A fluidity prediction method and device for fluid-solidified soil with multi-feature input

Abstract

The present invention relates to the technical field of fluidized solidified soil, and particularly to a method and device for predicting the fluidity of fluidized solidified soil with multi-feature input. A method for predicting the fluidity of fluidized solidified soil with multi-feature input includes: S1, collecting raw material features; S2, collecting current features; S3, collecting discharging features; S4, constructing a prediction model; and S5, applying it to the production line. By pre-establishing a multi-input prediction model for the fluidity of fluidized solidified soil based on raw material features, current features, and discharging features and applying it to the production line, during the actual mixing process of fluidized solidified soil in the production line, by inputting corresponding feature data into the mature prediction model, the fluidity of the current batch of fluidized solidified soil can be pre-determined. The prediction of the fluidity of fluidized solidified soil by multi-input features has high accuracy. At the same time, it solves the problem in the prior art that the fluidity of fluidized solidified soil can only be measured offline, and realizes the online measurement of the fluidity of fluidized solidified soil.

Description

Technical Field

[0001] The present invention relates to the technical field of fluid-solidified soil, and in particular to a method and device for predicting the fluidity of fluid-solidified soil with multi-feature input. Background Art

[0002] Fluid-solidified soil is a new type of building material, which is widely used in engineering such as foundation reinforcement, trench backfilling, road subgrade, foundation pit support curtain wall, and mine goaf backfilling. Its main characteristics are good fluidity, can be constructed by pumping or pouring, has a fast filling and laying speed, can completely fill all the small voids and irregular shapes in the space, ensure full contact between the filler and the surrounding soil, and provide stable support and uniform distributed load.

[0003] However, there are some deficiencies in the existing methods for detecting the fluidity of fluid-solidified soil. For example, when detecting according to the method for determining the fluidity of cement mortar in GB / T 2419-2005, the accuracy of fluidity measurement is related to the operation specification degree of workers, the detection time is relatively long, about six minutes, and the detection has hysteresis, and the change of fluidity during the production process cannot be detected. Therefore, it is of great significance to develop a fast and accurate online method for predicting the fluidity of fluid-solidified soil. Summary of the Invention

[0004] Other features and advantages of the present invention will be described in the following specification, and will become apparent in part from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structures specifically pointed out in the specification and other specification drawings.

[0005] The objective of the present invention is to overcome the above deficiencies, and provide a method and device for predicting the fluidity of fluid-solidified soil with multi-feature input. By pre-establishing a multi-input fluidity prediction model of fluid-solidified soil based on raw material characteristics, current characteristics, and discharging characteristics and applying it to the production line, during the actual mixing process of fluid-solidified soil on the production line, by inputting corresponding characteristic data into the mature prediction model, the fluidity of the current batch of fluid-solidified soil can be pre-determined in advance. The prediction accuracy of the fluidity of fluid-solidified soil through multi-input characteristics is high. At the same time, the mature fluidity prediction model has high graftability and wide applicability, solves the problem that a large amount of labor is required for the fluidity of fluid-solidified soil in the prior art and can only be measured offline, and realizes the online measurement of the fluidity of fluid-solidified soil.

[0006] The present invention provides a method for predicting the fluidity of fluid-solidified soil with multi-feature input, including:

[0007] S1. Collect raw material characteristics: For multiple batches of fluid-solidified soil that have been produced, trace their corresponding production formula data, and obtain the raw material characteristics of the fluid-solidified soil after processing the production formula data.

[0008] S2. Collect current characteristics: For multiple batches of fluid-solidified soil, trace the current curves during their respective mixing processes, and obtain the current characteristics based on the current curves.

[0009] S3. Collect discharging characteristics: For multiple batches of fluid-solidified soil, trace the discharging images after their respective mixing is completed, and form the discharging characteristics after processing the discharging images.

[0010] S4. Build a prediction model: Build a fluid-solidified soil fluidity prediction model based on a neural network, and train the prediction model with training data to obtain a mature fluid-solidified soil fluidity prediction model. The training data specifically includes one-dimensional arrays formed by raw material characteristics, current characteristics, discharging characteristics, and corresponding fluidity labels.

[0011] S5. Application in the production line: Integrate the multi-characteristics obtained during the production process of the production line into a one-dimensional array and input it into the mature fluid-solidified soil fluidity prediction model to output the fluid-solidified soil fluidity prediction result.

[0012] In some embodiments, in step S1, the specific steps for collecting raw material characteristics are as follows:

[0013] S11. Obtain the actual production formula from the raw material metering equipment, specifically including muck, curing agent, admixture, and water.

[0014] S12. Obtain the water mass characteristic based on the water content of the muck raw material and the water usage, obtain the soil mass characteristic by calculating based on the mass of the muck raw material and the water content of the muck raw material, obtain the curing agent mass characteristic based on the type and mass of the curing agent, and obtain the admixture mass characteristic based on the type and mass of the admixture.

[0015] S13. Combine the water mass characteristic, soil mass characteristic, curing agent mass characteristic, and admixture mass characteristic to form the raw material characteristic.

[0016] In some embodiments, in step S2, the specific steps for collecting current characteristics are as follows:

[0017] S21. Collect the current values in the mixer at a fixed frequency to obtain the current curve, smooth the current curve, and take the current value at no-load as the first current value.

[0018] S22. When the mixer stirs the fluidized solidified soil, the current curve starts to rise from the no-load current. As the fluidized solidified soil is stirred evenly, the current value continuously decreases. The stage when the current value no longer decreases is taken as the sign of the evenly stirred stage, and the current value in the evenly stirred stage is taken as the second current value;

[0019] S23. The difference obtained by subtracting the first current value from the second current value is used as the current characteristic of the current batch of fluidized solidified soil.

[0020] In some embodiments, in step S3, the specific steps for collecting the discharging characteristics are as follows:

[0021] S31. Obtain the discharging image at the moment when the discharging port is opened as the first image;

[0022] S32. Obtain the discharging image 2 seconds after the discharging port is opened as the second image;

[0023] S33. Calculate the absolute value of the difference in pixel values between the first image and the second image, and obtain the third image after converting it into a grayscale image;

[0024] S34. Perform thresholding processing on the third image, and count the number of non-0 pixel points as the discharging characteristic of the current batch of fluidized solidified soil.

[0025] In some embodiments, in step S4, the fluidized solidified soil fluidity prediction model uses a fully connected neural network. The specific structure of the fully connected neural network includes an input layer, a first hidden layer, a second hidden layer, and an output layer. Among them, the number of neurons in the first hidden layer is 64, the number of neurons in the second hidden layer is 32, the activation function is ReLU, and the number of neurons in the output layer is 1 and is equipped with a linear activation function.

[0026] In some embodiments, the optimizer in the training process of the fully connected neural network uses the Adam optimizer, and the loss function uses the mean absolute error evaluation.

[0027] In some embodiments, in step S4, for the produced fluidized solidified soil, the producer divides the grades with reference to the fluidity standard and adds the fluidity labels into the corresponding arrays.

[0028] A fluidized solidified soil fluidity prediction device with multi-feature input, the device includes:

[0029] A double-shaft mixing bin, which is used to evenly stir the raw materials to form fluidized solidified soil;

[0030] A temporary storage bin, which is used to temporarily store the fluidized solidified soil and guide it to the tanker. A discharging port is provided at the bottom side of the temporary storage bin;

[0031] A camera, which is located on the side of the double-shaft mixing bin and the temporary storage bin, and is used to take horizontal pictures of the fluidized solidified soil below the discharging port;

[0032] A control center, which is connected to the biaxial mixing bin and the camera for data connection;

[0033] The rotating shaft of the biaxial mixing bin is connected to the control center for obtaining current characteristics. After the camera takes a horizontal picture, the picture is input into the control center, and the actual production formula obtained from the raw material metering equipment is stored in the control center.

[0034] In some embodiments, the control center includes a model construction module and a prediction module. The model construction module trains a prediction model using training data to form a mature prediction model for the fluidity of fluid-solidified soil. The mature prediction model for the fluidity of fluid-solidified soil is input into the prediction module, and after inputting a multi-feature one-dimensional array, the predicted value of the fluidity of fluid-solidified soil is output.

[0035] In some embodiments, the control center further includes a feature extraction module, which is used to obtain the actual production formula from the raw material metering equipment and convert it into raw material characteristics, to convert the current value into current characteristics, and to convert the horizontal picture into discharging characteristics.

[0036] By adopting the above technical solutions, the beneficial effects of the present invention are as follows:

[0037] The present invention pre-establishes a multi-input prediction model for the fluidity of fluid-solidified soil based on raw material characteristics, current characteristics and discharging characteristics and applies it to the production line. During the actual mixing process of fluid-solidified soil on the production line, by inputting the corresponding characteristic data into the mature prediction model, the fluidity of the current batch of fluid-solidified soil can be pre-determined in advance. The prediction of the fluidity of fluid-solidified soil by multi-input characteristics is highly accurate. At the same time, the mature fluidity prediction model has high graftability and wide applicability, solving the problem in the prior art that a large amount of labor is required to measure the fluidity of fluid-solidified soil and it can only be measured offline, and realizing the online measurement of the fluidity of fluid-solidified soil.

[0038] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure.

[0039] Undoubtedly, such purposes of the present invention and other purposes will become more apparent after the detailed description of the preferred embodiments described in multiple drawings and diagrams below.

[0040] To make the above and other purposes, features and advantages of the present invention more obvious and understandable, one or several preferred embodiments are specifically given below, and in conjunction with the attached drawings, the detailed description is as follows. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings are used to provide a further understanding of the present invention, and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, but do not constitute a limitation to the present invention.

[0042] In the accompanying drawings, the same components are denoted by the same reference numerals, and the drawings are schematic and not necessarily drawn to actual scale.

[0043] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only one or several embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on such accompanying drawings.

[0044] Figure 1 It is a schematic diagram of the overall process of the fluidity prediction method of fluid-solidified soil with multi-feature input in some embodiments of the present invention;

[0045] Figure 2 It is a schematic diagram of the overall structure of the fluidity prediction device of fluid-solidified soil with multi-feature input in some embodiments of the present invention;

[0046] Figure 3 It is a schematic diagram of the internal data flow of the control center in some embodiments of the present invention.

[0047] Main reference numeral description:

[0048] 1. Twin-shaft mixing bin; 2. Temporary storage bin; 3. Camera. Detailed implementation manners

[0049] In order to make the purpose, technical solutions and advantages of the present invention clearer, the following further details the present invention in conjunction with specific implementation manners. It should be understood that the specific implementation manners described herein are only used to explain the present invention, but not to limit the present invention.

[0050] In addition, in the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "axial", "radial", "circumferential", etc. is based on the orientation or positional relationship shown in the accompanying drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus cannot be understood as a limitation to the present invention.

[0051] In the present invention, unless otherwise clearly specified or defined, terms such as "installation", "connection", "linkage", "fixation" shall be understood in a broad sense. For example, it may be a fixed connection, a detachable connection, or an integral body; it may be a direct connection or an indirect connection through an intermediate medium, and it may be the internal communication of two components or the interaction relationship between two components. However, indicating a direct connection means that there is no connection relationship constructed through a transition structure between the two connected bodies, and they are only connected through a connection structure to form an integral body. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0052] In the present invention, unless otherwise clearly specified or defined, the first feature being "on" or "under" the second feature may be that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples.

[0053] Refer to Figures 1-3 , Figure 1 is a schematic diagram of the overall process of the fluidity prediction method of fluid-solidified soil with multi-feature input in some embodiments of the present invention; Figure 2 is a schematic diagram of the overall structure of the fluidity prediction device of fluid-solidified soil with multi-feature input in some embodiments of the present invention; Figure 3 is a schematic diagram of the internal data flow of the control center in some embodiments of the present invention.

[0054] According to some embodiments of the present invention, as Figure 1 shown, the present invention provides a fluidity prediction method of fluid-solidified soil with multi-feature input, including:

[0055] S1. Collect raw material features: For multiple batches of fluid-solidified soil that have been produced, trace the corresponding production formula data, and after processing the production formula data, obtain the raw material features of the fluid-solidified soil;

[0056] The specific steps for collecting raw material features are:

[0057] S11. Obtain the actual production formula from the raw material metering equipment, specifically including construction waste, solidifying agent, admixture, and water; the actual production formula includes the actual usage amounts of raw materials. Among them, construction waste generally includes building waste, engineering slurry, etc., the solidifying agent generally includes slag powder, fly ash, cement, etc., and the admixture generally includes expansive agent, water reducing agent, etc. Different production formulas are selected corresponding to different types and batches of the fluidized solidified soil actually produced.

[0058] S12. Obtain the water mass characteristic based on the water content of the construction waste raw material and the water usage amount, obtain the soil mass characteristic by calculating according to the mass of the construction waste raw material and the water content of the construction waste raw material, obtain the solidifying agent mass characteristic according to the type and mass of the solidifying agent, and obtain the admixture mass characteristic according to the type and mass of the admixture.

[0059] S13. The water mass characteristic, soil mass characteristic, solidifying agent mass characteristic, and admixture mass characteristic are combined to form the raw material characteristic.

[0060] S2. Collect the current characteristic: For multiple batches of fluidized solidified soil, trace the current curves during their respective mixing processes, and obtain the current characteristic based on the current curves.

[0061] The specific steps for collecting the current characteristic are as follows:

[0062] S21. Collect the current values in the mixer at a fixed frequency to obtain the current curve, smooth the current curve, and take the current value at no-load as the first current value; preferably, the collection frequency of the current value is 5 Hz.

[0063] S22. When the mixer mixes the fluidized solidified soil, the current curve rises from the no-load current. As the fluidized solidified soil is mixed evenly, the current value continuously decreases. Take the stage when the current value no longer decreases as the sign of the evenly mixed stage, and take the current value in the evenly mixed stage as the second current value.

[0064] S23. Use the difference between the second current value and the first current value as the current characteristic of the current batch of fluidized solidified soil.

[0065] S3. Collect the discharging characteristic: For multiple batches of fluidized solidified soil, trace the discharging images after their respective mixing is completed, and form the discharging characteristic after processing the discharging images.

[0066] The arrangement positions among the twin-shaft mixing silo 1, the temporary storage silo 2, and the camera 3 are as Figure 2 shown;

[0067] The specific steps for collecting the discharging characteristic are as follows:

[0068] S31. Obtain the discharging image at the moment when the discharging port is opened as the first image.

[0069] S32. Obtain the discharging image at the second after the discharging port is opened as the second image;

[0070] S33. Calculate the absolute value of the difference in pixel values between the first image and the second image, and obtain the third image after converting it into a grayscale image;

[0071] S34. Perform thresholding on the third image, and count the number of non-zero pixel points as the discharging feature of the current batch of fluid-solidified soil.

[0072] S4. Build a prediction model: Build a fluid-solidified soil fluidity prediction model based on a neural network, and use the training data to train the prediction model to obtain a mature fluid-solidified soil fluidity prediction model. The training data specifically includes a one-dimensional array formed by raw material features, current features, discharging features, and corresponding fluidity labels;

[0073] The fluid-solidified soil fluidity prediction model uses a fully connected neural network. The specific structure of the fully connected neural network includes an input layer, a first hidden layer, a second hidden layer, and an output layer. Among them, the number of neurons in the first hidden layer is 64, the number of neurons in the second hidden layer is 32, the activation function is ReLU, and the number of neurons in the output layer is 1 and has a linear activation function; The optimizer in the training process of the fully connected neural network uses the Adam optimizer, and the loss function uses the mean absolute error evaluation;

[0074] For the produced fluid-solidified soil, the producer divides the grades with reference to the fluidity standard and adds the fluidity label to the corresponding array. For example, the national standard stipulates that the fluid-solidified soil with a fluidity less than or equal to 120 mm is a plastic mixture, and the fluid-solidified soil with a fluidity greater than 120 mm is a fluid mixture. The producer can further divide and delimit the intervals for the plastic mixture and the fluid mixture according to the actual production situation, and add fluidity labels suitable for its own production line.

[0075] S5. Application in the production line: Integrate the multi-feature one-dimensional array obtained during the production process of the production line and input it into the mature fluid-solidified soil fluidity prediction model to output the fluid-solidified soil fluidity prediction result.

[0076] As Figure 2 shown, the present invention also provides a fluid-solidified soil fluidity prediction device with multi-feature input, and the device includes:

[0077] A double-shaft mixing bin 1, which is used to uniformly mix raw materials to form fluid-solidified soil;

[0078] A temporary storage bin 2, which is used to temporarily store the fluid-solidified soil and guide it to the tanker. The bottom side of the temporary storage bin 2 is provided with a discharging port;

[0079] A camera 3, which is located on the side of the biaxial mixing bin 1 and the temporary storage bin 2, is used to take horizontal pictures of the fluidized solidified soil below the discharge port;

[0080] A control center, which is data-connected to the biaxial mixing bin 1 and the camera 3;

[0081] The rotating shaft of the biaxial mixing bin 1 is connected to the control center for obtaining current characteristics. After the camera 3 takes horizontal pictures, the pictures are input into the control center, and the actual production formula obtained from the raw material metering equipment is stored in the control center;

[0082] Preferably, in the actual application process, there are requirements for the opening time of the discharge port of the biaxial mixing bin 1. Each time the discharge port is about to be opened, it will wait until the main shaft of the mixer rotates to a certain fixed position before opening.

[0083] According to some embodiments of the present invention, optionally, as Figure 3 shown, the control center includes a model construction module and a prediction module. The model construction module uses training data to train the prediction model to form a mature fluidized solidified soil fluidity prediction model, and inputs the mature fluidized solidified soil fluidity prediction model into the prediction module. After inputting a multi-feature one-dimensional array, the predicted value of the fluidized solidified soil fluidity is output.

[0084] The control center also includes a feature extraction module, which is used to obtain the actual production formula from the raw material metering equipment and convert it into raw material characteristics, to convert the current value into current characteristics, and to convert the horizontal picture into discharge characteristics.

[0085] It should be understood that the embodiments disclosed in the present invention are not limited to the specific processing steps or materials disclosed herein, but should extend to equivalent alternatives of such features understood by those of ordinary skill in the relevant art. It should also be understood that the terms used herein are only for the purpose of describing specific embodiments and do not mean to limit.

[0086] The "embodiments" mentioned in the specification mean that the specific features or characteristics described in connection with the embodiments are included in at least one embodiment of the present invention. Therefore, the phrase "embodiments" that appears throughout the specification does not necessarily refer to the same embodiment.

[0087] In addition, the described features or characteristics can be combined into one or more embodiments in any other suitable way. In the above description, some specific details, such as thickness, quantity, etc., are provided to provide a comprehensive understanding of the embodiments of the present invention. However, those skilled in the relevant art will understand that the present invention can be implemented without one or more of the above specific details or can also be implemented using other methods, components, materials, etc.

Claims

1. A method for predicting the fluidity of fluid-solidified soil with multi-feature input, characterized in that, Including S1. Collect raw material characteristics: For multiple batches of fluid-solidified soil after production, trace the corresponding production formula data, process the production formula data to obtain the raw material characteristics of the fluid-solidified soil. The specific steps for collecting raw material characteristics are as follows: S11. Obtain the actual production formula from the raw material metering equipment, specifically including muck, curing agent, admixture, and water; S12. Obtain the water mass characteristic based on the water content of the muck raw material and the water consumption, calculate the soil mass characteristic based on the mass of the muck raw material and the water content of the muck raw material, obtain the curing agent mass characteristic based on the type and mass of the curing agent, and obtain the admixture mass characteristic based on the type and mass of the admixture; S13. The water mass characteristic, soil mass characteristic, curing agent mass characteristic, and admixture mass characteristic are combined to form the raw material characteristic; S2. Collect current characteristics: For multiple batches of fluid-solidified soil, trace the current curves during their respective mixing processes, and obtain the current characteristics based on the current curves. The specific steps for collecting current characteristics are as follows: S21. Collect the current values in the mixer at a fixed frequency to obtain the current curve, smooth the current curve, and take the current value at no-load as the first current value; S22. When the mixer stirs the fluid-solidified soil, the current curve rises from the no-load current. As the fluid-solidified soil is stirred evenly, the current value continuously decreases. Take the stage where the current value no longer decreases as the sign of the evenly stirred stage, and take the current value in the evenly stirred stage as the second current value; S23. Use the difference between the second current value and the first current value as the current characteristic of the current batch of fluid-solidified soil; S3. Collect discharge characteristics: For multiple batches of fluid-solidified soil, trace the discharge images after their respective mixing is completed, and form discharge characteristics after data processing of the discharge images. The specific steps for collecting discharge characteristics are as follows: S31. Obtain the discharge image at the moment when the discharge port is opened as the first image; S32. Obtain the discharge image 2 seconds after the discharge port is opened as the second image; S33. Calculate the absolute value of the difference in pixel values between the first image and the second image, and obtain the third image after converting it into a grayscale image; S34. Perform thresholding processing on the third image, and count the number of non-zero pixel points as the discharge characteristic of the current batch of fluid-solidified soil; S4. Build a prediction model: Build a fluid-solidified soil fluidity prediction model based on a neural network, and train the prediction model with training data to obtain a mature fluid-solidified soil fluidity prediction model. The training data specifically includes one-dimensional arrays formed by raw material characteristics, current characteristics, discharge characteristics, and the corresponding fluidity labels; S5. Application in the production line: Integrate the multi-characteristics obtained during the production process of the production line into a one-dimensional array and input it into the mature fluid-solidified soil fluidity prediction model to output the fluid-solidified soil fluidity prediction result.

2. The fluidity prediction method of flowing and solidifying soil with multi-feature input according to claim 1, wherein, In step S4, the fluid-solidified soil fluidity prediction model uses a fully connected neural network. The specific structure of the fully connected neural network includes an input layer, a first hidden layer, a second hidden layer, and an output layer. Among them, the number of neurons in the first hidden layer is 64, the number of neurons in the second hidden layer is 32, the activation function is ReLU, and the number of neurons in the output layer is 1 and has a linear activation function.

3. The flowability prediction method of fluid-solidified soil with multi-feature input according to claim 2, characterized in that, During the training process of the fully connected neural network, the Adam optimizer is used, and the mean absolute error is used as the loss function for evaluation.

4. The fluidity prediction method of flowing state solidified soil with multi-feature input according to claim 1, characterized in that In step S4, for the produced fluid-solidified soil, the producer classifies it according to the fluidity standard and adds the fluidity label to the corresponding array.

5. A fluidity prediction device for fluid-solidified soil with multi-feature input, characterized in that, A device for implementing the fluid-solidified soil fluidity prediction method with multi-feature input described in any one of claims 1-4, the device includes A double-shaft mixing bin for uniformly mixing raw materials to form fluid-solidified soil; A temporary storage bin for temporarily storing the fluid-solidified soil and guiding it to a tanker. A discharge port is provided at the bottom side of the temporary storage bin; A camera located on the side of the double-shaft mixing bin and the temporary storage bin for taking horizontal pictures of the fluid-solidified soil under the discharge port; A control center, which is data-connected to the double-shaft mixing bin and the camera; The rotating shaft of the double-shaft mixing bin is connected to the control center for obtaining current characteristics. After the camera takes a horizontal picture, the picture is input into the control center, and the actual production formula obtained from the raw material metering device is stored in the control center; The control center includes a model construction module and a prediction module. The model construction module uses training data to train the prediction model to form a mature fluid-solidified soil fluidity prediction model, and inputs the mature fluid-solidified soil fluidity prediction model into the prediction module. After inputting a multi-feature one-dimensional array, the predicted value of the fluid-solidified soil fluidity is output.

6. The fluidity prediction device for fluid-solidified soil with multi-feature input according to claim 5, characterized in that The control center also includes a feature extraction module, which is used to obtain the actual production formula from the raw material metering device and convert it into raw material characteristics, to convert the current value into current characteristics, and to convert the horizontal picture into discharge characteristics.

Images