Construction equipment control method and device based on process level generalization aggregation, equipment and medium

Abstract

This disclosure presents a method, apparatus, device, and medium for controlling construction equipment based on process-level generalization aggregation. One specific implementation of the method includes: acquiring a historical construction dataset; performing process-homogeneous sample aggregation processing on the historical construction dataset; performing final-level process generalization aggregation processing on each historical construction data group; responding to a construction equipment control request sent by a user terminal containing the current final-level process identifier, construction area identifier, and current construction parameter information, filtering the target historical construction data group corresponding to the current final-level process identifier; performing lineage back-tracing matching processing on the target historical construction data group; training an initial construction control parameter prediction model; and driving the construction equipment corresponding to the construction area identifier to perform construction operations. This implementation reduces the number of engineering quality accidents.

Description

Technical Field

[0001] The embodiments disclosed herein relate to the field of computer technology, and more specifically to a construction equipment control method, apparatus, equipment, and medium based on process-level generalization aggregation. Background Technology

[0002] With the rapid development of informatization and intelligentization in construction, the precise control of automated construction equipment such as concrete mixing plants and intelligent tower cranes increasingly relies on machine learning models. Taking concrete mixing equipment as an example, the setting of its batching parameters (aggregate ratio, water-cement ratio, mixing time, etc.) needs to be predicted based on historical construction data to achieve adaptive control for different grades of concrete, different pouring locations, and different construction environments. Currently, the common practice is to finely group historical construction data according to multi-level process attributes and construction areas, then train a construction control parameter prediction model for each final-level process item, and control the construction equipment to perform construction operations based on this model.

[0003] However, when using the above methods to control construction equipment for construction operations, the following technical problems often arise: Due to the highly dispersed and diverse nature of construction tasks, historical construction data, after being divided according to the final stage of the process, exhibits a typical long-tail distribution. This means that the number of historical samples for a large number of final stage processes (such as the "C30 pumped beam" process in a remote area) is extremely small, far from meeting the minimum sample size requirement for training machine learning models (e.g., at least 500 samples are needed). When the sample size is insufficient, directly training the model on the final stage process can lead to the model failing to converge or severe overfitting. This, in turn, causes the predicted control parameters to deviate from the actual requirements, increasing the frequency of engineering quality accidents (such as substandard concrete strength (too high or too low water-cement ratio), segregation of the mixture, or uncontrolled slump (incorrect aggregate proportions), resulting in cracks in the poured structural components).

[0004] The information disclosed in this background section is only intended to enhance the understanding of the background of the inventive concept, and therefore may contain information that does not form prior art known to those skilled in the art. Summary of the Invention

[0005] The summary portion of this disclosure is intended to provide a brief overview of the concepts, which will be described in detail in the detailed description portion. This summary portion is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0006] Some embodiments of this disclosure provide construction equipment control methods, apparatuses, electronic devices, and computer-readable media to solve one or more of the technical problems mentioned in the background section above.

[0007] In a first aspect, some embodiments of this disclosure provide a construction equipment control method based on process-level generalized aggregation. The method includes: acquiring a historical construction dataset; performing process-homogeneous sample aggregation processing on the historical construction dataset to obtain various historical construction data groups; based on the historical construction dataset, performing final-level process generalized aggregation processing on each of the historical construction data groups to obtain a generalized construction data group set corresponding to the historical construction data group; and responding to receiving a construction equipment control request sent by a user terminal containing the current final-level process identifier, construction area identifier, and current construction parameter information, controlling the equipment from the aforementioned historical construction data groups. From the historical construction data set, a target historical construction data set corresponding to the current final-level process identifier is selected; in response to determining that the number of historical construction data in the target historical construction data set is less than or equal to a preset lower threshold, based on the obtained generalized construction data set, a lineage back-matching process is performed on the target historical construction data set to obtain a target generalized construction data set; based on the target generalized construction data set, an initial construction control parameter prediction model is trained to obtain a construction control parameter prediction model; the current construction parameter information is input into the construction control parameter prediction model to drive the construction equipment corresponding to the construction area identifier to perform construction operations.

[0008] Secondly, some embodiments of this disclosure provide a construction equipment control device based on process-level generalization aggregation. The device includes: an acquisition unit configured to acquire a historical construction dataset to acquire an image of a plate to be tested; an aggregation unit configured to perform process-homogeneous sample aggregation processing on the historical construction dataset to obtain various historical construction data groups; a generalization aggregation processing unit configured to perform final-level process generalization aggregation processing on each of the historical construction data groups based on the historical construction dataset to obtain a set of generalized construction data groups corresponding to the historical construction data groups; and a filtering unit configured to respond to a construction equipment control request sent by a user terminal containing the current final-level process identifier, construction area identifier, and current construction parameter information. The system selects target historical construction data groups from the aforementioned historical construction data groups that correspond to the current final-level process identifier. A lineage backtracking matching processing unit is configured to, in response to determining that the number of historical construction data in the target historical construction data group is less than or equal to a preset lower threshold, perform lineage backtracking matching processing on the target historical construction data group based on the obtained generalized construction data group sets to obtain a target generalized construction data group. A training unit is configured to train an initial construction control parameter prediction model based on the target generalized construction data group to obtain a construction control parameter prediction model. A driving unit is configured to input the current construction parameter information into the construction control parameter prediction model to drive the construction equipment corresponding to the construction area identifier to perform construction operations.

[0009] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation of the first aspect above.

[0010] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the method described in any of the implementations of the first aspect above.

[0011] The above-described embodiments of this disclosure have the following beneficial effects: the construction equipment control method based on process-level generalization aggregation in some embodiments of this disclosure reduces the number of engineering quality accidents. Specifically, the reason for the increase in the number of engineering quality accidents is that, due to the highly dispersed and diverse nature of construction tasks, historical construction data, after being divided according to the final-level process, exhibits a typical long-tail distribution, meaning that the number of historical samples for a large number of final-level process items (such as the "C30 pumping beam" process in a remote area) is extremely small, far from meeting the minimum sample size requirement for training machine learning models (e.g., at least 500 samples). When the sample size is insufficient, directly training the model on the final-level process item will lead to the model failing to converge or severe overfitting, thereby causing the predicted control parameters to deviate from the actual requirements, increasing the number of engineering quality accidents. Based on this, the construction equipment control method based on process-level generalization aggregation in some embodiments of this disclosure first obtains a historical construction dataset. Then, the historical construction dataset is subjected to process homogeneous sample aggregation processing to obtain various historical construction data groups. Historical construction data with similar process characteristics (such as having completely identical process attribute identification sequences and construction areas) can be grouped together. Next, based on the aforementioned historical construction datasets, a final-level process generalization aggregation process is performed on each of the aforementioned historical construction data groups to obtain a generalized construction data set corresponding to the aforementioned historical construction data groups. Thus, by progressively removing the attributes of the final-level process and re-aggregating, a multi-gradient ancestor generalized data set corresponding to the historical construction data groups—that is, a generalized construction data set—can be pre-constructed without introducing any virtual data, forming a backup data pool. Then, in response to receiving a construction equipment control request from the user terminal containing the current final-level process identifier, construction area identifier, and current construction parameter information, a target historical construction data group corresponding to the current final-level process identifier is selected from the aforementioned historical construction data groups. Subsequently, in response to determining that the number of historical construction data in the target historical construction data group is less than or equal to a preset lower threshold, a lineage back-matching process is performed on the target historical construction data group based on the obtained generalized construction data set to obtain the target generalized construction data group. Therefore, when the sample size of the target historical construction data set is insufficient, supplementary sample data with a lineage relationship to the target historical construction data set can be obtained from the generalized construction data set through lineage back-matching to solve the problem of insufficient sample size causing the model to be unable to train under long-tail distribution. Next, based on the aforementioned target generalized construction data set, the initial construction control parameter prediction model is trained to obtain the construction control parameter prediction model.Therefore, in cases of insufficient sample size, lineage backmatching is employed to obtain training samples that meet the minimum sample size requirement. This avoids the problem of model failure to converge or severe overfitting due to insufficient samples, allowing the trained model to more accurately learn the true mapping relationship between construction control parameters and construction data. The aforementioned current construction parameter information is input into the aforementioned construction control parameter prediction model to drive the construction equipment corresponding to the aforementioned construction area identifier to perform construction operations. This results in a construction control parameter prediction model trained with sufficient samples, enabling it to predict more accurate construction control parameters (such as water-cement ratio, aggregate ratio, slump, and other key parameters). This allows construction equipment to perform construction operations according to control parameters that better reflect actual needs, thereby reducing the number of engineering quality accidents (such as reducing the number of engineering quality accidents caused by control parameters deviating from actual needs, such as substandard concrete strength, mixture segregation, or uncontrolled slump). Attached Figure Description

[0012] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0013] Figure 1 This is a flowchart of some embodiments of the construction equipment control method based on process hierarchy generalization aggregation according to this disclosure; Figure 2 This is a schematic diagram of the structure of some embodiments of the construction equipment control device based on process hierarchy generalization aggregation according to this disclosure; Figure 3 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0014] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0015] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0016] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0017] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0018] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0019] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] Figure 1 A flowchart 100 is shown, illustrating some embodiments of a construction equipment control method based on process-level generalization aggregation according to this disclosure. This construction equipment control method based on process-level generalization aggregation includes the following steps: Step 101: Obtain historical construction dataset.

[0021] In some embodiments, the execution entity of the construction equipment control method based on hierarchical generalization of work processes (e.g., computing devices such as industrial control computers, cloud servers, or edge computing gateways) can acquire historical construction datasets. Each historical construction data point in the aforementioned historical construction dataset includes multi-level work process attribute data with a tree-like hierarchical structure and construction data identifiers. The multi-level work process attribute data is a sequence of work process attribute identifiers. The multi-level work process attribute data can refer to a set of hierarchical labels describing the path of a construction task in a work process classification tree. The work process classification tree is a tree structure formed by decomposing construction tasks level by level from coarse to fine. Specifically, the multi-level work process attribute data is embodied in a sequence of work process attribute identifiers, which is a sequence obtained by arranging the identifiers corresponding to each level of attribute in the multi-level work process attribute data sequentially from the first level to the last level. For example, the work process attribute identifier sequence could be {Main Engineering, Concrete Engineering, Beam Casting, C30 Pumped Beam} (first level is "Main Engineering", second level is "Concrete Engineering", third level is "Beam Casting", and last level is "C30 Pumped Beam"). The aforementioned construction data identifier refers to the primary key index field (e.g., 001) that uniquely identifies a record of historical construction data. In practice, the aforementioned execution entity can read historical construction datasets from local persistent storage media (e.g., solid-state drives or relational databases).

[0022] Optionally, the aforementioned implementing entity may also perform the following steps: The first step, in response to the determination that the number of historical construction data in the aforementioned target historical construction data set is greater than a preset lower threshold, is to generate a training sample set based on the target historical construction data set. It should be noted that the method for generating the training sample set based on the target historical construction data set is the same as the method for generating the training sample set based on the aforementioned target generalized construction data set, and will not be repeated here.

[0023] The second step is to train the initial construction control parameter prediction model based on the above training sample set to obtain the construction control parameter prediction model.

[0024] Step 102: Perform homogeneous sample aggregation processing on the historical construction dataset to obtain various historical construction data groups.

[0025] In some embodiments, the aforementioned execution entity may perform process homogenization sample aggregation processing on the aforementioned historical construction dataset to obtain various historical construction data groups.

[0026] In addressing the technical problems mentioned above, the application scenario—automatic control of concrete mixing plants in the construction industry—often presents the following challenges: Historical construction data, grouped according to process attribute identifier sequences (i.e., the complete path from the first-level process to the final-level process), may originate from multiple different construction areas within the same sequence. These areas exhibit significant differences in environmental conditions (temperature, humidity). Simply aggregating the historical construction data based solely on the process attribute identifier sequence and then training a construction control parameter prediction model on this data group fails to extract the truly applicable process mapping relationship for the current area due to the presence of numerous samples from other areas. This results in output control parameters (aggregate ratio, water-cement ratio, mixing time, etc.) deviating from the actual material requirements of the area, increasing the frequency of engineering quality accidents. Therefore, this application scenario requires the following: introducing clustering filtering based on construction area identifiers during aggregation to retain pure samples from the most important construction areas within the same process, adapting to the actual working conditions of the current area, thereby reducing the risk of engineering quality accidents. Faced with the above technical problems, we decided to adopt the following solution: In some optional implementations of certain embodiments, the aforementioned execution entity may perform process homogeneous sample aggregation processing on the aforementioned historical construction dataset through the following steps to obtain various historical construction data groups: The first step is to divide the historical construction data with the same process attribute identifier sequence into initial historical construction data groups, thus obtaining each initial historical construction data group. Each initial historical construction data group can consist of at least one historical construction data with the exact same process attribute identifier sequence.

[0027] The second step involves performing the following steps for each initial historical construction data group within each initial historical construction data group.

[0028] The first sub-step involves identifying each construction data identifier in the aforementioned initial historical construction data group as a construction data identifier set.

[0029] The second sub-step involves obtaining the construction area identifier set corresponding to the aforementioned construction data identifier set. Each construction area identifier in the construction area identifier set corresponds one-to-one with a construction data identifier in the aforementioned construction data identifier set. The aforementioned construction area identifier describes the location where the construction task occurs. In practice, for each construction data identifier in the construction data identifier set, the executing entity can query the corresponding construction area identifier from a pre-set construction detail database. Then, the retrieved construction area identifiers can be defined as the construction area identifier set. The aforementioned pre-set construction detail database refers to a pre-built and persistently stored relational or non-relational database used to store the complete field information of each historical construction record. The complete field information of the historical construction record may include, but is not limited to, the following: construction data identifier, process attribute identifier sequence, construction area identifier, quantity of work, working condition characteristic information, and construction equipment control parameter information. The aforementioned working condition characteristic information describes the environmental conditions during the construction task (e.g., the working condition characteristic information could be an ambient temperature of 26 degrees Celsius). The aforementioned construction equipment control parameter information refers to the set of process parameters actually used in historical construction processes to control the operation of construction equipment (e.g., if the construction equipment is a concrete mixing plant, the construction equipment control parameter information for the concrete mixing plant may include the reference quantity of coarse aggregate, the reference quantity of fine aggregate, the reference quantity of cement, the reference weight of water, and the reference mixing time).

[0030] The third sub-step involves clustering the aforementioned set of construction area identifiers to obtain various construction area identifier clusters. Each of these clusters corresponds to at least one initial historical construction data point from the initial historical construction data set. In practice, the execution entity can employ hash grouping technology to group the construction area identifier set, grouping identifiers with the same identifier into a single group to obtain various construction area identifier clusters. Specifically, an empty hash map is created, using the construction area identifier as the key and a dynamic array as the value. Each identifier in the set is iterated through, and the hash map is queried using that identifier as the key. If the hash map does not exist, a new array is created and the current identifier is stored therein; if it already exists, the current identifier is appended to the existing array. After the iteration is complete, the array corresponding to each key in the hash map constitutes a construction area identifier cluster, and all identifier strings within the same array are identical.

[0031] The fourth sub-step involves determining at least one initial historical construction data point corresponding to the aforementioned construction area identifier cluster group within each of the above-mentioned construction area identifier cluster groups as a historical construction data group.

[0032] The above-mentioned technical solution, "combining step 104 and the first step, in response to determining that the number of historical construction data in the target historical construction data group is greater than a preset lower threshold, generates a training sample set based on the target historical construction data group; the second step, based on the training sample set, trains the initial construction control parameter prediction model to obtain the construction control parameter prediction model, and in combination with step 107, inputs the current construction parameter information into the construction control parameter prediction model to drive the construction equipment corresponding to the construction area identifier to perform construction operations," and its related content, serves as an inventive point of this disclosure, solving the technical problem that "the control parameters output by the construction control parameter prediction model deviate from the actual needs of the current area, leading to an increase in engineering quality accidents." Factors that cause the output control parameters of the construction control parameter prediction model to deviate from the actual needs of the current region, leading to an increase in engineering quality accidents, are often as follows: After historical construction data is grouped according to the process attribute identification sequence (i.e., the complete path from the first-level process to the final-level process), construction data under the same process attribute identification sequence may come from multiple different construction areas. These different areas have significantly different construction conditions and environmental conditions (temperature, humidity). Simply aggregating the target historical construction data group based solely on the process attribute identification sequence, when training the construction control parameter prediction model based on this target historical construction data group, results in the model being unable to extract the process mapping relationship truly applicable to the current region due to the large number of samples from other regions mixed in with the training data. This causes the output control parameters (aggregate ratio, water-cement ratio, mixing time, etc.) to deviate from the actual material requirements of the region, increasing the frequency of engineering quality accidents. Solving these factors can reduce the frequency of engineering quality accidents. To achieve this, firstly, historical construction data with the same process attribute identification sequence in the above historical construction dataset are divided into initial historical construction data groups, resulting in various initial historical construction data groups. Thus, historical construction data with completely identical process paths can be grouped together. Next, for each initial historical construction data group within each initial historical construction data group, the following steps are performed: Each construction data identifier in the aforementioned initial historical construction data group is determined as a construction data identifier set. Then, a construction area identifier set corresponding to the aforementioned construction data identifier set is obtained, wherein each construction area identifier in the aforementioned construction area identifier set corresponds one-to-one with each construction data identifier in the aforementioned construction data identifier set. Thus, the construction data identifier set corresponding to each initial historical construction data in the initial historical construction data group can be determined based on the construction data identifier set. Next, the aforementioned construction area identifier sets are clustered to obtain various construction area identifier clusters, wherein each of the aforementioned construction area identifier clusters corresponds to at least one initial historical construction data in the aforementioned initial historical construction data group.This allows for the differentiation of sample distribution across different construction areas under each process, with each cluster representing a specific construction area. Subsequently, for each of the aforementioned construction area identifier clusters, at least one initial historical construction data point corresponding to the aforementioned initial historical construction data group is identified as a historical construction data group. This splits the originally mixed initial historical construction data group into multiple independent historical construction data groups, each group sharing the same process and construction area. In step 104, in response to receiving a construction equipment control request from the user terminal containing the current final-level process identifier and construction area identifier, a target historical construction data group corresponding to the current final-level process identifier is selected from the aforementioned historical construction data groups that have undergone regional segmentation. Since each historical construction data group is already bound to a unique construction area, the data in the selected target historical construction data group comes from an area completely consistent with the construction area identifier in the current request, eliminating cross-regional noise. When the number of historical construction data in the target historical construction data group exceeds a preset lower threshold, a training sample set is directly generated based on this group, and the initial construction control parameter prediction model is trained to obtain the construction control parameter prediction model. Thus, the mapping relationship learned by the model is based on the historical real construction data of the region corresponding to the construction area identifier sent by the user terminal, which can more accurately reflect the technological patterns of the current region under a specific process and is not affected by differences in other regions. Combined with step 107, the above-mentioned current construction parameter information is input into the above-mentioned construction control parameter prediction model to drive the construction equipment corresponding to the above-mentioned construction area identifier to perform construction operations. Therefore, the current construction parameter information can be input into the construction control parameter prediction model that has learned to reflect the technological patterns of the current region under a specific process to control the construction equipment corresponding to the construction area identifier to perform construction operations, reducing the proportion deviation caused by cross-regional noise and reducing the number of engineering quality accidents.

[0033] Step 103: Based on the historical construction dataset, perform final-level process generalization aggregation processing on each historical construction data group to obtain the generalized construction data group set corresponding to the historical construction data group.

[0034] In some embodiments, the execution entity may perform final-level process generalization aggregation processing on each of the aforementioned historical construction data groups based on the aforementioned historical construction dataset to obtain a generalized construction data group set corresponding to the aforementioned historical construction data group.

[0035] In some optional implementations of certain embodiments, the aforementioned execution entity may perform final-level process generalization aggregation processing on each of the aforementioned historical construction data groups based on the aforementioned historical construction dataset to obtain a generalized construction data group set corresponding to the aforementioned historical construction data groups: The first step is to perform the following final-level process generalization and aggregation processing on each of the above historical construction data groups: The first sub-step is to determine the target quantity of historical construction data in the aforementioned historical construction data group.

[0036] The second sub-step involves determining, in response to the determination that the aforementioned target quantity is less than or equal to a preset value, a sequence of process attribute identifiers included in one historical construction data set within the aforementioned historical construction data group, as a reference process attribute identifier sequence. Here, the aforementioned preset value refers to a pre-set lower limit threshold for the sample size, such as 50 or 500 records.

[0037] The third sub-step, based on the reference process attribute identifier sequence, performs the following generation steps: Sub-step one involves identifying at least one historical construction data set containing the last reference process attribute identifier in the reference process attribute identifier sequence within the historical construction dataset as the generalized construction data set corresponding to the aforementioned historical construction data set. For example, if the reference process attribute identifier sequence is {Main Structure, Concrete Engineering, Beam Casting, C30 Pumped Beam}, then the last identifier is “C30 Pumped Beam”. The historical construction dataset can be {Historical Construction Data 1, Historical Construction Data 2, Historical Construction Data 3}. Historical Construction Data 1 contains the process attribute identifier sequence {Main Structure, Concrete Engineering, Beam Casting, C30 Pumped Beam}. Historical Construction Data 2 contains the process attribute identifier sequence {Main Structure, Concrete Engineering, Beam Casting, C30 Pumped Beam, Adding Water-Reducing Agent}. Historical Construction Data 3 contains the process attribute identifier sequence {Main Structure, Concrete Engineering, Beam Casting, C35 Pumped Beam}. Therefore, the generalized construction data set can be {Historical Construction Data 1, Historical Construction Data 2} containing “C30 Pumped Beam”.

[0038] Sub-step two involves deleting the last reference process attribute identifier from the reference process attribute identifier sequence in order to update the reference process attribute identifier sequence.

[0039] Sub-step three: In response to the determination that the updated reference process attribute identifier sequence is not empty, the above generation steps are executed again based on the updated reference process attribute identifier sequence.

[0040] The second step is to determine that the updated reference process attribute identifier sequence is empty, and to identify at least one generalized construction data group as the generalized construction data group set corresponding to the above-mentioned historical construction data group.

[0041] Step 104: In response to receiving a construction equipment control request from the user terminal containing the current final-level process identifier, construction area identifier, and current construction parameter information, filter the target historical construction data group corresponding to the current final-level process identifier from each historical construction data group.

[0042] In some embodiments, the aforementioned execution entity may, in response to receiving a construction equipment control request from a user terminal containing the current final-level process identifier, construction area identifier, and current construction parameter information, filter the target historical construction data group corresponding to the current final-level process identifier from the aforementioned historical construction data groups. Here, the aforementioned user terminal refers to the front-end device of the construction management system, such as a tablet computer at the construction site, a desktop computer in the dispatch center, or a touch terminal integrated into the operation panel of a concrete mixing plant. The aforementioned construction equipment control request refers to a digital instruction issued by the aforementioned user terminal containing the current final-level process identifier, construction area identifier, and current construction parameter information. The aforementioned current final-level process identifier refers to the unique identifier of the finest-grained construction process targeted by the user terminal in this request, such as "C30 pumped beam" or "basement floor C35 impermeable concrete". The aforementioned construction area identifier refers to the geographical location label of the current construction, such as "Building A - Zone B - 3rd floor". The aforementioned current construction parameter information refers to information collected in real time for this construction task, containing the current process attribute identifier sequence and current working condition characteristic information. The aforementioned current process attribute identifier sequence can be a set of process attribute identifiers corresponding to the current construction. In practice, for each of the aforementioned historical construction data groups, the executing entity can determine the last process attribute identifier in the sequence of process attribute identifiers included in a historical construction data set as the process attribute identifier to be compared. The construction area identifier corresponding to the historical construction data can be determined as the construction area identifier to be compared. Then, the executing entity can compare the process attribute identifier to be compared with the current final-level process identifier and compare the construction area identifier to be compared with the area identifier sent by the user. In response to determining that the process attribute identifier to be compared is the same as the current final-level process identifier and that the construction area identifier to be compared is the same as the area identifier sent by the user, the executing entity can determine the aforementioned historical construction data group as the target historical construction data group corresponding to the current final-level process identifier.

[0043] Step 105: In response to determining that the number of historical construction data in the target historical construction data group is less than or equal to a preset lower threshold, based on the obtained generalized construction data group sets, the target historical construction data group is subjected to lineage back-tracing matching to obtain the target generalized construction data group.

[0044] In some embodiments, the execution entity may, in response to determining that the number of historical construction data in the target historical construction data group is less than or equal to a preset lower threshold, perform lineage tracing matching on the target historical construction data group based on the obtained generalized construction data group sets to obtain target generalized construction data groups. Each historical construction data group corresponds to one generalized construction data group set in the generalized construction data group sets.

[0045] In some optional implementations of certain embodiments, the aforementioned execution entity may perform lineage tracing matching on the aforementioned target historical construction data set based on the obtained various generalized construction data set sets to obtain the target generalized construction data set: The first step is to determine the set of generalized construction data groups corresponding to the target historical construction data groups as the set of generalized construction data groups to be screened.

[0046] The second step is to sort the selected generalized construction data groups in ascending order according to the number of generalized construction data contained in each data group in the selected generalized construction data group set, so as to obtain the sequence of selected generalized construction data groups, where each selected generalized construction data has a lineage pointer identifier.

[0047] The third step is to iterate through the above sequence of filtered generalized construction data groups and determine the first filtered generalized construction data group that meets the preset filtering conditions as the target generalized construction data group. The preset filtering conditions can be that the number of generalized construction data points in the generalized construction data group is greater than or equal to the preset lower threshold.

[0048] Step 106: Based on the target generalized construction data set, train the initial construction control parameter prediction model to obtain the construction control parameter prediction model.

[0049] In some embodiments, the aforementioned execution entity can train an initial construction control parameter prediction model based on the aforementioned target generalized construction data set to obtain a construction control parameter prediction model. Each target generalized construction data in the aforementioned target generalized construction data set includes a process attribute identifier sequence and a construction data identifier. The aforementioned initial construction control parameter prediction model refers to a machine learning model (e.g., a multilayer perceptron (MLP) model or a convolutional neural network (CNN) model) that has not yet been trained or whose parameters are in an initial state (e.g., randomly initialized), whose input is construction parameter information and whose output is control parameter information. The aforementioned control parameter information can be a set of process parameters predicted based on the construction parameter information for controlling the operation of construction equipment. The control parameter information may include the reference quantity of coarse aggregate, the reference quantity of fine aggregate, the reference quantity of cement, the reference weight of water, and the reference mixing time.

[0050] In some optional implementations of certain embodiments, the aforementioned execution entity may train the initial construction control parameter prediction model based on the aforementioned target generalized construction data set through the following steps to obtain the construction control parameter prediction model: The first step is to generate a training sample set based on the aforementioned target generalized construction data set. Each training sample in the training sample set includes sample construction parameter information and sample target control parameter information.

[0051] The second step is to train the initial construction control parameter prediction model based on the above training sample set to obtain the construction control parameter prediction model.

[0052] In some optional implementations of certain embodiments, the aforementioned execution entity can generate a training sample set based on the aforementioned target generalized construction data set through the following steps: The first step is to perform the following steps for each target generalization construction data in the above target generalization construction data group: The first sub-step involves determining the sequence of process attribute identifiers included in the aforementioned target generalized construction data as the target process attribute identifier sequence. Here, the aforementioned target process attribute identifier sequence refers to the sequence of process attribute identifiers stored in the target generalized construction data, from the first-level process to the final-level process.

[0053] The second sub-step involves retrieving the working condition characteristic information and construction equipment control parameter information corresponding to the construction data identifiers included in the aforementioned target generalized construction data from the preset construction detail database. In practice, the aforementioned executing entity can send a query request to the aforementioned preset construction detail database based on the aforementioned construction data identifier (e.g., 001) to retrieve the working condition characteristic information and construction equipment control parameter information corresponding to the construction data identifier from the preset construction detail database.

[0054] The third sub-step involves combining the target process attribute identifier sequence and the working condition feature information to obtain sample construction parameter information. In practice, the executing entity first encodes the target process attribute identifier sequence, converting it into a numerical vector as the first vector. Then, it encodes the working condition feature information to obtain a numerical vector corresponding to the working condition feature information as the second vector. Next, the first vector and the second vector can be concatenated, and the concatenated vector is determined as the sample construction parameter information.

[0055] The fourth sub-step involves determining the aforementioned construction equipment control parameter information as the sample target control parameter information corresponding to the aforementioned sample construction parameter information.

[0056] The fifth sub-step involves determining the above-mentioned sample construction parameter information and the above-mentioned sample target control parameter information as training samples.

[0057] The second step is to determine the obtained training samples as the training sample set.

[0058] In some optional implementations of certain embodiments, the aforementioned execution entity may train the aforementioned initial construction control parameter prediction model based on the aforementioned training sample set through the following steps to obtain the construction control parameter prediction model: The first step involves performing the following training steps based on at least one training sample in the training sample set: The first sub-step involves inputting at least one sample construction parameter information from at least one training sample into the initial construction control parameter prediction model to obtain the sample prediction control parameter information corresponding to each training sample in at least one training sample.

[0059] The second sub-step involves comparing the predicted control parameters of each training sample in at least one training sample with the corresponding target control parameters. In practice, for each training sample in at least one training sample, the difference between the predicted control parameters and the target control parameters can be calculated using a preset loss function (e.g., mean squared error loss function MSE or mean absolute error loss function MAE). The mean of these difference values ​​is then used as the overall difference value for comparison. This difference value reflects the degree of deviation between the model's predicted value (predicted control parameters) and the true value (target control parameters). A larger difference value indicates a less accurate prediction, while a smaller difference value indicates a prediction closer to the true value.

[0060] The third sub-step involves determining, based on the comparison results, whether the initial construction control parameter prediction model has achieved the preset optimization objective. This preset optimization objective can be that the overall difference value represented by the comparison results is less than a preset loss threshold.

[0061] The fourth sub-step, in response to determining that the initial construction control parameter prediction model has reached the optimization objective, uses the initial construction control parameter prediction model as the trained construction control parameter prediction model. In practice, when the executing entity determines that the model has reached the optimization objective, it terminates the training loop and saves all network parameters of the current initial construction control parameter prediction model (including the weight matrices and bias vectors of each layer) to non-volatile storage media (e.g., a preset file on a hard drive or uploading the model to a preset model repository). Simultaneously, the executing entity identifies this model as the construction control parameter prediction model and loads it into memory, and marks the state of the aforementioned construction control parameter prediction model as ready.

[0062] The fifth sub-step, in response to the determination that the initial construction control parameter prediction model has not achieved the optimization objective, involves adjusting the network parameters of the initial construction control parameter prediction model, using at least one unused training sample from the training sample set, and then using the adjusted initial construction control parameter prediction model as the new initial construction control parameter prediction model, and repeating the above training steps. As an example, the back propagation algorithm (BP algorithm) and gradient descent methods (such as mini-batch gradient descent) can be used to adjust the network parameters of the aforementioned initial construction control parameter prediction model.

[0063] Step 107: Input the current construction parameter information into the construction control parameter prediction model to drive the construction equipment corresponding to the construction area identifier to perform construction operations.

[0064] In some embodiments, the executing entity can input the current construction parameter information into the construction control parameter prediction model to drive the construction equipment corresponding to the construction area identifier to perform construction operations. The current construction parameter information includes a current process attribute identifier sequence and current working condition characteristic information. The last current process attribute identifier in the current process attribute identifier sequence is a pouring process identifier, and the construction equipment corresponding to the construction area identifier is a concrete mixing device. The pouring process identifier can be an identifier representing the type of concrete pouring process (e.g., C30 pumped beam).

[0065] In the process of using the above-mentioned technical solutions to solve the technical problems mentioned in the background of this disclosure, the following technical problems often arise in the application scenario: automated production control of concrete mixing plants in the construction field. The benchmark parameters output by the construction control parameter prediction model (such as benchmark coarse aggregate quantity, mixing time, etc.) are ideal process values ​​learned from historical data, without considering the current actual physical state of the concrete mixing equipment. For example, zero drift of the weighing sensor leads to weighing deviation, excessively high motor winding temperature leads to a shortened continuous running time, and increased vibration of the reducer leads to a decrease in the safe weighing value. If the benchmark parameters output by the model are directly used to drive the equipment without adjustment, it is easy to cause a single weighing to exceed the safe weighing value (causing weighing failure or loss of accuracy), or the total continuous mixing time to exceed the continuous running time of the motor (causing overheat protection shutdown or even motor burnout), thereby increasing the risk of damage to the construction equipment, i.e., the concrete mixing equipment. The following requirements are necessary for this application scenario: the concrete mixing plant needs to produce concrete stably and safely; the state of the concrete mixing equipment changes dynamically with usage time and environmental conditions; and the control of the construction equipment, i.e., the concrete mixing equipment, requires the ability to sense the equipment status and adaptively adjust parameters. To address these technical challenges, we have decided to adopt the following solution: In some optional implementations of certain embodiments, the execution entity may input the current construction parameter information into the construction control parameter prediction model through the following steps to drive the construction equipment corresponding to the construction area identifier to perform construction operations: The first step is to input the current construction parameter information into the construction control parameter prediction model to obtain the control parameter information. This control parameter information includes the benchmark quantities of coarse aggregate, fine aggregate, cement, water, and mixing time. The benchmark quantity of coarse aggregate refers to the total weight of coarse aggregate (usually crushed stone or gravel) predicted by the model for this construction, in kilograms (kg). The benchmark quantity of fine aggregate refers to the total weight of fine aggregate (usually natural sand or manufactured sand) predicted by the model for this construction, in kilograms. The benchmark quantity of cement refers to the total weight of cement predicted by the model for this construction, in kilograms. The benchmark weight of water refers to the total weight of water predicted by the model for this construction, in kilograms. The benchmark mixing time refers to the total time predicted by the model for uniformly mixing all materials under standard conditions, in seconds (s). In practice, the aforementioned executing entity encodes the sequence of process attribute identifiers in the current construction parameter information (e.g., integer mapping) and concatenates it with the numerical values ​​of the working condition feature information to form a feature vector, which is then input into the construction control parameter prediction model to obtain the control parameter information.

[0066] The second step is to obtain the equipment status parameter information of the construction equipment corresponding to the aforementioned construction area identification. This equipment status parameter information includes the current zero drift value of the weighing sensor, the current winding temperature of the mixing motor, the current vibration amplitude of the reducer, and the cumulative production volume. These equipment status parameters reflect the current operating status and health of the mixing equipment. The current zero drift value of the weighing sensor can be the value corresponding to the non-zero point output by the weighing sensor in the concrete mixing equipment when unloaded. The current winding temperature of the mixing motor can be the current temperature of the motor. The current vibration amplitude of the reducer can reflect gearbox wear or bearing failure. The cumulative production weight represents the total weight of concrete produced by the equipment since it was put into operation.

[0067] The third step involves inputting the aforementioned equipment status parameter information into a preset equipment performance degradation model to obtain equipment capacity constraint information. This constraint information includes a safe weighing value and a motor continuous operating time. The preset equipment performance degradation model is a pre-defined mathematical model (e.g., linear regression, exponential decay model, or physics-based wear model). Its input is the equipment status parameter information, and its output is information related to the safe operating limits of each key component of the equipment under the current operating conditions—that is, the equipment capacity constraint information. The safe weighing value can be the maximum total weight of material that the weighing sensor in the construction equipment, i.e., the concrete mixing equipment, can safely and accurately weigh in a single operation under the current zero drift and wear conditions. The motor continuous operating time can be the maximum duration for which the mixing motor can operate continuously at the current winding temperature without exceeding the maximum allowable temperature of the insulation class. As an example, the aforementioned equipment performance degradation model can calculate the current safe weighing value and the motor continuous operating time based on built-in degradation rules (e.g., for every 10,000 kg increase in cumulative production weight, the safe weighing value decreases by 1%; for every 10°C increase in motor temperature, the continuous operating time decreases by 15%). As an example, the equipment performance degradation model can be set with the following rules: the initial safe weighing value is set to 5000kg, the initial motor operating time is set to 120 seconds, and the safe weighing value decreases by 1% for every 10000kg increase in cumulative production weight; the motor operating time decreases by 15% for every 10℃ increase in motor winding temperature relative to ambient temperature (20℃). Assuming the currently read equipment status parameters are: cumulative production weight 85000kg, motor winding temperature 78℃, the model first calculates the degradation caused by the cumulative production weight: 85000kg ÷ 10000kg × 1% = 8.5%, and the current safe weighing value = 5000kg × (100% - 8.5%) = 4575kg. Then, it calculates the degradation caused by temperature: temperature rise (78℃ - 20℃) ÷ 10℃ × 15% = 87%, and the current motor operating time = 120 seconds × (100% - 87%) = 15.6 seconds. The model outputs a safe weighing value of 4575 kg, and the motor can be used continuously for 15.6 seconds.

[0068] The fourth step is to determine the total weight by summing the above-mentioned benchmark quantities of coarse aggregate, fine aggregate, cement, and water.

[0069] Fifth, based on the total weight and the safety weighing value mentioned above, generate batch mixing information, single batch coarse aggregate weight, single batch fine aggregate weight, single batch cement weight, and single batch water weight. In practice, the executing entity can divide the total weight by the safety weighing value and round up to the nearest integer, using this rounded value as the number of mixing cycles for batch mixing information. Next, the executing entity can determine the single batch coarse aggregate weight by the ratio of the reference coarse aggregate quantity to the number of mixing cycles. The ratio of the reference fine aggregate quantity to the number of mixing cycles can be used to determine the single batch fine aggregate weight. The ratio of the reference cement quantity to the number of mixing cycles can be used to determine the single batch cement weight. Finally, the ratio of the single batch water weight to the number of mixing cycles determines the single batch water weight.

[0070] Step 6: Based on the batch mixing information and the baseline mixing time, generate the total mixing time. In practice, the executing entity can determine the total mixing time by multiplying the number of mixing cycles represented by the batch mixing information by the baseline mixing time.

[0071] Step 7: In response to determining that the total stirring time is greater than the continuous motor duration, a stirring intermittent cooling interval is generated based on the total stirring time, the continuous motor duration, and the stirring batch information. The stirring intermittent cooling interval refers to the pause time inserted between every two batches during continuous stirring due to the temperature limitations of the motor or reducer (this interval is set to prevent heat accumulation in the motor or reducer from exceeding the allowable range, extending equipment life and ensuring safety). In practice, the executing entity can input the total stirring time, the continuous motor duration, and the stirring batch information into a preset stirring intermittent cooling interval formula to obtain the stirring intermittent cooling interval. The preset stirring intermittent cooling interval formula can be: Stirring intermittent cooling interval = ((Total stirring time - Continuous motor duration) ÷ (Number of batches represented by stirring batch information - 1)) × Preset time constant. The preset time constant can be a preset thermal time constant related to the equipment's heat dissipation characteristics.

[0072] Step 8: Based on the aforementioned batch mixing information, the aforementioned single batch coarse aggregate weight, the aforementioned single batch fine aggregate weight, the aforementioned single batch cement weight, the single batch water weight, the reference mixing time, and the aforementioned mixing interval cooling interval, a cyclic control command sequence and a cooling waiting command are generated. The aforementioned cooling waiting command is inserted between every two adjacent cyclic control commands in the cyclic control command sequence to drive the construction equipment corresponding to the aforementioned construction area identifier to perform construction operations. Each cyclic control command in the generated cyclic control command sequence refers to a set of digital commands used to control a complete construction operation (such as a single batching, mixing, and discharging), including the following fields: single batch coarse aggregate weight, single batch fine aggregate weight, single batch cement weight, single batch water weight, and reference mixing time. The aforementioned cooling waiting command can be a command to control the construction equipment to stop operating for a duration equal to the mixing interval cooling interval. In practice, each cyclic control command in the cyclic control command sequence controls the coarse aggregate bins, fine aggregate bins, cement bins, and water pumps of the construction equipment (i.e., the concrete mixing plant) to batch materials sequentially or in parallel according to the weight indicated by the cyclic control command. Weighing sensors provide real-time weight feedback. Once a safe weighing value is reached, the valves of the coarse aggregate bins, fine aggregate bins, cement bins, and water pumps are closed, and the mixing motor is started. The mixing time is maintained for the baseline mixing duration indicated by the cyclic control command. After mixing is complete, the discharge gate is opened, and the concrete is unloaded into a transport vehicle or placing equipment. When a cooling wait command is executed, the construction equipment is stopped, and the stop time is the mixing intermittent cooling interval.

[0073] The above-described technical solution and its related content, as an inventive point of this disclosure, solve the technical problem of "increased risk of damage to construction equipment, i.e., concrete mixing equipment." Factors leading to increased risk of damage to construction equipment, i.e., concrete mixing equipment, often include: the baseline parameters output by the construction control parameter prediction model (such as the baseline amount of coarse aggregate, mixing time, etc.) are ideal process values ​​learned from historical data, without considering the current actual physical state of the concrete mixing equipment. For example, zero drift of the weighing sensor leads to weighing deviation, excessively high motor winding temperature leads to a shortened continuous operating time, and increased vibration of the reducer leads to a decrease in the safe weighing value. If the baseline parameters output by the model are directly used to drive the equipment without adjustment, it is easy to cause a single weighing to exceed the safe weighing value (causing weighing failure or loss of accuracy), or the total continuous mixing time to exceed the continuous operating time of the motor (causing overheat protection shutdown or even motor burnout), thereby increasing the risk of damage to the construction equipment, i.e., concrete mixing equipment. Solving these factors can reduce the risk of damage to construction equipment, i.e., concrete mixing equipment. To achieve this effect, firstly, the current construction parameter information is input into the construction control parameter prediction model to obtain control parameter information, which includes the reference quantity of coarse aggregate, the reference quantity of fine aggregate, the reference quantity of cement, the reference weight of water, and the reference mixing time. This yields the control parameter information used to control the concrete mixing equipment. Next, the equipment status parameter information of the construction equipment corresponding to the construction area identifier is obtained, including the current zero drift value of the weighing sensor, the current winding temperature of the mixing motor, the current vibration amplitude of the reducer, and the cumulative production weight. This provides equipment status parameter information characterizing the current physical health state of the construction equipment. Then, the equipment status parameter information is input into a preset equipment performance degradation model to obtain equipment capacity constraint information, including the safe weighing value and the continuous operating time of the motor. This determines the maximum safe value for a single weighing under the current equipment condition and the maximum allowable continuous operating time of the motor. Finally, the sum of the reference quantities of coarse aggregate, fine aggregate, cement, and water is determined as the total weight. Based on the total weight and the safety weighing value, batch mixing information, the weight of coarse aggregate in a single batch, the weight of fine aggregate in a single batch, the weight of cement in a single batch, and the weight of water in a single batch are generated. This allows the total weight exceeding the safety weighing value to be automatically broken down into multiple sub-batches not exceeding the safety weighing value, ensuring that each batch of materials does not overload the weighing sensor. Next, based on the batch mixing information and the baseline mixing time, the total mixing time is generated. Then, in response to determining that the total mixing time is greater than the continuous motor duration, a mixing interval cooling interval is generated based on the total mixing time, the continuous motor duration, and the batch mixing information.Therefore, when the total continuous mixing time may cause the motor to overheat, the required cooling waiting time between each batch is automatically calculated. Finally, based on the above mixing batch information, the above single batch coarse aggregate weight, the above single batch fine aggregate weight, the above single batch cement weight, the single batch water weight, the reference mixing time, and the above mixing interval cooling interval, a cyclic control command sequence and a cooling waiting command are generated. The cooling waiting command is inserted between every two adjacent cyclic control commands in the cyclic control command sequence to drive the construction equipment corresponding to the above construction area identifier to perform construction operations. Thus, the equipment will perform batching and mixing according to the safe weight after batching, and pause cooling between every two batches to keep the motor temperature within the allowable range. Also, because it adopts the method of acquiring equipment status parameters, dynamically determining the safe weighing value and the motor's sustainable duration through the equipment performance degradation model, and automatically batching according to the relationship between the total weight and the safe weighing value, and automatically inserting cooling waiting intervals according to the relationship between the total mixing time and the motor's sustainable duration, the risk of damage to the weighing sensor caused by single weighing overload and the risk of motor burnout caused by continuous mixing overheating are reduced, thus lowering the risk of damage to the concrete mixing equipment.

[0074] The above-described embodiments of this disclosure have the following beneficial effects: the construction equipment control method based on process-level generalization aggregation in some embodiments of this disclosure reduces the number of engineering quality accidents. Specifically, the reason for the increase in the number of engineering quality accidents is that, due to the highly dispersed and diverse nature of construction tasks, historical construction data, after being divided according to the final-level process, exhibits a typical long-tail distribution, meaning that the number of historical samples for a large number of final-level process items (such as the "C30 pumping beam" process in a remote area) is extremely small, far from meeting the minimum sample size requirement for training machine learning models (e.g., at least 500 samples). When the sample size is insufficient, directly training the model on the final-level process item will lead to the model failing to converge or severe overfitting, thereby causing the predicted control parameters to deviate from the actual requirements, increasing the number of engineering quality accidents. Based on this, the construction equipment control method based on process-level generalization aggregation in some embodiments of this disclosure first obtains a historical construction dataset. Then, the historical construction dataset is subjected to process homogeneous sample aggregation processing to obtain various historical construction data groups. Historical construction data with similar process characteristics (such as having completely identical process attribute identification sequences and construction areas) can be grouped together. Next, based on the aforementioned historical construction datasets, a final-level process generalization aggregation process is performed on each of the aforementioned historical construction data groups to obtain a generalized construction data set corresponding to the aforementioned historical construction data groups. Thus, by progressively removing the attributes of the final-level process and re-aggregating, a multi-gradient ancestor generalized data set corresponding to the historical construction data groups—that is, a generalized construction data set—can be pre-constructed without introducing any virtual data, forming a backup data pool. Then, in response to receiving a construction equipment control request from the user terminal containing the current final-level process identifier, construction area identifier, and current construction parameter information, a target historical construction data group corresponding to the current final-level process identifier is selected from the aforementioned historical construction data groups. Subsequently, in response to determining that the number of historical construction data in the target historical construction data group is less than or equal to a preset lower threshold, a lineage back-matching process is performed on the target historical construction data group based on the obtained generalized construction data set to obtain the target generalized construction data group. Therefore, when the sample size of the target historical construction data set is insufficient, supplementary sample data with a lineage relationship to the target historical construction data set can be obtained from the generalized construction data set through lineage back-matching to solve the problem of insufficient sample size causing the model to be unable to train under long-tail distribution. Next, based on the aforementioned target generalized construction data set, the initial construction control parameter prediction model is trained to obtain the construction control parameter prediction model.Therefore, in cases of insufficient sample size, lineage backmatching is employed to obtain training samples that meet the minimum sample size requirement. This avoids the problem of model failure to converge or severe overfitting due to insufficient samples, allowing the trained model to more accurately learn the true mapping relationship between construction control parameters and construction data. The aforementioned current construction parameter information is input into the aforementioned construction control parameter prediction model to drive the construction equipment corresponding to the aforementioned construction area identifier to perform construction operations. This results in a construction control parameter prediction model trained with sufficient samples, enabling it to predict more accurate construction control parameters (such as water-cement ratio, aggregate ratio, slump, and other key parameters). This allows construction equipment to perform construction operations according to control parameters that better reflect actual needs, thereby reducing the number of engineering quality accidents (such as reducing the number of engineering quality accidents caused by control parameters deviating from actual needs, such as substandard concrete strength, mixture segregation, or uncontrolled slump).

[0075] Further reference Figure 2 As an implementation of the methods shown in the figures, this disclosure provides some embodiments of a construction equipment control device based on process-level generalization aggregation. These device embodiments are similar to... Figure 1 Corresponding to the method embodiments shown, the device can be specifically applied to various electronic devices.

[0076] like Figure 2As shown, a construction equipment control device 200 based on process-level generalization aggregation in some embodiments includes: an acquisition unit 201, an aggregation unit 202, a generalization aggregation processing unit 203, a filtering unit 204, a lineage backtracking matching processing unit 205, a training unit 206, and a driving unit 207. The acquisition unit 201 is configured to acquire historical construction datasets; the aggregation unit 202 is configured to perform process-level homogeneous sample aggregation processing on the historical construction datasets to obtain various historical construction data groups; the generalization aggregation processing unit 203 is configured to perform final-level process generalization aggregation processing on each of the historical construction data groups based on the historical construction datasets to obtain a set of generalized construction data groups corresponding to the historical construction data groups; the filtering unit 204 is configured to, in response to receiving a construction equipment control request from a user terminal containing the current final-level process identifier, construction area identifier, and current construction parameter information, filter data from the various historical construction data groups that correspond to the current final-level process. The sequence identifier corresponds to the target historical construction data group; the lineage backtracking matching processing unit 205 is configured to, in response to determining that the number of historical construction data in the target historical construction data group is less than or equal to a preset lower threshold, perform lineage backtracking matching processing on the target historical construction data group based on the obtained generalized construction data group sets, to obtain the target generalized construction data group; the training unit 206 is configured to, based on the target generalized construction data group, train the initial construction control parameter prediction model to obtain the construction control parameter prediction model; the driving unit 207 is configured to input the current construction parameter information into the construction control parameter prediction model to drive the construction equipment corresponding to the construction area identifier to perform construction operations.

[0077] It is understandable that the units described in the device 200 are related to the reference. Figure 1 The steps in the method described above correspond to each other. Therefore, the operations, features, and beneficial effects described above for the method also apply to the device 200 and the units contained therein, and will not be repeated here.

[0078] The following is for reference. Figure 3 It shows a schematic diagram of the structure of an electronic device 300 suitable for implementing some embodiments of the present disclosure. Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0079] like Figure 3As shown, the electronic device 300 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0080] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 3 Each box shown can represent a device or multiple devices as needed.

[0081] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 309, or installed from storage device 308, or installed from ROM 302. When the computer program is executed by processing device 301, it performs the functions defined in the methods of some embodiments of this disclosure.

[0082] It should be noted that, in some embodiments of this disclosure, the computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0083] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0084] The computer-readable medium may be contained within an electronic device or may exist independently, not assembled into the electronic device. The computer-readable medium carries one or more programs that, when executed by the electronic device, cause the electronic device to: acquire a historical construction dataset; perform process homogenization sample aggregation processing on the historical construction dataset to obtain various historical construction data groups; based on the historical construction dataset, perform final-level process generalization aggregation processing on each of the historical construction data groups to obtain a generalized construction data group set corresponding to the historical construction data group; and, in response to receiving a construction equipment control request from a user terminal containing the current final-level process identifier, construction area identifier, and current construction parameter information, acquire the aforementioned historical construction data dataset. From the historical construction data set, a target historical construction data set corresponding to the current final-level process identifier is selected; in response to determining that the number of historical construction data in the target historical construction data set is less than or equal to a preset lower threshold, based on the obtained generalized construction data set, a lineage back-matching process is performed on the target historical construction data set to obtain a target generalized construction data set; based on the target generalized construction data set, an initial construction control parameter prediction model is trained to obtain a construction control parameter prediction model; the current construction parameter information is input into the construction control parameter prediction model to drive the construction equipment corresponding to the construction area identifier to perform construction operations.

[0085] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—and conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0086] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0087] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including an acquisition unit, an aggregation unit, a generalization aggregation processing unit, a filtering unit, a lineage backtracking matching processing unit, a training unit, and a driving unit. The names of these units do not necessarily limit the unit itself; for example, the acquisition unit may also be described as "a unit for acquiring historical construction datasets."

[0088] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0089] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of technical features, but should also cover other technical solutions formed by arbitrary combinations of technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A construction equipment control method based on process-level generalization and aggregation, characterized by: Obtain historical construction datasets; The historical construction dataset is subjected to process homogeneous sample aggregation to obtain various historical construction data groups; Based on the historical construction dataset, each historical construction data group in each historical construction data group is subjected to final-level process generalization aggregation processing to obtain a generalized construction data group set corresponding to the historical construction data group. In response to receiving a construction equipment control request from the user terminal containing the current final-level process identifier, construction area identifier, and current construction parameter information, the target historical construction data group corresponding to the current final-level process identifier is selected from the various historical construction data groups; In response to determining that the number of historical construction data in the target historical construction data group is less than or equal to a preset lower threshold, based on the obtained generalized construction data group sets, the target historical construction data group is subjected to lineage tracing matching to obtain the target generalized construction data group. Based on the target generalized construction data set, the initial construction control parameter prediction model is trained to obtain the construction control parameter prediction model. The current construction parameter information is input into the construction control parameter prediction model to drive the construction equipment corresponding to the construction area identifier to perform construction operations.

2. The method of claim 1, wherein, The method of training the initial construction control parameter prediction model based on the target generalized construction data set to obtain the construction control parameter prediction model is characterized by: Based on the target generalized construction data set, a training sample set is generated; Based on the training sample set, the initial construction control parameter prediction model is trained to obtain the construction control parameter prediction model.

3. The method of claim 2, wherein, Each target generalized construction data in the target generalized construction data set includes a process attribute identifier sequence and a construction data identifier. The step of generating a training sample set based on the target generalized construction data set is characterized by: For each target generalized construction data in the target generalized construction data group, perform the following steps: The sequence of process attribute identifiers included in the target generalized construction data is determined as the target process attribute identifier sequence; Obtain working condition characteristic information and construction equipment control parameter information corresponding to the construction data identifiers included in the target generalized construction data from the preset construction detail database; The target process attribute identifier sequence and the working condition feature information are combined to obtain sample construction parameter information; The construction equipment control parameter information is determined as the sample target control parameter information corresponding to the sample construction parameter information; The sample construction parameter information and the sample target control parameter information are used as training samples; The obtained training samples are defined as the training sample set.

4. The method of claim 2, wherein, Each training sample in the training sample set includes sample construction parameter information and sample target control parameter information. The initial construction control parameter prediction model is trained based on the training sample set to obtain the construction control parameter prediction model, characterized by: Perform the following training steps based on at least one training sample in the training sample set: Input at least one sample construction parameter information from at least one training sample into the initial construction control parameter prediction model to obtain sample prediction control parameter information corresponding to each training sample in at least one training sample. Compare the sample prediction control parameter information corresponding to each training sample in at least one training sample with the corresponding sample target control parameter information; Based on the comparison results, determine whether the initial construction control parameter prediction model has achieved the preset optimization target; In response to the determination that the initial construction control parameter prediction model has reached the optimization objective, the initial construction control parameter prediction model is used as the completed construction control parameter prediction model. In response to the determination that the initial construction control parameter prediction model has not achieved the optimization objective, the network parameters of the initial construction control parameter prediction model are adjusted, and at least one training sample that has not been used in the training sample set is used as the initial construction control parameter prediction model, and the training steps are performed again.

5. The method of claim 1, wherein, Each historical construction data point in the historical construction dataset includes multi-level process attribute data with a tree-like hierarchical structure and construction data identifiers. The multi-level process attribute data is a sequence of process attribute identifiers. Furthermore, based on the historical construction dataset, a final-level process generalization aggregation process is performed on each historical construction data group to obtain a generalized construction data group set corresponding to the historical construction data group. The characteristic of this process is: For each of the aforementioned historical construction data groups, the following final-level process generalization aggregation process is performed: The number of historical construction data in the historical construction data group is determined as the target number; In response to determining that the target quantity is less than or equal to a preset value, the process attribute identifier sequence included in one of the historical construction data in the historical construction data group is determined as a reference process attribute identifier sequence; Based on the reference process attribute identifier sequence, perform the following generation steps: The historical construction data set containing at least one reference process attribute identifier in the last reference process attribute identifier sequence is identified as the generalized construction data set corresponding to the historical construction data set. Remove the last reference process attribute identifier from the reference process attribute identifier sequence to update the reference process attribute identifier sequence; In response to the determination that the updated reference process attribute identifier sequence is not empty, the generation step is performed again based on the updated reference process attribute identifier sequence; In response to determining that the updated reference process attribute identifier sequence is empty, at least one generalized construction data group is identified as the generalized construction data group set corresponding to the historical construction data group.

6. The method of claim 1, wherein, Each historical construction data group in the various historical construction data groups corresponds to a set of generalized construction data groups in the various generalized construction data group sets, and the target historical construction data group is subjected to lineage tracing matching based on the obtained sets of generalized construction data groups to obtain the target generalized construction data group, characterized in that: The generalized construction data set corresponding to each generalized construction data set and the target historical construction data set is determined as the screened generalized construction data set; Based on the number of generalized construction data contained in each data group in the generalized construction data set, the generalized construction data groups in the generalized construction data set are sorted in ascending order to obtain the generalized construction data group sequence. Each generalized construction data group has a lineage pointer identifier. Traverse the sequence of filtered generalized construction data groups and determine the first filtered generalized construction data group that meets the preset filtering conditions as the target generalized construction data group.

7. A construction equipment control device based on process-level generalization and aggregation, characterized in that: The acquisition unit is configured to acquire historical construction datasets; The aggregation unit is configured to perform process homogeneous sample aggregation processing on the historical construction dataset to obtain various historical construction data groups. The generalization aggregation processing unit is configured to perform final-level process generalization aggregation processing on each historical construction data group based on the historical construction dataset, so as to obtain a generalized construction data group set corresponding to the historical construction data group. The filtering unit is configured to, in response to receiving a construction equipment control request sent by the user terminal containing the current final-level process identifier, construction area identifier, and current construction parameter information, filter the target historical construction data group corresponding to the current final-level process identifier from the various historical construction data groups; The lineage tracing and matching processing unit is configured to, in response to determining that the number of historical construction data in the target historical construction data group is less than or equal to a preset lower threshold, perform lineage tracing and matching processing on the target historical construction data group based on the obtained generalized construction data group sets, to obtain the target generalized construction data group. The training unit is configured to train the initial construction control parameter prediction model based on the target generalized construction data set to obtain the construction control parameter prediction model. The driving unit is configured to input the current construction parameter information into the construction control parameter prediction model to drive the construction equipment corresponding to the construction area identifier to perform construction operations.

8. An electronic device, characterized in that: One or more processors; A storage device on which one or more programs are stored; When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1 to 6.

9. A computer readable medium having stored thereon a computer program, characterized in that: When the program is executed by the processor, it implements the method as described in any one of claims 1 to 6.

Images