Intelligent Analysis Method and System for Energy Consumption in Building Construction

Abstract

This invention discloses an intelligent analysis method and system for energy consumption in building construction, belonging to the field of data processing technology. The method includes: acquiring energy consumption monitoring datasets for building construction nodes; performing consumption density analysis on each building construction node and selecting identified building construction nodes with a consumption density greater than a preset value; constructing a tree-shaped construction energy consumption decomposition model; performing multi-dimensional factor matrix decomposition on the identified building construction nodes to obtain an energy consumption distribution output matrix; and optimizing the execution strategy for the identified building construction nodes to obtain a low-consumption execution strategy. This invention solves the technical problem that existing building construction energy consumption management methods are unable to accurately analyze increasingly complex construction scenarios, resulting in low energy management accuracy and energy utilization. It achieves the technical effect of accurately identifying key consumption nodes and optimizing execution strategies, thereby improving the accuracy and efficiency of energy management in complex construction processes.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, specifically to a method and system for intelligent analysis of energy consumption in building construction. Background Technology

[0002] Energy consumption is a significant cost factor in building construction. Traditional energy management methods often rely on manual monitoring and experience-based adjustments, which are ill-suited to increasingly complex construction scenarios. With the expansion of building scale and the diversification of construction techniques, traditional energy management methods cannot achieve precise analysis and optimization, resulting in low energy management accuracy and poor energy utilization. This is especially true in complex construction processes involving multiple construction nodes and various equipment, materials, and resources, where effective energy consumption monitoring and real-time optimization strategies are lacking. Furthermore, existing methods often lack systematicity and operability in formulating optimization strategies, making them difficult to directly apply to actual construction processes. Summary of the Invention

[0003] This application provides a method and system for intelligent analysis of energy consumption in building construction, which addresses the technical problem that existing building construction energy management methods are unable to accurately analyze increasingly complex construction scenarios, resulting in low energy management accuracy and energy utilization.

[0004] The first aspect of this application provides an intelligent analysis method for energy consumption in building construction. The method includes: acquiring an energy consumption monitoring dataset of building construction nodes; performing energy consumption density analysis on each building construction node according to the energy consumption monitoring dataset to obtain a energy consumption density evaluation index; selecting identified building construction nodes with a density greater than a preset energy consumption density based on the energy consumption density evaluation index, wherein the energy consumption density analysis includes time density, efficiency density, spatial density, and value density; constructing a tree-shaped construction energy consumption decomposition model; using the tree-shaped construction energy consumption decomposition model to perform multi-dimensional factor matrix decomposition on the identified building construction nodes to obtain an energy consumption distribution output matrix, wherein the multi-dimensional factor matrix includes equipment factors, process factors, environmental factors, and resource factors; and optimizing the execution strategy of the identified building construction nodes based on the energy consumption distribution output matrix to obtain a low-energy consumption execution strategy.

[0005] A second aspect of this application provides an intelligent analysis system for energy consumption in building construction. The system includes: a monitoring data acquisition module for acquiring energy consumption monitoring datasets of building construction nodes; a consumption density analysis module for performing consumption density analysis on each building construction node according to the energy consumption monitoring dataset, obtaining a consumption density evaluation index, and selecting identified building construction nodes with a consumption density greater than a preset value based on the consumption density evaluation index, wherein the consumption density analysis includes time density, efficiency density, spatial density, and value density; an energy consumption distribution analysis module for constructing a tree-shaped construction energy consumption decomposition model, using the tree-shaped construction energy consumption decomposition model to perform multi-dimensional factor matrix decomposition on the identified building construction nodes, obtaining an energy consumption distribution output matrix, wherein the multi-dimensional factor matrix includes equipment factors, process factors, environmental factors, and resource factors; and an execution strategy optimization module for optimizing the execution strategy of the identified building construction nodes according to the energy consumption distribution output matrix to obtain a low-consumption execution strategy.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The intelligent analysis method and system for building construction energy consumption provided in this application pertains to the field of data processing technology. Through consumption density analysis and a tree-shaped energy consumption decomposition model, it accurately identifies key energy consumption factors in building construction nodes and obtains energy consumption distribution based on multi-dimensional factor matrix decomposition. By analyzing consumption density evaluation indicators, it optimizes the execution strategies of construction nodes. This solves the technical problem that existing building construction energy consumption management methods are unable to accurately analyze increasingly complex construction scenarios, resulting in low energy management accuracy and energy utilization. It achieves the technical effect of accurately identifying key consumption nodes and optimizing execution strategies, thereby improving the accuracy and energy utilization of energy management in complex construction processes. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic diagram of the intelligent analysis method for building construction energy consumption provided in the embodiments of this application;

[0010] Figure 2 This is a schematic diagram of the intelligent analysis system for building construction energy consumption provided in an embodiment of this application.

[0011] Figure labeling: 11 Monitoring data acquisition module, 12 Consumption density analysis module, 13 Energy consumption distribution analysis module, 14 Execution strategy optimization module. Detailed Implementation

[0012] This application provides a method and system for intelligent analysis of energy consumption in building construction, which addresses the technical problem that existing building construction energy management methods are unable to accurately analyze increasingly complex construction scenarios, resulting in low energy management accuracy and energy utilization.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0015] Example 1, as Figure 1 As shown, this application provides an intelligent analysis method for energy consumption in building construction, the method comprising:

[0016] P10: Obtain the energy consumption monitoring dataset for building construction nodes.

[0017] Specifically, the first step is to deploy appropriate energy monitoring equipment to acquire energy consumption monitoring datasets for building construction nodes. These construction nodes refer to specific construction stages or locations with particular functions and energy consumption characteristics during the building construction process, such as concrete pouring nodes, steel structure installation nodes, and electrical system wiring nodes. These nodes consume different types of energy during construction, such as electricity, fuel oil, and heat. To comprehensively understand the energy consumption of each node, high-precision energy monitoring equipment, such as smart meters, flow sensors, and heat sensors, needs to be installed at key locations on the construction site. These devices can collect energy consumption data in real time and transmit it to a data processing center.

[0018] Next, various data related to energy consumption at each construction node are collected using the aforementioned monitoring equipment to generate an energy consumption monitoring dataset. This dataset includes, but is not limited to, the following: energy consumption at each construction node over different time periods (e.g., hourly and daily electricity and fuel consumption), equipment operating status data (e.g., equipment start-up and shutdown times, operating power), and construction environment parameters (e.g., temperature, humidity, and light intensity at the construction site, as these environmental factors may affect energy consumption). This data is collected and stored through a data acquisition module, forming a structured energy consumption monitoring dataset that provides the foundation for subsequent analysis and processing.

[0019] Furthermore, to ensure data standardization and comparability, all collected data must be processed in a unified format. For example, for power consumption monitoring data, it may be necessary to convert the power consumption of different devices and standardize the units to ensure that data from all nodes can be compared and analyzed using a unified standard throughout the entire construction period. The data format standardization process includes standardizing the timestamps, units, and numerical ranges of the data to ensure seamless data exchange between different devices and nodes on the same platform.

[0020] Meanwhile, considering the potential for different energy consumption patterns during construction, the data acquisition module should possess good scalability, enabling it to dynamically add or adjust monitoring nodes based on the progress of the construction project or changes in different construction stages, ensuring the integrity of the monitoring data. Furthermore, it should have certain data preprocessing capabilities, such as data cleaning and outlier removal, to avoid invalid data interfering with subsequent analysis.

[0021] P20: Perform energy consumption density analysis on each building construction node according to the energy consumption monitoring dataset to obtain the energy consumption density evaluation index. Based on the energy consumption density evaluation index, select the building construction nodes that have a higher than the preset energy consumption density. The energy consumption density analysis includes time density, efficiency density, spatial density and value density.

[0022] Wherein, the time density is the ratio of the total energy consumption of a building construction node to the continuous construction time of the corresponding building construction node, the efficiency density is the ratio of the total energy consumption of a building construction node to the BIM physical engineering quantity of the corresponding building construction node, the spatial density is the ratio of the total energy consumption of a building construction node to the construction area occupied by the corresponding building construction node, and the value density is the ratio of the total energy consumption of a building construction node to the construction cost of the corresponding building construction node.

[0023] Optionally, based on the acquired energy consumption monitoring dataset, a detailed energy consumption density analysis is conducted on each construction node. Through multi-dimensional evaluation indicators, the energy consumption density of each node during construction is quantified, and nodes exceeding preset standards are identified to further optimize resource allocation and construction strategies. The energy consumption density analysis covers four key dimensions: time density, efficiency density, spatial density, and value density. Each dimension quantifies and evaluates the energy consumption of each construction node from different perspectives, thereby comprehensively and accurately grasping the energy utilization status of each node.

[0024] First, time density is calculated by dividing the total energy consumption of a construction node by the duration of continuous construction at that node. This indicator clearly reflects the energy consumption intensity of the construction node per unit time. For example, if a construction node consumes a large amount of energy in a short period of time, its time density value will be significantly higher, which may indicate that the node has insufficient energy utilization efficiency and requires further analysis and optimization.

[0025] Efficiency density is derived by comparing the total energy consumption of a construction node with its BIM physical quantities. BIM physical quantities are the actual workload of a construction node calculated based on the Building Information Model (BIM), such as the volume of concrete poured or the weight of steel structures. Calculating efficiency density allows for the assessment of energy consumption per unit of work. A high efficiency density for a node indicates that it consumes more energy to complete the same amount of work, potentially indicating energy waste that requires attention and improvement. Therefore, this indicator helps identify construction stages where energy consumption does not match the actual workload.

[0026] The calculation of spatial density involves the ratio of the total energy consumption of a building construction node to the construction area occupied by that node. This indicator is used to measure energy consumption per unit construction area. For example, if a construction node consumes a large amount of energy within a relatively small construction area, its spatial density will be high. This may indicate that the construction layout of that node is not reasonable enough, or that there is an inefficiency in energy utilization, requiring optimization and adjustment of the construction layout or energy utilization methods.

[0027] Finally, value density is measured by the ratio of the total energy consumption of a construction node to the construction cost of that node. Construction cost refers to the combined expense of all resources used in the construction process, including labor, materials, and equipment. A construction node with high value density means that its energy consumption is higher relative to the cost invested, and may require cost control and energy efficiency optimization.

[0028] After completing the density calculations across the four dimensions mentioned above, we obtain the energy consumption density assessment indicators for each construction node, namely, the time density, efficiency density, spatial density, and value density values ​​for each node. These indicators are presented numerically, intuitively reflecting the energy consumption characteristics of each node. Subsequently, based on these energy consumption density assessment indicators, construction nodes exceeding a preset energy consumption density threshold are selected as identified construction nodes. The preset energy consumption density threshold is a reference value set according to industry standards, historical data, or project objectives, used to determine which nodes' energy consumption exceeds the normal range. Through this screening process, nodes with abnormally high energy consumption during construction can be accurately identified. These identified nodes represent construction phases with abnormal energy consumption in one or more dimensions and will become the focus of subsequent optimization strategies.

[0029] P30: Construct a tree-shaped construction energy consumption decomposition model. Using this model, decompose the identified building construction nodes into a multi-dimensional factor matrix to obtain an energy consumption distribution output matrix. This multi-dimensional factor matrix includes equipment factors, process factors, environmental factors, and resource factors. The equipment factors include equipment performance factors, equipment matching factors, and equipment status factors. The process factors include technical process factors, construction path factors, and process connection factors. The environmental factors include spatial location factors, temporal distribution factors, and climate state factors. The resource factors include information resource factors, collaborative resource factors, and resource supply factors.

[0030] It should be understood that constructing a tree-structured construction energy consumption decomposition model involves a detailed breakdown of energy consumption at each stage of the construction process, identifying various potential factors affecting energy consumption. This construction energy consumption decomposition model treats the energy consumption of each construction stage as the result of the combined effects of multiple factors, which encompass four key dimensions: equipment, technology, environment, and resources.

[0031] Specifically, the construction of the tree-shaped construction energy consumption decomposition model involves breaking down the complex construction process into hierarchical structures, decomposing the overall energy consumption into the energy consumption of multiple subsystems and sub-nodes. This tree structure clearly demonstrates the energy flow relationships between different levels, progressively refining the sources of energy consumption from the macroscopic construction stages to the microscopic equipment operations. For example, the entire building construction process can be decomposed into the foundation construction stage, the main structure construction stage, and the decoration construction stage. Each stage can be further decomposed into specific construction nodes, such as the foundation construction stage, which can be decomposed into nodes such as foundation pit excavation and foundation pouring. Each node is then associated with specific equipment and processes.

[0032] After constructing a tree-structured construction energy consumption decomposition model, a multi-dimensional factor matrix decomposition is performed on the identified construction nodes. This involves decomposing the energy consumption data into multiple factor matrices, each representing a specific dimension, thereby revealing the key factors influencing energy consumption. These factors include equipment factors, process factors, environmental factors, and resource factors. To achieve this decomposition, methods such as nonnegative matrix factorization (NMF), tensor decomposition (e.g., CP / PARAFAC), principal component analysis (PCA), or factor decomposition based on sparse coding can be employed. Each method has its advantages. For example, NMF ensures that the elements of the decomposed factor matrix are nonnegative, consistent with the actual physical meaning of energy consumption; tensor decomposition can handle multi-dimensional data and is suitable for complex construction scenarios; PCA can extract the main directions of variation, simplifying the data structure; and factor decomposition based on sparse coding can yield a more concise and interpretable factor representation.

[0033] The decomposed factor matrix comprises multiple dimensions. The equipment factor considers various factors related to equipment used in building construction, including equipment performance factors, equipment matching factors, and equipment status factors. Equipment performance factors reflect the energy efficiency level of the equipment itself, such as its rated power and operating efficiency. Equipment matching factors consider the collaborative working relationship between equipment, such as power matching and operating speed matching between different devices. Equipment status factors relate to the real-time operating status of the equipment, such as its wear and tear and failure frequency. By decomposing these factors, the specific role of equipment in energy consumption can be clarified, providing a basis for equipment selection, maintenance, and optimization.

[0034] Process factors focus on the impact of technology and operational methods on energy consumption during construction, including technological process factors, construction path factors, and process connection factors. Technological process factors reflect the efficiency of the technical methods and processes used in construction; different technologies and processes have significantly different energy requirements. Construction path factors involve the selection of work paths during construction; unreasonable paths can lead to energy waste, such as unnecessary transportation or equipment movement. Process connection factors focus on the efficiency of connections between different processes, such as waiting time between processes and the timeliness of material supply. Analyzing these factors helps optimize construction processes and reduce energy waste caused by unreasonable processes.

[0035] The decomposition results of environmental factors include spatial location factors, temporal distribution factors, and climate state factors. Spatial location factors analyze the location of construction nodes within the construction site. Some construction nodes, if located in remote areas, may require additional energy support, such as long-distance transportation or remote equipment operation, leading to higher energy consumption. Temporal distribution factors analyze the timing of construction nodes, particularly whether they occur during peak hours or overtime work. Improper time allocation can concentrate energy consumption and increase construction costs. Climate state factors focus on climate changes during construction. Extreme weather (high temperatures, heavy rain, etc.) may lead to additional energy demands, such as the operation of air conditioning and dehumidification equipment, or construction delays. Analyzing these factors can help rationally schedule construction time and location, reducing the adverse impact of environmental factors on energy consumption.

[0036] Resource factors encompass various resources required during construction, including information resources, collaborative resources, and resource supply factors. Information resources relate to information management at the construction site, such as the timely transmission of construction progress information and real-time monitoring of energy consumption data. Collaborative resources involve the efficiency of collaboration among construction parties, such as coordination between different construction teams and cooperation between equipment leasing companies and construction companies. Resource supply factors focus on the stability of energy and material supply, such as the impact of energy supply interruptions and material supply delays on construction. Analyzing these factors helps optimize resource management at the construction site, improve resource utilization efficiency, and reduce energy consumption increases caused by resource issues.

[0037] The energy consumption distribution output matrix obtained through the multi-dimensional factor matrix decomposition clearly demonstrates the contribution of each factor to energy consumption. This not only helps to deeply understand the complexity of energy consumption during building construction but also provides a strong basis for subsequent optimization strategies. By analyzing the impact of different factors, high-consumption construction nodes can be accurately identified, and practical suggestions can be provided for implementing energy-saving and consumption-reducing measures, thereby improving overall construction efficiency and reducing energy consumption.

[0038] Furthermore, in constructing a tree-shaped construction energy consumption decomposition model, step P30 of this application embodiment also includes:

[0039] P31: Define a fourth-order nonnegative tensor based on the equipment factor, process factor, environmental factor, and resource factor; P32: Preconstruct a tree-like node structure, which includes at least first-level, second-level, and third-level tree nodes. The second-level tree nodes are the child node factors corresponding to the first-level tree nodes, and the third-level tree nodes are the factor value ranges of the second-level tree nodes; P33: Obtain the preprocessed historical energy consumption sample dataset and input it into the fourth-order nonnegative tensor to generate model input samples; P34: Select an initial decomposition rank, and train the tree-like node structure using nonnegative tensor decomposition based on the selected initial decomposition rank and the model input samples to obtain a tree-like construction energy consumption decomposition model.

[0040] Optionally, the construction process of the tree-shaped construction energy consumption decomposition model can be further refined.

[0041] First, a fourth-order nonnegative tensor is defined based on equipment, process, environment, and resource factors. Each dimension of this fourth-order nonnegative tensor represents a factor, and the tensor values ​​are all nonnegative, which aligns with the actual situation of energy consumption, as energy consumption is always positive. By transforming the four factors of equipment, process, environment, and resources into the dimensions of a tensor, their interactions and their impact on energy consumption at construction nodes can be clearly represented. A tensor is a multidimensional array that can effectively represent the complex relationships between multidimensional data. A fourth-order nonnegative tensor means that it has four dimensions, each corresponding to a factor category: equipment, process, environment, and resource. Equipment factors cover aspects such as equipment performance, matching, and status; process factors involve elements such as technology, construction paths, and process connections; environmental factors include conditions such as spatial location, temporal distribution, and climate conditions; and resource factors relate to information resources, collaborative resources, and resource supply.

[0042] Next, a pre-constructed tree-like node structure is built, which includes at least first-level, second-level, and third-level tree nodes. First-level tree nodes represent major energy consumption categories, such as equipment, process, environment, and resources; second-level tree nodes are the child node factors corresponding to the first-level tree nodes, further refining the specific factors for each major category; third-level tree nodes are the factor value ranges of the second-level tree nodes, used to further subdivide the specific numerical range of each factor. For example, second-level nodes under the equipment factor can include equipment performance, equipment matching, and equipment status, while the numerical range of the equipment performance factor can be divided into high, medium, and low levels. This structure will help to deeply analyze the energy consumption of each node and its underlying causes, more accurately identifying which factors play a key role in energy consumption through hierarchical relationships.

[0043] Subsequently, a preprocessed historical energy consumption sample dataset is acquired and input into a fourth-order nonnegative tensor to generate model input samples. The historical energy consumption sample dataset includes energy consumption data from multiple construction projects, as well as working conditions and environmental factors related to construction milestones. The preprocessing process includes removing noisy data, standardizing data format, and imputing missing values ​​to ensure data accuracy and consistency. The processed data is then used as input samples in the fourth-order nonnegative tensor to generate model input samples. In this way, the data can fully reflect the changes of various factors at different construction milestones, providing effective training data for subsequent tensor decomposition.

[0044] Finally, an initial decomposition rank is selected, and the tree node structure is trained using non-negative tensor decomposition based on the selected initial decomposition rank and the model input samples. The choice of the initial decomposition rank is crucial to the model's training effect, determining the complexity of the decomposed factor matrix and the model's fitting ability. Through non-negative tensor decomposition training, complex energy consumption data can be decomposed into multiple non-negative factor matrices, thereby revealing the hidden structure in the data. Specifically, the non-negative tensor decomposition algorithm is used to train the model input samples and the tree node structure. Through an iterative optimization process, the tensor decomposition results are continuously adjusted so that the decomposed factor matrix can reflect the actual energy consumption situation as accurately as possible. In this way, the final tree-shaped construction energy consumption decomposition model can reveal the specific contribution of each factor (equipment, process, environment, resources) to energy consumption and provide data support for subsequent energy optimization and construction strategy improvement.

[0045] Furthermore, in constructing a tree-shaped construction energy consumption decomposition model, step P34 of this application embodiment also includes:

[0046] P34-1: Define the optimization objective for model training using tree regularization, time smoothing regularization, and sparsity regularization; wherein, the tree regularization term includes a constraint condition, the constraint condition being that the parent node factor is equal to the weighted sum of the child factors of the parent node factor; P34-2: Use the alternating optimization algorithm of non-negative least squares to optimize the weights of each node in the tree node structure based on the optimization objective, to obtain a convergent tree-shaped construction energy consumption decomposition model.

[0047] Specifically, the process of constructing a tree-shaped construction energy consumption decomposition model can be further refined to ensure that the model training process can accurately reflect the energy consumption patterns of construction nodes and effectively improve the model's predictive ability and practicality.

[0048] Specifically, by introducing tree regularization, temporal smoothing regularization, and sparsity regularization, the optimization objective for model training is defined. The introduction of these regularization terms ensures the stability and interpretability of the model, while satisfying the hierarchical relationship of the tree structure and the smoothness of the time series.

[0049] First, the tree regularization term includes a key constraint: the parent node factor equals the weighted sum of its child factors. This constraint ensures that the hierarchical relationships between nodes in the tree structure are maintained, allowing the model to accurately reflect the dependencies between factors at different levels. For example, in a tree structure, the value of a first-level node (such as the equipment factor) should be equal to the weighted sum of all its child nodes (such as the equipment performance factor, equipment matching factor, and equipment status factor). This hierarchical constraint not only contributes to the interpretability of the model but also ensures the reasonableness of the decomposition results.

[0050] Secondly, the introduction of a time smoothing regularization term ensures the smoothness of the model over time. During construction, energy consumption often exhibits temporal continuity and regularity. By constraining factor changes between adjacent time points, the time smoothing regularization term enables the model to better capture the dynamic characteristics of energy consumption over time, thereby improving the model's ability to fit time-series data.

[0051] Finally, the introduction of a sparse regularization term aims to improve the sparsity of the model, making the decomposition results more concise and interpretable. Sparsity means that the model tends to select fewer factors to explain the data, thereby reducing model complexity and the risk of overfitting. The sparse regularization term ensures that the model maintains interpretability while also possessing good generalization ability.

[0052] After defining the optimization objective, an alternating optimization algorithm using non-negative least squares can be employed to optimize the weights of each node in the tree-like structure based on the aforementioned objective. Non-negative least squares is a commonly used optimization method, particularly suitable for handling optimization problems with non-negative constraints. The alternating optimization algorithm iteratively optimizes each factor matrix, gradually approaching the global optimum. In each iteration, the algorithm fixes other factor matrices and optimizes the current factor matrix until the model converges, resulting in a tree-like construction energy consumption decomposition model. This model accurately reflects the energy consumption characteristics of each node during construction and provides strong support for subsequent energy consumption optimization.

[0053] Furthermore, the energy consumption decomposition model of the tree-structured construction is used to perform multi-dimensional factor matrix decomposition on the identified building construction nodes to obtain the energy consumption distribution output matrix. Step P30 of this embodiment also includes:

[0054] P35: Obtain the energy consumption monitoring dataset corresponding to the marked building construction node; P36: Construct the fourth-order non-negative monitoring tensor of the marked energy consumption monitoring dataset, use the tree-shaped construction energy consumption decomposition model to solve the energy consumption contribution value of the fourth-order non-negative monitoring tensor, and parse the energy consumption contribution value to obtain the energy consumption distribution output matrix.

[0055] Optionally, the process of using a tree-structured construction energy consumption decomposition model to perform multi-dimensional factor matrix decomposition on the construction nodes of the marked buildings can be further refined. After obtaining a converged tree-structured construction energy consumption decomposition model, this model can be used to conduct in-depth analysis of the construction nodes of the marked buildings to obtain detailed energy consumption distribution.

[0056] Specifically, the first step is to obtain the energy consumption monitoring dataset corresponding to the construction nodes of the marker buildings. These datasets are collected in real time by high-precision monitoring equipment installed during the construction process, covering the energy consumption data of each marker node at different time points.

[0057] Subsequently, based on these energy consumption monitoring datasets, a fourth-order nonnegative monitoring tensor was constructed. The dimensions of this tensor correspond to the four previously defined factors (equipment, process, environment, and resources), with the size of each dimension representing the diversity and variability of different factors in energy consumption. This tensor structure allows for the systematization and mathematical representation of energy consumption information at key construction nodes, ensuring that subsequent energy contribution analysis comprehensively and accurately considers every factor that may affect energy consumption. Specifically, the equipment factor dimension covers equipment performance, matching, and status; the process factor dimension involves technical processes, construction paths, and process connections; the environmental factor dimension includes spatial location, temporal distribution, and climate conditions; and the resource factor dimension relates to information resources, collaborative resources, and resource supply. This approach provides a more comprehensive reflection of the intrinsic structure and interrelationships of energy consumption.

[0058] Next, using the previously constructed tree-structured construction energy consumption decomposition model, the fourth-order non-negative monitoring tensor is solved to obtain the energy consumption contribution value of each factor. By analyzing and decomposing the monitoring tensor, the model can identify the specific contribution of each factor (equipment, process, environment, resources) to the energy consumption of construction nodes. Specifically, the tree-structured model will calculate the energy consumption contribution value of each construction node based on the known factor matrix, reflecting the weight and role of different factors in energy consumption.

[0059] Finally, by analyzing these energy consumption contribution values, the multi-dimensional factor matrix is ​​refined, and an energy consumption distribution output matrix is ​​constructed. This matrix comprehensively displays the energy consumption distribution at different construction nodes and reveals the energy consumption contribution of each factor at each node. This output matrix not only helps analyze the distribution of energy consumption but also provides specific evidence for further optimizing energy management during construction. For example, if an abnormally high energy consumption contribution value is found for a certain piece of equipment within a specific time period, the equipment can be specifically inspected and optimized; if the energy consumption contribution value of a certain construction process is generally high, improvements to the process flow can be considered to reduce energy consumption.

[0060] Furthermore, after obtaining the energy consumption distribution output matrix, step P30 in this embodiment of the application further includes:

[0061] P31a: Extract the energy consumption distribution health matrix of the building construction node; P32a: Compare the energy consumption distribution health matrix with the energy consumption distribution output matrix to generate fault reminder information.

[0062] In one possible embodiment of this application, after obtaining the energy consumption distribution output matrix, the energy consumption health status of the building construction nodes can be further assessed to ensure the health of energy consumption management and to promptly detect potential faults or anomalies.

[0063] Specifically, the first step is to extract the energy consumption distribution health matrix for each construction node. This energy consumption distribution health matrix is ​​a baseline matrix constructed by analyzing historical data, industry standards, and construction specifications. It reflects the energy consumption distribution of each construction node under normal construction conditions. For example, for a specific construction node, the distribution ratios of equipment energy consumption, process energy consumption, environmental energy consumption, and resource energy consumption under normal conditions are recorded in the energy consumption distribution health matrix.

[0064] Subsequently, the energy consumption distribution health matrix is ​​compared with the energy consumption distribution output matrix. The purpose of this comparison is to identify the deviation between the actual energy consumption distribution and the health status. The comparison process can be achieved by calculating a difference metric between the two matrices, such as using Euclidean distance, cosine similarity, or other statistical measures. If the difference metric exceeds a preset threshold, it indicates a significant deviation between the actual energy consumption distribution and the health status, which may mean problems such as energy waste, equipment failure, or unreasonable construction techniques during construction. For example, if the actual equipment energy consumption ratio at a certain construction node is much higher than the corresponding value in the health matrix, it may mean that the equipment is faulty or operating inefficiently.

[0065] Based on the comparison results, fault alerts are generated. These alerts should include a detailed description of the deviation, possible cause analysis, and suggested inspection or maintenance measures. For example, the alert might indicate abnormally high energy consumption at a specific construction node and suggest checking the equipment's operating status, including maintenance records and operating parameter settings. Furthermore, temporary energy-saving measures can be provided to reduce energy waste until the problem is completely resolved. These alerts will be sent to construction management personnel or maintenance teams so they can take timely action to correct abnormal energy consumption and ensure energy efficiency and construction quality throughout the process.

[0066] P40: Optimize the execution strategy of the identified building construction node based on the energy consumption distribution output matrix to obtain a low-consumption execution strategy.

[0067] Furthermore, step P40 in this embodiment of the application also includes:

[0068] P41: Locate the key consumption factors of the identified building construction node based on the energy consumption distribution output matrix; P42: Extract the task execution strategy of the key consumption factors, and perform directed optimization on the task execution strategy with the energy consumption contribution value of the key consumption factors being less than the average energy consumption contribution value as the optimization objective, to obtain the task execution optimization strategy of the key consumption factors; P43: Output the task execution optimization strategy as the low-consumption execution strategy of the identified building construction node.

[0069] It should be understood that the execution strategy for identifying building construction nodes is optimized based on the energy consumption distribution output matrix in order to obtain a low-energy-consumption execution strategy and ultimately achieve more efficient energy use.

[0070] First, based on the energy consumption distribution output matrix, a detailed analysis is conducted on the identified construction nodes to pinpoint the key consumption factors for these nodes. These key consumption factors are those that significantly impact energy consumption during construction. Analysis of the energy consumption distribution output matrix identifies which factors (such as equipment usage, process flow, environmental impact, and resource allocation) contribute the most to overall energy consumption at each construction node. For example, if equipment usage or certain steps in the work process at a particular node result in high energy consumption, these devices or processes are considered key consumption factors.

[0071] After identifying key energy consumption factors, task execution strategies for these factors are extracted. These strategies are then optimized in a directed manner, with the energy consumption contribution value being less than the average energy consumption contribution value. The task execution strategies specifically include the execution time windows for each process or task in each construction node, resource allocation schemes (such as equipment and personnel scheduling and configuration), equipment power settings and parallelism recommendations, and priority ranking of each process. The optimization objective is to reduce the contribution of these key factors to overall energy consumption. Specifically, optimization can be achieved through the following aspects: optimizing time windows by adjusting the start and end times of each process to reduce equipment idle time and unnecessary running time. For example, scheduling the running time of high-energy-consuming equipment during periods of lower energy prices; rationally allocating equipment and personnel resources to ensure that equipment operates in an efficient state and avoids equipment overload or idleness. For example, rationally arranging the equipment usage sequence and personnel configuration based on the equipment's energy consumption characteristics and construction needs; adjusting equipment power settings according to actual equipment needs to avoid equipment operating at high power. For example, for some adjustable power equipment, adjusting the power according to the actual load to reduce energy consumption; and optimizing the parallel operation strategy of equipment to reduce redundant operation between equipment. For example, by rationally arranging the parallel operation of equipment, the utilization rate of equipment can be improved and unnecessary energy consumption can be reduced; construction tasks can be prioritized according to their energy consumption contribution value, and high-energy-consuming tasks can be handled first. For example, the maintenance and optimization of high-energy-consuming equipment can be prioritized to reduce their impact on energy consumption.

[0072] These optimization measures effectively reduce the energy contribution of key consumption factors, bringing them below the average energy contribution, thereby optimizing energy consumption. After completing the above optimizations, the optimized task execution strategies are output as low-consumption execution strategies to identify construction nodes. These optimized execution strategies will directly guide operations during construction, ensuring full utilization of resources and minimizing energy consumption. The output low-consumption execution strategies will specifically guide the operations of each construction node during task execution, including equipment use, work sequence arrangement, and personnel scheduling, thereby improving construction efficiency while reducing unnecessary energy waste.

[0073] Furthermore, after obtaining the task execution optimization strategy for the key consumption factors, step P43 in this embodiment of the application further includes:

[0074] P43-1a: Obtain the construction process including the identified building construction node; P43-2a: Identify the remaining building construction nodes of the construction process, and the task execution strategy set corresponding to the remaining building construction nodes; P43-3a: Perform adaptive collaborative processing on the task execution strategy set according to the task execution optimization strategy of the identified building construction node to obtain a low-consumption execution strategy based on the construction process.

[0075] Specifically, after obtaining the task execution optimization strategy for key consumption factors, the entire construction process is further optimized through collaborative optimization.

[0076] Specifically, the first step is to obtain the construction procedures that identify building construction nodes. A construction procedure refers to the sequence of operations in the construction process, including different task phases such as design, material preparation, equipment installation, and construction. Each construction node involves one or more procedures. By clarifying the construction procedures for each node, it is possible to understand the nature of the task, the resources required, and the energy consumption pattern of each stage.

[0077] Subsequently, based on the identified construction procedures, it is necessary to further identify the remaining construction nodes and the corresponding task execution strategy sets for these nodes. Remaining construction nodes refer to all construction nodes in the current construction procedure other than the identified nodes. These nodes also have their own task execution strategies, including time scheduling, resource allocation, and equipment operating parameters. By identifying these nodes and their strategy sets, a comprehensive understanding of the execution status of the entire construction procedure can be obtained.

[0078] Next, based on the task execution optimization strategies for the identified construction nodes, adaptive collaborative processing is applied to the task execution strategy set for the remaining nodes. The purpose of adaptive collaborative processing is to ensure that the optimization strategies not only apply to individual nodes but also effectively coordinate and optimize the execution strategies of multiple related nodes. This means that each node in the construction process may be affected by preceding and following stages; therefore, when adjusting the task execution strategies for the remaining nodes, the interrelationships and dependencies between nodes must be considered. For example, some processes may need to be completed within a specific time window, while delays or advancements in other nodes may affect overall energy efficiency. Specific operations may include:

[0079] Time Coordination: Adjust the start and end times of each construction node to ensure that the optimization strategy for key consumption factors does not negatively impact the construction progress of other nodes. For example, if the optimization strategy suggests reducing equipment uptime within a specific time period, the construction times of other nodes need to be adjusted accordingly to avoid construction delays.

[0080] Resource coordination: Reallocate equipment and personnel resources to ensure that optimization strategies for key consumption factors can be implemented without affecting the construction efficiency of other nodes. For example, if the optimization strategy suggests reducing the use of a certain piece of equipment, the resources of that equipment need to be reallocated to other nodes to ensure the smooth progress of the construction process.

[0081] Equipment coordination: Adjust equipment operating parameters and parallelism to ensure that optimization strategies for key consumption factors are coordinated with equipment operation strategies throughout the construction process. For example, if the optimization strategy suggests reducing the power of a certain piece of equipment, the power settings of other equipment need to be adjusted accordingly to maintain energy efficiency during construction.

[0082] Prioritization and Coordination: The priorities of construction tasks are reordered to ensure that optimization strategies for key energy-consuming factors are implemented preferentially without affecting the construction quality of other nodes. For example, if the optimization strategy suggests prioritizing a high-energy-consuming task, the priorities of other tasks need to be adjusted accordingly to ensure the smooth progress of the construction process.

[0083] Ultimately, through these adaptive and collaborative processes, the resulting low-energy-consumption execution strategies based on construction procedures comprehensively optimize energy use during construction. These strategies not only improve construction efficiency but also reduce unnecessary energy consumption, ensuring optimal energy management throughout the project. In this way, energy consumption at each stage and step of the construction process can be kept within optimal control, thereby achieving energy conservation and high efficiency in the building construction process.

[0084] In summary, the embodiments of this application have at least the following technical effects:

[0085] This application uses multi-dimensional factor matrix decomposition and tree model analysis to accurately locate key nodes with high energy consumption during building construction, providing clear targets for subsequent optimization. Based on a detailed energy consumption distribution output matrix, it generates targeted low-consumption execution strategies to effectively reduce energy consumption. By comprehensively considering multiple dimensions such as equipment, processes, environment, and resources, it achieves comprehensive analysis and refined management of energy consumption during building construction, improving the scientific nature and accuracy of energy management. Through the implementation of optimization strategies, it reduces energy waste, improves energy utilization efficiency, lowers construction costs, and reduces environmental impact, meeting the requirements of sustainable development. By comparing the healthy energy consumption distribution matrix with the actual output matrix, it promptly identifies abnormal energy consumption and generates early warning information, helping construction management personnel to respond quickly and ensuring efficient energy utilization during the construction process.

[0086] It achieves the technical effect of accurately identifying key consumption nodes and optimizing execution strategies, thereby improving the accuracy of energy management and energy utilization in complex construction processes.

[0087] Example 2, based on the same inventive concept as the intelligent analysis method for building construction energy consumption in the foregoing examples, such as... Figure 2 As shown, this application provides an intelligent analysis system for energy consumption in building construction. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0088] The monitoring data acquisition module 11 is used to acquire the energy consumption monitoring dataset of building construction nodes.

[0089] The consumption density analysis module 12 is used to perform consumption density analysis on each building construction node according to the energy consumption monitoring dataset, obtain consumption density evaluation index, and select the building construction node with a consumption density greater than the preset consumption density based on the consumption density evaluation index. The consumption density analysis includes time density, efficiency density, spatial density and value density.

[0090] The energy consumption distribution analysis module 13 is used to construct a tree-shaped construction energy consumption decomposition model. The tree-shaped construction energy consumption decomposition model is used to decompose the marked building construction nodes into a multi-dimensional factor matrix to obtain an energy consumption distribution output matrix. The multi-dimensional factor matrix includes equipment factors, process factors, environmental factors, and resource factors.

[0091] The execution strategy optimization module 14 is used to optimize the execution strategy of the identified building construction node according to the energy consumption distribution output matrix to obtain a low-consumption execution strategy.

[0092] Furthermore, in the consumption density analysis module 12:

[0093] The time density is the ratio of the total energy consumption of a construction node to the continuous construction time of the corresponding construction node; the efficiency density is the ratio of the total energy consumption of a construction node to the BIM physical engineering quantity of the corresponding construction node; the spatial density is the ratio of the total energy consumption of a construction node to the construction area occupied by the corresponding construction node; and the value density is the ratio of the total energy consumption of a construction node to the construction cost of the corresponding construction node.

[0094] Furthermore, in the energy consumption distribution analysis module 13:

[0095] The multi-dimensional factor matrix includes equipment factors, process factors, environmental factors, and resource factors; wherein, the equipment factors include equipment performance factors, equipment matching factors, and equipment status factors; the process factors include technical process factors, construction path factors, and process connection factors; the environmental factors include spatial location factors, temporal distribution factors, and climate status factors; and the resource factors include information resource factors, collaborative resource factors, and resource supply factors.

[0096] Furthermore, the energy consumption distribution analysis module 13 is also used to perform the following steps:

[0097] A fourth-order nonnegative tensor is defined based on the equipment factor, process factor, environmental factor, and resource factor. A tree-like node structure is pre-constructed, which includes at least first-level, second-level, and third-level tree nodes. The second-level tree nodes are the child node factors corresponding to the first-level tree nodes, and the third-level tree nodes are the factor value ranges of the second-level tree nodes. A pre-processed historical energy consumption sample dataset is obtained and input into the fourth-order nonnegative tensor to generate model input samples. An initial decomposition rank is selected, and the tree-like node structure is trained by nonnegative tensor decomposition based on the selected initial decomposition rank and the model input samples to obtain a tree-like construction energy consumption decomposition model.

[0098] Furthermore, the energy consumption distribution analysis module 13 is also used to perform the following steps:

[0099] The optimization objective for model training is defined by tree regularization, time smoothing regularization, and sparsity regularization. The tree regularization term includes a constraint that the parent node factor is equal to the weighted sum of the child factors of the parent node factor. The alternating non-negative least squares optimization algorithm is used to optimize the weights of each node in the tree node structure based on the optimization objective, resulting in a convergent tree-shaped construction energy consumption decomposition model.

[0100] Furthermore, the energy consumption distribution analysis module 13 is also used to perform the following steps:

[0101] Obtain the energy consumption monitoring dataset corresponding to the marked building construction node; construct the fourth-order non-negative monitoring tensor of the marked energy consumption monitoring dataset; use the tree-shaped construction energy consumption decomposition model to solve the energy consumption contribution value of the fourth-order non-negative monitoring tensor; and analyze the energy consumption contribution value to obtain the energy consumption distribution output matrix.

[0102] Furthermore, the energy consumption distribution analysis module 13 is also used to perform the following steps:

[0103] Extract the energy consumption distribution health matrix of the building construction node; compare the energy consumption distribution health matrix with the energy consumption distribution output matrix to generate fault reminder information.

[0104] Furthermore, the execution strategy optimization module 14 is also used to perform the following steps:

[0105] Based on the energy consumption distribution output matrix, locate the key consumption factors of the identified building construction node; extract the task execution strategy of the key consumption factors, and perform directed optimization of the task execution strategy with the energy consumption contribution value of the key consumption factors being less than the average energy consumption contribution value as the optimization objective, to obtain the task execution optimization strategy of the key consumption factors; output the task execution optimization strategy as the low-consumption execution strategy of the identified building construction node.

[0106] Furthermore, the execution strategy optimization module 14 is also used to perform the following steps:

[0107] Obtain the construction process including the identified building construction node; identify the remaining building construction nodes of the construction process and the task execution strategy set corresponding to the remaining building construction nodes; perform adaptive collaborative processing on the task execution strategy set according to the task execution optimization strategy of the identified building construction node to obtain a low-consumption execution strategy based on the construction process.

[0108] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0109] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0110] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A method for intelligent analysis of energy consumption in construction, characterized in that, The method comprises: acquiring an energy consumption monitoring data set of a building construction node; performing consumption density analysis on each building construction node according to the energy consumption monitoring data set, obtaining a consumption density evaluation index, and selecting an identified building construction node with a consumption density greater than a preset consumption density based on the consumption density evaluation index, wherein the consumption density analysis comprises time density, efficacy density, space density, and value density; constructing a tree-shaped construction energy consumption decomposition model, performing multi-dimensional factor matrix decomposition on the identified building construction node by using the tree-shaped construction energy consumption decomposition model, and obtaining an energy consumption distribution output matrix, wherein the multi-dimensional factor matrix comprises equipment factors, process factors, environmental factors, and resource factors; optimizing an execution strategy of the identified building construction node according to the energy consumption distribution output matrix, and acquiring a low-consumption execution strategy; The method for constructing a tree-shaped construction energy consumption decomposition model comprises: defining a fourth-order non-negative tensor according to the equipment factors, the process factors, the environmental factors, and the resource factors; pre-constructing a tree-shaped node structure, wherein the tree-shaped node structure at least comprises a first-level tree-shaped node, a second-level tree-shaped node, and a third-level tree-shaped node, the second-level tree-shaped node is a child node factor corresponding to the first-level tree-shaped node, and the third-level tree-shaped node is a factor numerical interval of the second-level tree-shaped node; acquiring a preprocessed historical energy consumption sample data set, and inputting the preprocessed historical energy consumption sample data set into the fourth-order non-negative tensor to generate a model input sample; selecting an initial decomposition rank, performing non-negative tensor decomposition training on the tree-shaped node structure according to the selected initial decomposition rank and the model input sample, and obtaining a tree-shaped construction energy consumption decomposition model; performing consumption density analysis on each building construction node according to the energy consumption monitoring data set, obtaining a consumption density evaluation index, and the method further comprises: the time density is a ratio of total energy consumption of a building construction node to a corresponding continuous construction time of the building construction node, the efficacy density is a ratio of total energy consumption of a building construction node to a corresponding BIM physical engineering quantity of the building construction node, the space density is a ratio of total energy consumption of a building construction node to a corresponding occupied construction area of the building construction node, and the value density is a ratio of total energy consumption of a building construction node to a corresponding construction cost of the building construction node.

2. The method of claim 1, wherein, The multi-dimensional factor matrix comprises equipment factors, process factors, environmental factors, and resource factors; wherein the equipment factors comprise equipment performance factors, equipment matching factors, and equipment state factors, the process factors comprise technical process factors, construction path factors, and process connection factors, the environmental factors comprise space position factors, time distribution factors, and climate state factors, and the resource factors comprise information resource factors, collaboration resource factors, and resource supply factors.

3. The method of claim 1, wherein, The method for performing non-negative tensor decomposition training on the tree-shaped node structure according to the selected initial decomposition rank and the model input sample comprises: defining an optimization objective of model training through a tree-shaped regularization term, a time smoothing regularization, and a sparse regularization; wherein the tree-shaped regularization term comprises a constraint condition, and the constraint condition is that a parent node factor is equal to a weighted sum of child factors of the parent node factor. The tree-shaped construction energy consumption decomposition model is obtained by optimizing the weight of each node in the tree-shaped node structure based on an optimization target by using an alternating optimization algorithm of non-negative least squares.

4. The method of claim 1, wherein, The tree-shaped construction energy consumption decomposition model is used to perform multi-dimensional factor matrix decomposition on the identified building construction node to obtain an energy consumption distribution output matrix, and the method comprises: An identified energy consumption monitoring data set corresponding to the identified building construction node is obtained. A fourth-order non-negative monitoring tensor of the identified energy consumption monitoring data set is constructed, and an energy consumption contribution value of the fourth-order non-negative monitoring tensor is solved by using the tree-shaped construction energy consumption decomposition model, and an energy consumption distribution output matrix is obtained by analyzing the energy consumption contribution value.

5. The method of claim 1, wherein, An execution strategy optimization module is configured to optimize the execution strategy of the identified building construction node according to the energy consumption distribution output matrix to obtain a low-consumption execution strategy. The key consumption factor of the identified building construction node is located according to the energy consumption distribution output matrix. The task execution optimization strategy of the key consumption factor is obtained by performing directed optimization on the task execution strategy of the key consumption factor with the energy consumption contribution value of the key consumption factor being less than the average energy consumption contribution value as an optimization target. The task execution optimization strategy is output as the low-consumption execution strategy of the identified building construction node.

6. The method of claim 5, wherein, After obtaining the task execution optimization strategy of the key consumption factor, the method further comprises: A construction process including the identified building construction node is obtained. The remaining building construction nodes of the construction process and a task execution strategy set corresponding to the remaining building construction nodes are identified. The task execution strategy set is adaptively processed according to the task execution optimization strategy of the identified building construction node to obtain a low-consumption execution strategy based on the construction process.

7. The method of claim 1, wherein, After obtaining the energy consumption distribution output matrix, the method further comprises: An energy consumption distribution health matrix of the building construction node is extracted. The energy consumption distribution health matrix is compared with the energy consumption distribution output matrix to generate a fault reminder information.

8. An intelligent analysis system for energy consumption in building construction, characterized by, The system comprises: A monitoring data acquisition module is configured to obtain an energy consumption monitoring data set of a building construction node. A consumption density analysis module is configured to perform consumption density analysis on each building construction node according to the energy consumption monitoring data set to obtain a consumption density evaluation index, and select an identified building construction node with a consumption density greater than a preset consumption density based on the consumption density evaluation index, wherein the consumption density analysis includes time density, efficacy density, space density, and value density. An energy consumption distribution analysis module is configured to construct a tree-shaped construction energy consumption decomposition model, and perform multi-dimensional factor matrix decomposition on the identified building construction node by using the tree-shaped construction energy consumption decomposition model to obtain an energy consumption distribution output matrix, wherein the multi-dimensional factor matrix includes equipment factors, process factors, environmental factors, and resource factors. An execution strategy optimization module is configured to optimize the execution strategy of the identified building construction node according to the energy consumption distribution output matrix to obtain a low-consumption execution strategy. Further, in the consumption density analysis module, The time density is the ratio of the total energy consumption of the construction node to the corresponding construction node continuous construction time, the efficacy density is the ratio of the total energy consumption of the construction node to the corresponding construction node BIM physical engineering quantity, the space density is the ratio of the total energy consumption of the construction node to the corresponding construction node occupied construction area, and the value density is the ratio of the total energy consumption of the construction node to the corresponding construction node construction cost price; Further, the energy consumption distribution analysis module is further used to execute the following steps: Defining a four-order non-negative tensor according to the equipment factor, process factor, environment factor and resource factor; Pre-constructing a tree node structure, the tree node structure at least including a first-level tree node, a second-level tree node and a third-level tree node, the second-level tree node being a child node factor corresponding to the first-level tree node, and the third-level tree node being a factor numerical interval of the second-level tree node; Obtaining the pre-processed historical energy consumption sample data set and inputting the pre-processed historical energy consumption sample data set into the four-order non-negative tensor to generate a model input sample; Selecting an initial decomposition rank, and performing non-negative tensor decomposition training on the tree node structure according to the selected initial decomposition rank and the model input sample to obtain a tree-shaped construction energy consumption decomposition model.

Images