, the addition of water vapor is stopped until the pressure in the sealed container reaches 0.5-5MPa and the temperature reaches 150-260°C.

[0038] Further, in step S130, after the pressurization process is completed, the steam valve of the sealed container is closed, and the pressure and temperature in the container are kept constant. This process generally takes 5-20 minutes. In a specific example of the present application, the sealed container needs to maintain constant temperature and pressure for 5-20 minutes. During this process, the temperature and pressure in the container need to be monitored to ensure that the temperature and pressure in the monitored container remain constant.

[0039] Further, in step S140, the pressure relief valve of the sealed container is opened, and the pressure in the container begins to gradually decrease. When the pressure drops to normal pressure, the sand in the sealed container is collected. In this process, it is necessary to monitor the pressure and temperature in the container to ensure that the pressure and temperature in the monitored container can gradually decrease within a suitable range. Among them, the position of the pressure relief valve of the sealed container is preferably set at or near the highest point of the container, so as to ensure that when excessive pressure is generated inside the container, the pressure relief valve can discharge excess gas or steam in time, thereby ensuring that the pressure inside the container is always within a safe range.

[0040] Further, in step S150, after the sand is collected, it needs to be dried and separated and air-selected to obtain regenerated sand. Drying and separation can be carried out by placing the sand in a drying chamber, and air-selection can be carried out by screening the sand through a sieve. During the drying process, it is necessary to ensure that the sand can be fully dried so as not to affect the quality of the regenerated sand. During the air-selection process, it is necessary to ensure that the sand can be fully screened to remove impurities and particles that do not meet the requirements.

[0041] It is worth mentioning that the monitoring of the pressure and temperature in the container can be achieved using sensors. In one example, for the monitoring of pressure, a pressure sensor can be used, for example, various types of pressure sensors such as piezoresistive, capacitive, and inductive can be used. Among them, the piezoresistive sensor is a common pressure sensor, and the magnitude of the pressure can be determined by measuring the change in the resistance value. In a specific implementation, the piezoresistive sensor can be installed on the wall of the container, and the pressure data can be transmitted to the monitoring system through the sensor, thereby realizing real-time monitoring of the pressure in the container. In one example, for the monitoring of temperature, a temperature sensor can be used. For example, various types of temperature sensors such as thermocouples, thermistors, and infrared temperature sensors can be used. Among them, the thermocouple is a common temperature sensor, and the magnitude of the temperature can be determined by measuring the temperature difference at the contact point of two different metals. In a specific implementation, the thermocouple can be installed inside the container, and the temperature data can be transmitted to the monitoring system through the sensor, thereby realizing real-time monitoring of the temperature in the container.

[0042] It is worth mentioning that regarding drying and detachment, an oven can also be used as a drying equipment. Specifically, first, evenly distribute the treated foundry waste sand into the oven; then, start the oven and set the temperature to an appropriate value, which can generally be set between 150-200 degrees Celsius; then, close the oven door and open the oven vents to allow moisture to drain; then, start heating, which generally takes several hours, depending on the size of the oven and the temperature setting; finally, after the drying is completed, turn off the oven, wait for the oven to cool down, and then take out the treated foundry waste sand. It should be noted that the temperature and time can be appropriately adjusted according to the amount of foundry waste sand to ensure the treatment effect and product quality.

[0043] It is worth mentioning that for air separation, a screen or screening equipment can be used. In one example, a screen is used for processing. First, a suitable screen is prepared and different screen hole sizes are selected according to needs; then, the dried and separated sand is poured into the screen and the screen is gently shaken to distribute the sand naturally; then, during the screening process, the angle and shaking force of the screen can be adjusted as needed to better separate the sand; finally, after the screening is completed, the regenerated sand that has been screened out is collected, and the impurities and particles that do not meet the requirements on the screen are cleaned up.

[0044] It should be noted that during the implementation process, it is necessary to select drying equipment and screening equipment suitable for the specific application scenario, and to reasonably install and debug the equipment to ensure the processing effect and product quality.

[0045] Accordingly, in the actual regeneration and preparation process of foundry waste sand, the mixture of impurity film and sand that has been broken into fine particles needs to be dried, separated and air-selected, so as to use the wind force and the residual heat of the sand surface to boil and separate the impurity film on the sand surface. In this process, increasing the wind force can enhance the friction between the impurity film and the sand, thereby improving the separation efficiency of the impurity film. However, if the wind force is increased blindly, it will also cause a waste of wind resources. Moreover, in the early stage of air selection, most of the plasma film is not separated from the sand, and more wind resources are needed at this time. As time goes by, the plasma film and the sand will gradually separate, and the demand for wind resources will decrease.

[0046] Based on this, in the technical solution of the present application, during the air separation process, it is expected to judge the air separation stage based on the analysis of the friction strength between the impurity film and the sand, so as to adaptively perform intelligent control of the wind strength. Considering that the friction strength between the impurity film and the sand is difficult to detect during the actual air separation process, and the change in friction strength will cause the sound signal to change, and also considering that during the actual air separation, there is a time-series correlation between the fan speed value and the detected sound signal. Therefore, the friction strength is represented by the detected sound signal, and the wind strength is represented by the fan speed value. Through the time-series analysis of the detected sound signal and the fan speed value, the current fan speed value is adaptively adjusted based on the time-series change characteristics of the fan speed value and the time-series distribution characteristics of the detected sound signal, so as to optimize the efficiency and quality of the waste sand regeneration process while avoiding energy waste.

[0047] Accordingly, the steps of drying, separating and air separation the sand to obtain regenerated sand are further described. Figure 2 is a flow chart of sub-step S150 of the method for regenerating waste foundry sand according to an embodiment of the present application, Figure 3 FIG. 1 is a schematic diagram of the structure of the method for regenerating waste foundry sand according to an embodiment of the present application. Figure 2 and Figure 3 As shown, the sand is dried, separated and air-selected to obtain regenerated sand, including: S151, obtaining the fan speed values ​​at multiple predetermined time points within a predetermined time period, and the detection sound signal of the predetermined time period; S152, performing time series feature analysis on the fan speed values ​​at the multiple predetermined time points and the detection sound signal of the predetermined time period to obtain a mapping associated feature vector; and, S153, based on the mapping associated feature vector, determining whether the fan speed value at the current time point should be increased or decreased.

[0048] More specifically, in step S151, the fan speed values ​​at multiple predetermined time points within a predetermined time period and the detection sound signal of the predetermined time period are obtained. It should be understood that the air separation process is to accelerate the friction between the impurity film and the sand by wind force to improve the detachment efficiency of the impurity film. The wind force required for the air separation process needs to be dynamically adjusted according to the current contact condition between the waste sand impurity film and the sand mixture. Therefore, in this process, it is very necessary to timely monitor and analyze the fan speed value, detection sound signal and other data, which can reflect the wind force of the fan and the change in the friction strength of the waste sand impurity film and the sand mixture, and can help determine the current required wind force and adjustment plan, thereby improving the preparation efficiency and quality of waste sand regeneration.

[0049] More specifically, in step S152, the fan speed values ​​at the multiple predetermined time points and the detection sound signal of the predetermined time period are subjected to time series feature analysis to obtain a mapping correlation feature vector. Considering that for the detection sound signal, since the detection sound signal is a time domain signal, although its dominance in time correlation is more intuitive, the detection sound signal is weak and easily interfered by external noise, resulting in low accuracy in extracting the implicit features of the friction intensity of the impurity film and sand in the detection sound signal, thereby affecting the adaptive control accuracy of the fan speed value. Since the characteristics of the frequency domain signal are different from those of the time domain signal, the detection sound signal is converted into the frequency domain, and the friction intensity feature information can be determined by the implicit feature distribution information of the detection sound signal in the frequency domain, but it is not intuitive in the feature dominance of the detection sound signal, and the temporal correlation feature information is ignored. Therefore, in the technical solution of the present application, the extraction of friction intensity features and the adaptive regulation of the fan speed value are performed by combining the implicit features of the detection sound signal in the time domain and the frequency domain.

[0050] Accordingly, in a specific example, Figure 4 As shown, the fan speed values ​​at the plurality of predetermined time points and the detection sound signal of the predetermined time period are subjected to time series feature analysis to obtain a mapping association feature vector, including: S1521, the detection sound signal is subjected to frequency domain feature analysis based on Fourier transform to obtain a plurality of sound frequency domain feature statistics; S1522, the plurality of sound frequency domain feature statistics and the waveform diagram of the detection sound signal are subjected to associated feature coding to obtain an optimized detection sound feature matrix; S1523, the fan speed values ​​at the plurality of predetermined time points are subjected to feature extraction based on a deep neural network model as a fan speed time series feature extractor to obtain a fan speed time series feature vector; and, S1524, the mapping association feature vector is obtained based on the responsiveness association between the fan speed time series feature vector and the detection sound feature matrix. Considering that the particle size of the foundry waste sand is not fixed, the friction strength between the sand of different particle sizes and the impurity film is also different, which is manifested in the detection sound signal with different feature distribution information, and such feature information is not intuitive in the detection sound signal. Therefore, in the technical solution of the present application, the detected sound signal is further subjected to a frequency domain feature analysis based on Fourier transform to obtain a plurality of sound frequency domain feature statistical values, so as to fully capture the friction intensity characteristic information between the waste sand and the impurity film at different particle sizes based on the characteristic correlation between the plurality of sound frequency domain feature statistical values.

[0051] Furthermore, for the fan speed values ​​at the multiple predetermined time points, since the fan speed values ​​have a time-series variation rule in the time dimension, that is, the fan speed values ​​at the respective predetermined time points have a time-series correlation relationship. Therefore, the fan speed values ​​at the multiple predetermined time points are further arranged as a fan speed time-series input vector according to the time dimension, and then feature mining is performed in a time-series feature extractor based on a one-dimensional convolutional neural network model to extract the time-series correlation feature information of the fan speed values ​​in the time dimension, so as to obtain a fan speed time-series feature vector.

[0052] Then, the responsiveness transfer of the fan speed timing feature vector relative to the detection sound feature matrix is ​​calculated to obtain a mapping correlation feature vector, so as to represent the correlation feature information between the timing correlation feature of the fan speed value and the timing implicit feature of the detection sound signal, that is, the characteristic information about the friction intensity between the impurity film and the sand at different fan speeds. Accordingly, in a specific example, based on the responsiveness association between the fan speed timing feature vector and the detection sound feature matrix to obtain the mapping correlation feature vector, it includes: calculating the responsiveness transfer of the fan speed timing feature vector relative to the detection sound feature matrix to obtain the mapping correlation feature vector. It is worth mentioning that responsiveness transfer refers to the relationship between the input signal and the output signal during signal processing, and the responsiveness transfer refers to mapping the fan speed timing feature vector and the detection sound feature matrix to obtain a mapping correlation feature vector, which is used to represent the correlation feature information between the fan speed and the detection sound signal. This mapping relationship can be learned by machine learning methods, such as a neural network model, to capture the complex relationship between the fan speed and the detection sound signal. At different fan speeds, due to the different friction strengths between the impurity film and the sand, the temporal implicit characteristics of the detection sound signal will also change. Therefore, it is necessary to represent this correlation feature information through responsive transfer.

[0053] Accordingly, in a specific example, Figure 5As shown, the multiple sound frequency domain feature statistics and the waveform of the detection sound signal are associated with feature coding to obtain an optimized detection sound feature matrix, including: S15221, the multiple sound frequency domain feature statistics and the waveform of the detection sound signal are passed through a CLIP model including a waveform encoder and a sequence encoder to obtain a detection sound feature matrix; S15222, the sequence feature vector and the image feature vector are subjected to spatial adaptive point learning on a non-homogeneous Hilbert surface to obtain a fused feature vector; S15223, the fused feature vector is calculated and multiplied by its own transpose to obtain a fused feature matrix; and, S15224, the fused feature matrix and the detection sound feature matrix are fused to obtain an optimized detection sound feature matrix.

[0054] In order to further improve the adequacy of capturing the implicit feature information about the friction strength between the sand and the impurity film in the detection sound signal, so as to improve the control accuracy of the fan speed value, the CLIP model including the waveform encoder and the sequence encoder is further used to complete the feature mining fusion of the time domain and frequency domain of the detection sound signal. That is, specifically, the waveform of the detection sound signal is processed through the waveform encoder of the CLIP model, so as to extract the time domain implicit feature distribution information about the detection sound signal in the waveform of the detection sound signal through the convolutional neural network model of the waveform encoder as a filter, so as to obtain the detection sound waveform feature vector. Then, the multiple sound frequency domain feature statistics are encoded through the sequence encoder of the CLIP model to obtain the sound frequency domain feature vector. Here, the sequence encoder uses a one-dimensional convolutional neural network model to perform feature mining of the multiple sound frequency domain feature statistics to extract the associated feature information between the various sound frequency domain feature statistics. Furthermore, the acoustic frequency domain feature vector and the detection sound waveform feature vector are fused to optimize the time domain distribution characteristics of the detection sound signal based on the correlation characteristics between the various acoustic frequency domain feature statistical values, so as to extract the temporal implicit feature distribution information between the friction intensity of the impurity film and the sand, thereby obtaining the detection sound feature matrix.

[0055] In particular, in the technical solution of the present application, the classification feature vector is obtained by calculating the responsive transfer of the fan speed timing feature vector relative to the detection sound feature matrix, taking into account the domain transfer characteristics of the waveform image semantic characteristics of the detection sound signal under the sequence characteristic constraints of the time-series local correlation characteristics of the fan speed value expressed by the classification feature vector as the fan speed timing feature vector relative to the sound frequency domain feature statistics expressed by the detection sound feature matrix, if the association effect of the detection sound feature matrix on the sequence characteristics of the sound frequency domain feature statistics and the waveform image semantic characteristics of the detection sound signal can be improved, the expression effect of the classification feature vector can be improved.

[0056] Here, when the detection sound feature matrix is ​​obtained by passing the multiple sound frequency domain feature statistics and the waveform of the detection sound signal through the CLIP model including a waveform encoder and a sequence encoder, the sequence feature vector obtained by the sequence encoder for the multiple sound frequency domain feature statistics and the image feature vector obtained by the waveform encoder for the detection sound signal are point-by-point associated by position to obtain the detection sound feature matrix. Therefore, there is a non-homogeneous point-by-point correspondence between the sequence feature vector and the image feature vector, that is, the sequence feature vector expresses the sample-to-sample association feature of the sound frequency domain feature statistics based on a one-dimensional convolution scale, while the image feature vector expresses the local image semantic association feature of the waveform of the detection sound signal based on a two-dimensional convolution scale. Therefore, the sequence feature vector, for example, denoted as V1, and the image feature vector, for example, denoted as V2, are first subjected to spatial adaptive point learning on a non-homogeneous Hilbert surface to obtain a fused feature vector, for example, denoted as V r .

[0057] Accordingly, in a specific example, spatial adaptive point learning on a non-homogeneous Hilbert surface is performed on the sequence feature vector and the image feature vector to obtain a fused feature vector, including:

[0058] Performing spatial adaptive point learning on a non-homogeneous Hilbert surface on the sequence feature vector and the image feature vector using the following fusion formula to obtain the fused feature vector;

[0059] Among them, the fusion formula is:

[0060]

[0061]

[0062]

[0063] Wherein, V1 represents the sequence feature vector, V2 represents the image feature vector, p1≠p2, and represents the non-homogeneous Mind-type distance based on Gilbert space, and p1 and p2 are hyperparameters, and are the global feature means of feature vectors V1 and V2, respectively, and both feature vectors V1 and V2 are row vectors, d1 represents the first distance, d2 represents the second distance, Cov represents convolution processing, represents positional addition, ⊙ represents positional dot multiplication, V r represents the fused feature vector.

[0064] In this way, by performing one-dimensional convolution on the vector point association between the sequence feature vector V1 and the image feature vector V2 using the non-homogeneous Gilbert space metric, the feature manifold represented by the high-dimensional features of the sequence feature vector V1 and the image feature vector V2 can be used for the non-axis alignment (non-axisalignment) characteristics in the high-dimensional feature space, and adaptive point learning can be performed on the surface space of the manifold convergence hyperplane based on the Hilbert space toward the hyperplane, and the aerial measurement (aerial measurement) oriented to the distribution convergence direction of each of the sequence feature vector V1 and the image feature vector V2 is used as a correction to improve the non-homogeneous point-by-point fusion between the sequence feature vector V1 and the image feature vector V2. Then, the fused feature vector V r After multiplying its own transpose to obtain a fusion feature matrix and further fusing it with the detection sound feature matrix, the point-by-point correlation fusion effect of the detection sound feature matrix for the sequence characteristics of the sound frequency domain feature statistics and the waveform image semantic characteristics of the detection sound signal can be improved, thereby improving the expression effect of the classification feature vector. In this way, the current fan speed value can be adaptively controlled in real time and accurately based on the actual friction between the impurity film and the sand, so as to optimize the efficiency and quality of the waste sand regeneration process while avoiding energy waste.

[0065] Accordingly, in a specific example, the deep neural network model is a one-dimensional convolutional neural network model. It should be understood that a one-dimensional convolutional neural network (1D CNN) is a deep learning model for processing one-dimensional data such as time series data or signal data. Compared with a traditional fully connected neural network, a one-dimensional convolutional neural network can automatically extract features from data through a combination of convolutional layers and pooling layers, thereby better capturing local and global patterns of data and having better generalization capabilities. A one-dimensional convolutional neural network is usually composed of a plurality of convolutional layers and pooling layers alternating, wherein the convolutional layer is used to extract features, and the pooling layer is used to reduce data dimensions and reduce model parameters. In the convolutional layer, the model extracts local features of the data by sliding a convolutional kernel of a fixed size, and the size and number of the convolutional kernels can be adjusted according to the complexity of the data and the needs of the model. In the pooling layer, the model reduces the dimension and compression features of the data by performing maximum or average pooling on the convolution results, thereby reducing the model complexity and overfitting risk.

[0066] More specifically, in step S153, based on the mapping associated feature vector, it is determined whether the fan speed value at the current time point should be increased or decreased. Accordingly, in a specific example, based on the mapping associated feature vector, it is determined whether the fan speed value at the current time point should be increased or decreased, including: passing the mapping associated feature vector through a classifier to obtain a classification result, and the classification result is used to indicate that the fan speed value at the current time point should be increased or decreased. That is, in the technical solution of the present application, the label of the classifier includes the fan speed value at the current time point should be increased (first label), and the fan speed value at the current time point should be decreased (second label), wherein the classifier determines which classification label the mapping associated feature vector belongs to by a soft maximum function. It is worth noting that the first label p1 and the second label p2 here do not contain artificially set concepts. In fact, in the training process, the computer model does not have the concept of "the fan speed value at the current time point should be increased or decreased". It only has two classification labels and the probability of the output feature under these two classification labels, that is, the sum of p1 and p2 is one. Therefore, the classification result that the fan speed value should be increased or decreased is actually converted into a binary class probability distribution that conforms to natural laws through the classification label. In essence, the physical meaning of the natural probability distribution of the label is used, rather than the linguistic text meaning of "the fan speed value at the current time point should be increased or decreased". It should be understood that in the technical solution of the present application, the classification label of the classifier is the control strategy label that the fan speed value at the current time point should be increased or decreased. Therefore, after obtaining the classification result, the fan speed value change at the current time point can be adaptively adjusted based on the classification result, so as to perform adaptive adjustment of the current fan speed value, so as to optimize the efficiency and quality of the waste sand regeneration process while avoiding energy waste.

[0067] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression and SVM are often used to solve binary classification problems. For multi-classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification, but this is prone to errors and inefficient. Commonly used multi-classification methods include the Softmax classification function.

[0068] In summary, the regeneration method of waste foundry sand based on the embodiment of the present application is explained, which can regenerate and recycle the waste foundry sand to avoid waste of resources and environmental pollution.

[0069] Figure 6 FIG. 1 is a block diagram of a foundry waste sand regeneration device 100 according to an embodiment of the present application. Figure 6 As shown, the regeneration equipment 100 of foundry waste sand according to the embodiment of the present application includes: a foundry waste sand loading module 110, which is used to load the foundry waste sand into a sealed container; a water vapor adding module 120, which is used to open the steam valve of the sealed container and close the pressure relief valve of the sealed container, and add water vapor to the sealed container for pressurization treatment until the pressure in the sealed container reaches 0.5-5MPa and the temperature reaches 150-260°C, and then stop adding water vapor; a constant temperature and pressure maintaining module 130, which is used to close the steam valve of the sealed container and maintain the constant temperature and pressure in the sealed container for 5-20min; a pressure reducing module 140, which is used to open the pressure relief valve of the sealed container for pressure reducing treatment, and when the pressure in the sealed container drops to normal pressure, collect the sand in the sealed container; and a drying and air separation module 150, which is used to dry, separate and air separate the sand to obtain regenerated sand.

[0070] In one example, in the above-mentioned foundry waste sand regeneration equipment 100, the drying and air separation module 150 includes: a data acquisition unit, used to obtain the fan speed values ​​at multiple predetermined time points within a predetermined time period, and the detection sound signal of the predetermined time period; a timing feature analysis unit, used to perform timing feature analysis on the fan speed values ​​at the multiple predetermined time points and the detection sound signal of the predetermined time period to obtain a mapping associated feature vector; and a speed value control unit, used to determine whether the fan speed value at the current time point should be increased or decreased based on the mapping associated feature vector.

[0071] Here, those skilled in the art will appreciate that the specific functions and operations of the various modules in the above-mentioned foundry waste sand regeneration equipment 100 have been described in detail above. Figures 1 to 5 The description of the regeneration method of the foundry waste sand has been described in detail, and therefore, its repeated description will be omitted.

[0072] As described above, the foundry waste sand regeneration device 100 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a foundry waste sand regeneration algorithm. In one example, the foundry waste sand regeneration device 100 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the foundry waste sand regeneration device 100 can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the foundry waste sand regeneration device 100 can also be one of the many hardware modules of the wireless terminal.

[0073] Alternatively, in another example, the foundry waste sand regeneration device 100 and the wireless terminal may also be separate devices, and the foundry waste sand regeneration device 100 may be connected to the wireless terminal via a wired and / or wireless network, and transmit interactive information in accordance with an agreed data format.

[0074] Figure 7 FIG. 1 is an application scenario diagram of the method for regenerating waste foundry sand according to an embodiment of the present application. Figure 7 As shown, in this application scenario, first, the fan speed values ​​at multiple predetermined time points within a predetermined time period are obtained (for example, Figure 7 D1 as shown in the figure), and the detection sound signal of the predetermined time period (for example, Figure 7 D2) as shown in FIG. Figure 7 The C shown in the figure is a sound detection sensor. Then, the fan speed values ​​at the plurality of predetermined time points and the detection sound signals of the predetermined time period are input into a server (for example, Figure 7 S) as shown in the figure, wherein the server is able to use the regeneration algorithm of the foundry waste sand to process the fan speed values ​​of the multiple predetermined time points and the detection sound signals of the predetermined time period to obtain a classification result indicating that the fan speed value at the current time point should be increased or decreased.

[0075] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the invention belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0076] The above is an explanation of the present invention and should not be considered as a limitation thereof. Although several exemplary embodiments of the present invention have been described, it will be readily appreciated by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present invention. Therefore, all such modifications are intended to be included within the scope of the present invention as defined in the claims. It should be understood that the above is an explanation of the present invention and should not be considered as being limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present invention is defined by the claims and their equivalents.

Claims

1. A method for regenerating foundry waste sand, characterized in that: include: Put the waste foundry sand into a sealed container; Open the steam valve of the sealed container and close the pressure relief valve of the sealed container, and add water vapor to the sealed container for pressurization treatment until the pressure in the sealed container reaches 0.5-5MPa and the temperature reaches 150-260°C, and then stop adding water vapor; Close the steam valve of the sealed container and maintain the temperature and pressure in the sealed container at a constant level for 5-20 minutes; Opening the pressure relief valve of the sealed container to reduce the pressure, and collecting the sand in the sealed container when the pressure in the sealed container drops to normal pressure; as well as The sand is dried, separated and air-selected to obtain regenerated sand.

2. The method for regenerating waste foundry sand according to claim 1, characterized in that: The sand is dried, separated and air-selected to obtain regenerated sand, comprising: Obtaining the fan speed values ​​at a plurality of predetermined time points within a predetermined time period, and the detection sound signal of the predetermined time period; Performing time series feature analysis on the fan speed values ​​at the plurality of predetermined time points and the detection sound signal in the predetermined time period to obtain a mapping associated feature vector; and Based on the mapping associated feature vector, it is determined whether the fan speed value at the current time point should be increased or decreased.

3. The method for regenerating waste foundry sand according to claim 2, characterized in that: Performing time series feature analysis on the fan speed values ​​at the plurality of predetermined time points and the detection sound signal in the predetermined time period to obtain a mapping associated feature vector includes: Performing frequency domain feature analysis based on Fourier transform on the detection sound signal to obtain a plurality of sound frequency domain feature statistical values; Performing correlation feature coding on the plurality of sound frequency domain feature statistics and the waveform diagram of the detection sound signal to obtain an optimized detection sound feature matrix; Based on the deep neural network model as a fan speed time series feature extractor, feature extraction is performed on the fan speed values ​​at the plurality of predetermined time points to obtain a fan speed time series feature vector; and The mapping association feature vector is obtained based on the responsiveness association between the fan speed time series feature vector and the detection sound feature matrix.

4. The method for regenerating waste foundry sand according to claim 3, characterized in that: The method of performing associated feature coding on the plurality of sound frequency domain feature statistics and the waveform diagram of the detection sound signal to obtain an optimized detection sound feature matrix comprises: Passing the plurality of audio frequency domain feature statistics and the waveform of the detection sound signal through a CLIP model including a waveform encoder and a sequence encoder to obtain a detection sound feature matrix; Perform spatial adaptive point learning on the non-homogeneous Hilbert surface on the sequence feature vector and the image feature vector to obtain a fused feature vector; Calculate the fused feature vector multiplied by its own transpose to obtain a fused feature matrix; and The fused feature matrix and the detected sound feature matrix are fused to obtain an optimized detected sound feature matrix.

5. The method for regenerating waste foundry sand according to claim 4, characterized in that: The sequence feature vector and the image feature vector are subjected to spatial adaptive point learning on the non-homogeneous Hilbert surface to obtain a fused feature vector, including: Performing spatial adaptive point learning on a non-homogeneous Hilbert surface on the sequence feature vector and the image feature vector using the following fusion formula to obtain the fused feature vector; Among them, the fusion formula is: Wherein, V1 represents the sequence feature vector, V2 represents the image feature vector, p1≠p2, and represents the non-homogeneous Mind-type distance based on Gilbert space, and p1 and p2 are hyperparameters, and are the global feature means of feature vectors V1 and V2, respectively, and both feature vectors V1 and V2 are row vectors, d1 represents the first distance, d2 represents the second distance, Cov represents convolution processing, represents positional addition, ⊙ represents positional multiplication, V r represents the fused feature vector.

6. The method for regenerating waste foundry sand according to claim 5, characterized in that: The deep neural network model is a one-dimensional convolutional neural network model.

7. The method for regenerating waste foundry sand according to claim 6, characterized in that: The mapping association feature vector is obtained based on the responsiveness association between the fan speed timing feature vector and the detection sound feature matrix, including: calculating the responsiveness transfer of the fan speed timing feature vector relative to the detection sound feature matrix to obtain the mapping association feature vector.

8. The method for regenerating waste foundry sand according to claim 7, characterized in that: Based on the mapping associated feature vector, determining whether the fan speed value at the current time point should be increased or decreased includes: passing the mapping associated feature vector through a classifier to obtain a classification result, and the classification result is used to indicate whether the fan speed value at the current time point should be increased or decreased.

9. A regeneration device for waste foundry sand, characterized in that: include: A foundry waste sand loading module, used for loading the foundry waste sand into a sealed container; A water vapor adding module, used to open the steam valve of the sealed container and close the pressure relief valve of the sealed container, and add water vapor to the sealed container for pressurization treatment until the pressure in the sealed container reaches 0.5-5MPa and the temperature reaches 150-260°C, and then stop adding water vapor; A constant temperature and pressure maintaining module, used to close the steam valve of the sealed container and maintain a constant temperature and pressure in the sealed container for 5-20 minutes; A pressure reduction module, used for opening the pressure relief valve of the sealed container to reduce the pressure, and collecting the sand in the sealed container when the pressure in the sealed container drops to normal pressure; as well as The drying and air separation module is used for drying, separating and air separation of the sand to obtain regenerated sand.

10. The regeneration equipment for waste foundry sand according to claim 9, characterized in that: The drying and air separation module comprises: A data acquisition unit, used to acquire the fan speed values ​​at a plurality of predetermined time points within a predetermined time period, and the detection sound signal of the predetermined time period; a time series characteristic analysis unit, configured to perform time series characteristic analysis on the fan speed values ​​at the plurality of predetermined time points and the detection sound signal in the predetermined time period to obtain a mapping associated characteristic vector; and The speed value control unit is used to determine whether the fan speed value at the current time point should be increased or decreased based on the mapping associated characteristic vector.

Images