Concrete mixing control method and equipment based on data processing and medium

Abstract

The invention relates to the technical field of computers, in particular to a concrete mixing control method and device based on data processing and a medium. According to the method, a reference dimension reduction feature is obtained by conducting dimension reduction processing on a mixing parameter matrix, and feature prediction is conducted based on the reference dimension reduction feature; compared with the prior art in which prediction is carried out directly according to the stirring parameter matrix, the calculation amount of prediction reasoning is effectively reduced, the real-time performance of concrete stirring control carried out in a prediction mode is further guaranteed, the efficiency of concrete stirring control is improved, prediction features are verified, and the accuracy of concrete stirring control is improved. According to the method, the reliability of predicting by adopting the reference dimension reduction features is ensured, and the situation that the prediction accuracy is reduced due to feature prediction based on the reference dimension reduction features is avoided, so that the reliability of concrete mixing control is ensured.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a concrete mixing control method, equipment, and medium based on data processing. Background Technology

[0002] In manufacturing scenarios using concrete mixing equipment, it is usually necessary to adjust the operating parameters of the concrete mixing equipment according to the real-time situation of concrete mixing to ensure that the final concrete mixing result is good and can meet the needs of production or use.

[0003] However, the real-time status of concrete mixing is often not directly obtainable. In existing technologies, sensors deployed on concrete mixing equipment are typically used to collect information such as temperature, humidity, and pressure. The concrete mixing status is then indirectly determined based on the collected mixing parameter information. Obviously, since the concrete mixing status can only be determined after the mixing parameter information is collected, this method has poor real-time performance, making it difficult to carry out concrete mixing under the condition that the concrete mixing equipment is under optimal control parameters, thus resulting in low concrete mixing control efficiency.

[0004] To address the aforementioned issues, existing technologies typically employ time-series prediction models to predict mixing parameters and determine control parameters based on these predictions, thereby effectively improving the real-time performance of concrete mixing control. However, due to the numerous types of mixing parameters and the large amount of input data parameters for time-series prediction models, a significant computational burden is incurred. In scenarios with low processing power, it becomes difficult to predict mixing parameters in a timely manner, resulting in limited improvement in concrete mixing control efficiency.

[0005] Therefore, how to improve the efficiency of concrete mixing control has become an urgent problem to be solved. Summary of the Invention

[0006] To address the aforementioned technical problems, the present invention provides a data-processing-based concrete mixing control method, which includes the following steps: S101, obtain the stirring parameter vectors corresponding to M preset time points respectively, wherein the stirring parameter vectors include sensor acquisition values ​​corresponding to N preset stirring parameter types respectively, and M and N are both positive integers.

[0007] S102 is obtained by concatenating M stirring parameter vectors to form a stirring parameter matrix.

[0008] S103, input the stirring parameter matrix into the trained parameter dimensionality reduction model to obtain the reference dimensionality reduction features.

[0009] S104, The reference dimensionality reduction features are input into the trained feature prediction model to predict the predicted features corresponding to the preset target time point.

[0010] S105, determine the verification requirements for the target time point based on the preset reference verification frequency, the target time point, and the Mth preset time point.

[0011] S106, when the verification requirement of the target time point meets the first preset condition, the predicted feature corresponding to the target time point is verified to obtain the verification result.

[0012] S107, when the verification result meets the second preset condition, the prediction feature corresponding to the target time point is input into the trained control parameter prediction model to obtain the prediction parameters corresponding to K preset equipment control parameter types, where K is a positive integer.

[0013] S108, at the target time point, the concrete mixing equipment is controlled according to the predicted parameters corresponding to the K equipment control parameter types respectively.

[0014] The present invention also provides a data processing-based concrete mixing control device, which includes: The parameter acquisition module is used to acquire stirring parameter vectors corresponding to M preset time points, wherein the stirring parameter vectors include sensor acquisition values ​​corresponding to N preset stirring parameter types, and M and N are both positive integers.

[0015] The vector splicing module is used to splice M stirring parameter vectors to obtain a stirring parameter matrix.

[0016] The matrix dimensionality reduction module is used to input the stirring parameter matrix into the trained parameter dimensionality reduction model to obtain reference dimensionality reduction features.

[0017] The feature prediction module is used to input the reference dimensionality reduction features into the trained feature prediction model to predict the predicted features corresponding to the preset target time point.

[0018] The verification and judgment module is used to determine the verification requirements of the target time point based on the preset reference verification frequency, the target time point, and the Mth preset time point.

[0019] The feature verification module is used to verify the predicted features corresponding to the target time point when the verification requirement of the target time point meets the first preset condition, and obtain the verification result.

[0020] The parameter prediction module is used to input the prediction features corresponding to the target time point into the trained control parameter prediction model when the verification result meets the second preset condition, so as to obtain the prediction parameters corresponding to K preset equipment control parameter types respectively, where K is a positive integer.

[0021] The equipment control module is used to control the concrete mixing equipment at the target time point according to the predicted parameters corresponding to the K equipment control parameter types.

[0022] The present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described data processing-based concrete mixing control method.

[0023] The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described data processing-based concrete mixing control method.

[0024] The present invention has at least the following beneficial effects: by performing dimensionality reduction processing on the mixing parameter matrix to obtain reference dimensionality reduction features, and performing feature prediction based on the reference dimensionality reduction features, the computational load of prediction inference is effectively reduced compared with the existing technology of directly predicting based on the mixing parameter matrix, thereby ensuring the real-time performance of concrete mixing control using the prediction method and improving the efficiency of concrete mixing control. By verifying the predicted features, the reliability of prediction using the reference dimensionality reduction features is ensured, avoiding the reduction in prediction accuracy due to feature prediction based on the reference dimensionality reduction features, thereby ensuring the reliability of concrete mixing control. Attached Figure Description

[0025] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a schematic flowchart of a data processing-based concrete mixing control method provided in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of a data processing-based concrete mixing control device provided in Embodiment 2 of the present invention. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It is understood that, where appropriate, the terms used to distinguish similar objects can be interchanged so that the invention can also be implemented in other embodiments besides the illustrated or described embodiments. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or devices.

[0029] Example 1 This embodiment provides a data processing-based method for controlling concrete mixing, such as... Figure 1 The diagram shown is a flowchart illustrating a data-processing-based concrete mixing control method according to Embodiment 1 of the present invention. This data-processing-based concrete mixing control method includes the following steps: S101, obtain the stirring parameter vectors corresponding to M preset time points respectively, wherein the stirring parameter vectors include sensor acquisition values ​​corresponding to N preset stirring parameter types respectively, and M and N are both positive integers; S102, the stirring parameter matrix is ​​obtained by concatenating M stirring parameter vectors; S103, Input the stirring parameter matrix into the trained parameter dimensionality reduction model to obtain the reference dimensionality reduction features; S104, The reference dimensionality reduction features are input into the trained feature prediction model to predict the predicted features corresponding to the preset target time point. S105, determine the verification requirements for the target time point based on the preset reference verification frequency, the target time point, and the Mth preset time point; S106, when the verification requirement of the target time point meets the first preset condition, the predicted feature corresponding to the target time point is verified to obtain the verification result; S107, when the verification result meets the second preset condition, the prediction feature corresponding to the target time point is input into the trained control parameter prediction model to obtain the prediction parameters corresponding to K preset equipment control parameter types, where K is a positive integer; S108, at the target time point, the concrete mixing equipment is controlled according to the predicted parameters corresponding to the K equipment control parameter types respectively.

[0030] The time interval between adjacent preset time points is the same. The concrete mixing equipment is equipped with N sensors that collect different preset mixing parameters. The preset mixing parameter types may include pressure, temperature, humidity, etc. The sensor collection value is the mixing parameter collected by the sensor of the corresponding preset mixing parameter type at the corresponding preset time point.

[0031] The stirring parameter vector is represented by an N×1 vector. M stirring parameter vectors are column-joined according to the chronological order of their corresponding preset time points to obtain a stirring parameter matrix, which is represented by an N×M matrix.

[0032] A parametric dimensionality reduction model can include several convolutional modules. Each convolutional module can include convolutional layers, pooling layers, and normalization layers. The input size of the parametric dimensionality reduction model is N×M, and the output size is R×T, where R and T are both positive integers, and R... <N,T<M。

[0033] Reference dimensionality reduction features can refer to the processing results of reference dimensionality reduction features after being processed by a trained parameter dimensionality reduction model. Reference dimensionality reduction features are represented by a matrix of size R×T.

[0034] The feature prediction model can use the U-Net neural network, and the specific architecture of the feature prediction model will not be elaborated here.

[0035] The predicted features are also represented as an R×T matrix.

[0036] The reference verification frequency can be preset by the implementer to verify the predicted features according to the reference verification frequency, thereby ensuring the reliability of the predicted features.

[0037] The control parameter prediction model can adopt a fully connected layer architecture. The input size of the control parameter prediction model is R×T, and the output size is 1×K.

[0038] Equipment control parameters can include the speed, power, and material ratio of the concrete mixing equipment.

[0039] In one specific implementation, the step of inputting the reference dimensionality reduction features into a trained feature prediction model to predict the predicted features corresponding to a preset target time point includes: Determine the reference time interval between adjacent preset time points; Calculate the target time interval between the target time point and the Mth preset time point; Calculate the ratio of the target time interval to the reference time interval, and use the ratio calculation result as the target iteration number Q; Initialize the real-time iteration number P = 1 with the reference dimensionality-reduced feature as the intermediate input data; If P < Q, input the intermediate input data into the trained feature prediction model to obtain an intermediate feature, and use the intermediate feature as the intermediate input data to update P = P + 1; If P = Q, input the intermediate input data into the trained feature prediction model to obtain an intermediate feature, and use the intermediate feature as the predicted feature corresponding to the target time point.

[0040] Among them, the reference time interval can refer to a fixed time interval between adjacent preset time points. The target time point can be set by the implementer himself, but it should be ensured that the ratio of the target time interval determined based on the target time point to the reference time interval is a positive integer, that is, to ensure that the target iteration number Q is a positive integer.

[0041] In a specific implementation manner, the determining the verification requirement of the target time point according to the preset reference verification frequency, the target time point and the Mth preset time point includes: Take the reciprocal of the reference verification frequency to obtain the reference verification period T, where T is a positive integer less than M; Calculate the target time interval between the target time point and the Mth preset time point; If the ratio of the target time interval to the reference verification period is a positive integer, determine that the verification requirement of the target time point is to be verified; If the ratio of the target time interval to the reference verification period is not a positive integer, determine that the verification requirement of the target time point is not to be verified.

[0042] Among them, the reference verification frequency is preset by the implementer, and the reference verification period T determined based on the reference verification frequency is a positive integer multiple of the reference time interval;

[0043] In a specific implementation manner, the first preset condition is: the verification requirement of the target time point is to be verified.

[0044] Among them, when the verification requirement of the target time point is to be verified, verify the predicted feature corresponding to the target time point to obtain a verification result.

[0045] In one specific implementation, the step of verifying the predicted features corresponding to the target time point to obtain the verification result includes: Subtract the target time point from the reference verification period to obtain the reference time point; The predicted features corresponding to the target time point are input into the trained feature recovery model to obtain the first reference parameter matrix; When the reference time point is the same as the Mth preset time point, the stirring parameter matrix is ​​determined as the second reference parameter matrix; When the reference time point is different from the Mth preset time point, the predicted features corresponding to the reference time point are input into the trained feature recovery model to obtain the second reference parameter matrix; Extract the first column to the MTth column of the first reference parameter matrix to obtain the first temporary submatrix; Extract columns T+1 to M from the second reference parameter matrix to obtain the second temporary submatrix; When the first temporary submatrix and the second temporary submatrix are the same, the verification result is determined to be successful. When the first temporary submatrix and the second temporary submatrix are the same, the verification result is determined to be verification failure.

[0046] Among them, the feature recovery model is used to recover the predicted features into the predicted stirring parameter matrix.

[0047] Specifically, by introducing a feature recovery model, the interpretability of feature prediction based on reference dimensionality reduction features is enhanced. The predicted features obtained in each iteration can be recovered into the predicted stirring parameter matrix through the feature recovery model, which is equivalent to the result of prediction using a conventional time series prediction model based on the stirring parameter matrix.

[0048] When the first temporary submatrix and the second temporary submatrix are the same, it can be assumed that the feature prediction process strictly follows the time sequence for iterative prediction, and the verification result is determined to be successful.

[0049] In one specific implementation, the second preset condition is: the verification result is that the verification is passed.

[0050] When the verification result is successful, the predicted features corresponding to the target time point are input into the trained control parameter prediction model to obtain the predicted parameters corresponding to K preset equipment control parameter types.

[0051] In one specific implementation, the training process of the parameter dimensionality reduction model, the feature prediction model, the feature recovery model, and the control parameter prediction model includes the following steps: Obtain the historical parameter vectors corresponding to W historical time points and the control parameter annotation information corresponding to W historical time points, where W is an integer greater than M; A first historical parameter matrix is ​​formed by the historical parameter vectors corresponding to the Rth historical time point to the R+M-1th historical time point respectively, and the control parameter annotation information corresponding to the R+M-1th historical time point is used as the first label information corresponding to the first historical parameter matrix, where R+M+1≤W; The second historical parameter matrix is ​​formed by the historical parameter vectors corresponding to the (R+1)th historical time point to the (R+M)th historical time point, and the control parameter annotation information corresponding to the (R+M)th historical time point is used as the second label information corresponding to the second historical parameter matrix. The first historical parameter matrix is ​​input into the parameter dimensionality reduction model to obtain the first sample dimensionality reduction features; The second historical parameter matrix is ​​input into the parameter dimensionality reduction model to obtain the second sample dimensionality reduction features; The dimensionality reduction features of the first sample are input into the feature prediction model to obtain the predicted sample features; Based on the predicted sample features and the second sample dimensionality reduction features, the feature prediction loss is calculated; The first sample dimensionality reduction features are input into the feature recovery model to obtain the first sample recovery matrix; The second sample dimensionality reduction features are input into the feature recovery model to obtain the second sample recovery matrix; The first recovery sub-loss is calculated based on the first sample recovery matrix and the first historical parameter matrix; The second recovery sub-loss is calculated based on the second sample recovery matrix and the second historical parameter matrix; The first sample dimensionality reduction feature is input into the control parameter prediction model to obtain the first sample control parameter vector; The second sample dimensionality reduction features are input into the control parameter prediction model to obtain the second sample control parameter vector; The first control parameter prediction sub-loss is calculated based on the first sample control parameter vector and the first label information. The second control parameter prediction sub-loss is calculated based on the second sample control parameter vector and the second label information; The target training loss is determined by the feature prediction loss, the first recovery sub-loss, the second recovery sub-loss, the first control parameter prediction sub-loss, and the second control parameter prediction loss. Based on the target training loss, the parameter dimensionality reduction model, the feature prediction model, the feature recovery model, and the control parameter prediction model are jointly trained until the target training loss converges, resulting in the trained parameter dimensionality reduction model, the trained feature prediction model, the trained feature recovery model, and the trained control parameter prediction model.

[0052] Among them, the control parameter annotation information can be pre-annotated manually based on W historical parameter vectors.

[0053] Specifically, the mean squared error loss is calculated by performing a mean squared error on the predicted sample features and the dimensionality-reduced features of the second sample to obtain the feature prediction loss. The feature prediction loss is used to supervise the prediction process of the feature prediction model to strictly meet the time constraints.

[0054] The first sample recovery matrix and the first historical parameter matrix are subjected to mean squared error loss calculation to obtain the first recovery sub-loss. The second sample recovery matrix and the second historical parameter matrix are subjected to mean squared error loss calculation to obtain the second recovery sub-loss. The first recovery sub-loss and the second recovery sub-loss are used to supervise the representation ability of the feature after dimensionality reduction.

[0055] The mean squared error loss is calculated using the first sample control parameter vector and the first label information to obtain the first control parameter prediction sub-loss. The mean squared error loss is calculated using the second sample control parameter vector and the second label information to obtain the second control parameter prediction sub-loss.

[0056] In this first embodiment, by performing dimensionality reduction processing on the mixing parameter matrix, reference dimensionality reduction features are obtained. Feature prediction is then performed based on these reference dimensionality reduction features. Compared with the existing technology that directly predicts based on the mixing parameter matrix, this effectively reduces the computational load of prediction inference, thereby ensuring the real-time performance of concrete mixing control using the prediction method and improving the efficiency of concrete mixing control. By verifying the predicted features, the reliability of prediction using the reference dimensionality reduction features is ensured, avoiding a decrease in prediction accuracy due to feature prediction based on the reference dimensionality reduction features, thus ensuring the reliability of concrete mixing control.

[0057] Example 2 This second embodiment provides a concrete mixing control device based on data processing, such as... Figure 2 The diagram shown is a structural schematic of a data processing-based concrete mixing control device according to Embodiment 2 of the present invention. This data processing-based concrete mixing control device includes: The parameter acquisition module 201 is used to acquire stirring parameter vectors corresponding to M preset time points, wherein the stirring parameter vectors include sensor acquisition values ​​corresponding to N preset stirring parameter types, and M and N are both positive integers.

[0058] The vector splicing module 202 is used to splice M stirring parameter vectors to obtain a stirring parameter matrix.

[0059] The matrix dimensionality reduction module 203 is used to input the stirring parameter matrix into the trained parameter dimensionality reduction model to obtain reference dimensionality reduction features.

[0060] The feature prediction module 204 is used to input the reference dimensionality reduction features into the trained feature prediction model to predict the predicted features corresponding to the preset target time point.

[0061] The verification judgment module 205 is used to determine the verification requirements of the target time point based on the preset reference verification frequency, the target time point, and the Mth preset time point.

[0062] The feature verification module 206 is used to verify the predicted features corresponding to the target time point when the verification requirement of the target time point meets the first preset condition, and obtain the verification result.

[0063] The parameter prediction module 207 is used to input the prediction features corresponding to the target time point into the trained control parameter prediction model when the verification result meets the second preset condition, so as to obtain the prediction parameters corresponding to K preset equipment control parameter types respectively, where K is a positive integer.

[0064] The equipment control module 208 is used to control the concrete mixing equipment at the target time point according to the predicted parameters corresponding to the K equipment control parameter types respectively.

[0065] It should be noted that the specific limitations of the data processing-based concrete mixing control device can be found in the limitations of the data processing-based concrete mixing control method above, and will not be repeated here. The information interaction and execution process between the above modules are based on the same concept as the method embodiments of this invention, and their specific functions and technical effects can be found in the method embodiments section, and will not be repeated here.

[0066] Example 3 This embodiment provides a computer device, which can be a server. The computer device may include a processor, memory, network interface, and database connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a data processing-based concrete mixing control method.

[0067] Example 4 This fourth embodiment provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the data-processing-based concrete mixing control method described in the above embodiments. To avoid repetition, this will not be repeated here. Alternatively, when executed by a processor, the computer program implements the functions of each module / unit in the above embodiment of the data-processing-based concrete mixing control device. To avoid repetition, this will not be repeated here.

[0068] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0069] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional units and modules is used as an example. In practical applications, the above functions can be assigned to different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0070] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any simple modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A concrete mixing control method based on data processing, characterized in that, The concrete mixing control method based on data processing includes the following steps: S101. Obtain the mixing parameter vectors corresponding to M preset time points respectively, where the mixing parameter vector includes the sensor acquisition values corresponding to N preset mixing parameter types, and both M and N are positive integers; S102. Concatenate the M mixing parameter vectors to obtain a mixing parameter matrix; S103. Input the mixing parameter matrix into the trained parameter dimensionality reduction model to obtain a reference dimensionality reduction feature; S104. Input the reference dimensionality reduction feature into the trained feature prediction model to predict the prediction feature corresponding to a preset target time point; S105. Determine the verification requirement of the target time point according to the preset reference verification frequency, the target time point and the Mth preset time point; S106. When the verification requirement of the target time point meets the first preset condition, verify the prediction feature corresponding to the target time point to obtain a verification result; S107. When the verification result meets the second preset condition, input the prediction feature corresponding to the target time point into the trained control parameter prediction model to obtain the prediction parameters corresponding to K preset equipment control parameter types, where K is a positive integer; S108. At the target time point, control the concrete mixing equipment according to the prediction parameters corresponding to K equipment control parameter types respectively.

2. The concrete mixing control method based on data processing according to claim 1, characterized in that, The step of inputting the reference dimensionality reduction feature into the trained feature prediction model to predict the prediction feature corresponding to a preset target time point includes: Determine the reference time interval between adjacent preset time points; Calculate the target time interval between the target time point and the Mth preset time point; Perform a ratio calculation on the target time interval and the reference time interval, and use the ratio calculation result as the target iteration number Q; Use the reference dimensionality reduction feature as the intermediate input data, and initialize the real-time iteration number P = 1; If P < Q, input the intermediate input data into the trained feature prediction model to obtain an intermediate feature, and use the intermediate feature as the intermediate input data to update P = P + 1; If P = Q, input the intermediate input data into the trained feature prediction model to obtain an intermediate feature, and use the intermediate feature as the prediction feature corresponding to the target time point.

3. The concrete mixing control method based on data processing according to claim 1, characterized in that, The step of determining the verification requirement of the target time point according to the preset reference verification frequency, the target time point and the Mth preset time point includes: Take the reciprocal of the reference verification frequency to obtain a reference verification period T, where T is a positive integer less than M; Calculate the target time interval between the target time point and the Mth preset time point; If the ratio of the target time interval to the reference verification period is a positive integer, determine that the verification requirement of the target time point is to be verified; If the ratio of the target time interval to the reference verification period is not a positive integer, determine that the verification requirement of the target time point is not to be verified.

4. The concrete mixing control method based on data processing according to claim 3, characterized in that, The first preset condition is that the verification requirement of the target time point is to be verified.

5. The concrete mixing control method based on data processing according to claim 3, characterized in that, The step of verifying the prediction feature corresponding to the target time point to obtain a verification result includes: Subtract the target time point from the reference verification period to obtain the reference time point; The predicted features corresponding to the target time point are input into the trained feature recovery model to obtain the first reference parameter matrix; When the reference time point is the same as the Mth preset time point, the stirring parameter matrix is ​​determined as the second reference parameter matrix; When the reference time point is different from the Mth preset time point, the predicted features corresponding to the reference time point are input into the trained feature recovery model to obtain the second reference parameter matrix; Extract the first column to the MTth column of the first reference parameter matrix to obtain the first temporary submatrix; Extract columns T+1 to M from the second reference parameter matrix to obtain the second temporary submatrix; When the first temporary submatrix and the second temporary submatrix are the same, the verification result is determined to be successful. When the first temporary submatrix and the second temporary submatrix are the same, the verification result is determined to be verification failure.

6. The concrete mixing control method based on data processing according to claim 5, characterized in that, The second preset condition is: the verification result is that the verification is successful.

7. The concrete mixing control method based on data processing according to claim 5, characterized in that, The training process of the parameter dimensionality reduction model, the feature prediction model, the feature recovery model, and the control parameter prediction model includes the following steps: Obtain the historical parameter vectors corresponding to W historical time points and the control parameter annotation information corresponding to W historical time points, where W is an integer greater than M; A first historical parameter matrix is ​​formed by the historical parameter vectors corresponding to the Rth historical time point to the R+M-1th historical time point respectively, and the control parameter annotation information corresponding to the R+M-1th historical time point is used as the first label information corresponding to the first historical parameter matrix, where R+M+1≤W; The second historical parameter matrix is ​​formed by the historical parameter vectors corresponding to the (R+1)th historical time point to the (R+M)th historical time point, and the control parameter annotation information corresponding to the (R+M)th historical time point is used as the second label information corresponding to the second historical parameter matrix. The first historical parameter matrix is ​​input into the parameter dimensionality reduction model to obtain the first sample dimensionality reduction features; The second historical parameter matrix is ​​input into the parameter dimensionality reduction model to obtain the second sample dimensionality reduction features; The dimensionality reduction features of the first sample are input into the feature prediction model to obtain the predicted sample features; Based on the predicted sample features and the second sample dimensionality reduction features, the feature prediction loss is calculated; The first sample dimensionality reduction features are input into the feature recovery model to obtain the first sample recovery matrix; The second sample dimensionality reduction features are input into the feature recovery model to obtain the second sample recovery matrix; The first recovery sub-loss is calculated based on the first sample recovery matrix and the first historical parameter matrix; The second recovery sub-loss is calculated based on the second sample recovery matrix and the second historical parameter matrix; The first sample dimensionality reduction feature is input into the control parameter prediction model to obtain the first sample control parameter vector; The second sample dimensionality reduction features are input into the control parameter prediction model to obtain the second sample control parameter vector; The first control parameter prediction sub-loss is calculated based on the first sample control parameter vector and the first label information. The second control parameter prediction sub-loss is calculated based on the second sample control parameter vector and the second label information; The target training loss is determined by the feature prediction loss, the first recovery sub-loss, the second recovery sub-loss, the first control parameter prediction sub-loss, and the second control parameter prediction loss. Based on the target training loss, the parameter dimensionality reduction model, the feature prediction model, the feature recovery model, and the control parameter prediction model are jointly trained until the target training loss converges, resulting in the trained parameter dimensionality reduction model, the trained feature prediction model, the trained feature recovery model, and the trained control parameter prediction model.

8. A concrete mixing control device based on data processing, characterized in that, The data processing-based concrete mixing control device includes: The parameter acquisition module is used to acquire stirring parameter vectors corresponding to M preset time points, wherein the stirring parameter vectors include sensor acquisition values ​​corresponding to N preset stirring parameter types, and M and N are both positive integers; The vector concatenation module is used to concatenate M stirring parameter vectors to obtain a stirring parameter matrix. The matrix dimensionality reduction module is used to input the stirring parameter matrix into the trained parameter dimensionality reduction model to obtain reference dimensionality reduction features; The feature prediction module is used to input the reference dimensionality reduction features into the trained feature prediction model to predict the predicted features corresponding to the preset target time point. The verification and judgment module is used to determine the verification requirements of the target time point based on the preset reference verification frequency, the target time point, and the Mth preset time point. The feature verification module is used to verify the predicted features corresponding to the target time point when the verification requirement of the target time point meets the first preset condition, and obtain the verification result. The parameter prediction module is used to input the prediction features corresponding to the target time point into the trained control parameter prediction model when the verification result meets the second preset condition, so as to obtain the prediction parameters corresponding to K preset equipment control parameter types respectively, where K is a positive integer; The equipment control module is used to control the concrete mixing equipment at the target time point according to the predicted parameters corresponding to the K equipment control parameter types.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the data processing-based concrete mixing control method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the data processing-based concrete mixing control method according to any one of claims 1 to 7.

Images