Energy consumption partition monitoring method and system for intelligent mold factory

Abstract

The invention discloses an energy consumption partition monitoring method and system for an intelligent mold factory, and the method comprises the steps: obtaining the energy consumption data of the intelligent mold factory through a sensor network, and generating an energy consumption data set; performing spatial clustering processing on the energy consumption data set through a spatial clustering algorithm to obtain energy consumption spatial distribution characteristics; calculating an energy consumption difference value and a physical distance between two adjacent production areas according to the plurality of energy consumption space distribution characteristics, and generating an energy consumption gradient quantization matrix; determining a high-energy-consumption area list according to the energy consumption gradient quantization matrix, and analyzing an energy flow path guided in a gradient direction by adopting a flow direction tracking algorithm to obtain position coordinates of an abnormal loss point; determining an energy consumption propagation influence range of the abnormal loss point according to the position coordinates and the energy consumption gradient quantization matrix, calculating power distribution schemes of the plurality of production areas through a dynamic programming algorithm, and generating power upper limit adjustment data according to the plurality of power distribution schemes; and updating the power distribution of the plurality of production areas through a real-time control algorithm.

Description

Technical Field

[0001] The present application relates to the field of intelligent manufacturing, and specifically, to a method and system for monitoring energy consumption zones in an intelligent mold factory. Background Art

[0002] As a core area of ​​industrial transformation and upgrading, intelligent manufacturing plays a crucial role in improving production efficiency and reducing resource consumption. Energy management is particularly important in high-precision industries such as mold processing, where it directly impacts production costs and environmental sustainability. Currently, energy consumption monitoring in mold factories relies heavily on data collection and aggregate energy consumption statistics from a single device, lacking detailed analysis and dynamic optimization across spatial dimensions. This approach fails to effectively capture the energy consumption distribution characteristics of processing equipment across different areas, making it difficult to accurately locate high-energy consumption areas and insufficiently analyzing energy flows. This in turn limits the depth and real-time nature of energy optimization.

[0003] Regarding related technologies, current energy consumption monitoring in mold factories mostly relies on data collection and total energy consumption statistics of a single device, which cannot effectively capture the energy consumption distribution characteristics of processing equipment in different areas, resulting in difficulty in accurately locating high-energy consumption areas and insufficient analysis of energy flow, which in turn limits the depth and real-time nature of energy consumption optimization. No effective solution has yet been proposed. Summary of the Invention

[0004] The embodiments of the present application provide a method and system for energy consumption zoning monitoring for an intelligent mold factory, so as to at least solve the problem in the related art that the current energy consumption monitoring of mold factories mostly relies on data collection and total energy consumption statistics of a single device, and cannot effectively capture the energy consumption distribution characteristics of processing equipment in different areas, resulting in difficulty in accurately locating high-energy consumption areas and insufficient analysis of energy flow, which in turn limits the depth and real-time performance of energy consumption optimization.

[0005] According to one embodiment of the present application, a method for monitoring energy consumption zoning in an intelligent mold factory is provided, comprising: acquiring energy consumption data of multiple production areas of the intelligent mold factory through a sensor network, and generating an energy consumption data set of the intelligent mold factory that includes time series and spatial coordinates; performing spatial clustering processing on the energy consumption data set through a spatial clustering algorithm to obtain energy consumption spatial distribution characteristics of the multiple production areas; sequentially calculating the energy consumption difference and physical distance between two adjacent production areas in the multiple production areas based on the multiple energy consumption spatial distribution characteristics, and generating an energy consumption gradient quantization matrix including gradient values, spatial coordinate pairs, and gradient directions based on multiple groups of the energy consumption differences and the physical distances, wherein the gradient value is the ratio of the energy consumption difference to the physical distance; and determining a list of high energy consumption areas based on the energy consumption gradient quantization matrix. And through the high-energy consumption area list and the energy consumption data set, a flow tracking algorithm is used to analyze the energy flow path guided by the gradient direction to obtain the position coordinates of the abnormal loss point, wherein the high-energy consumption area list includes multiple high-energy consumption areas, and the abnormal loss point is the gradient mutation point in the gradient direction; according to the position coordinates and the energy consumption gradient quantization matrix, the energy consumption propagation influence range of the abnormal loss point is determined, and according to the energy consumption propagation influence range and the real-time energy consumption change data of the multiple production areas, the power allocation scheme of the multiple production areas is calculated through a dynamic programming algorithm, and the power upper limit adjustment data of the intelligent mold factory is generated according to the multiple power allocation schemes; according to the power upper limit adjustment data, the power allocation of the multiple production areas is updated through a real-time control algorithm to obtain an optimized energy consumption distribution state.

[0006] According to another embodiment of the embodiment of the present application, an energy consumption zoning monitoring system for an intelligent mold factory is also provided, including: an acquisition module, used to acquire energy consumption data of multiple production areas of the intelligent mold factory through a sensor network, and generate an energy consumption data set of the intelligent mold factory, which includes time series and spatial coordinates; a spatial clustering module, used to perform spatial clustering processing on the energy consumption data set through a spatial clustering algorithm, and obtain energy consumption spatial distribution characteristics of the multiple production areas; a first generation module, used to calculate the energy consumption difference and physical distance of two adjacent production areas in the multiple production areas in turn according to multiple energy consumption spatial distribution characteristics, and generate an energy consumption gradient quantization matrix including gradient values, spatial coordinate pairs and gradient directions according to multiple groups of energy consumption differences and physical distances, wherein the gradient value is the ratio of energy consumption difference to physical distance; a determination module, used to determine the energy consumption gradient quantization matrix according to the energy consumption gradient quantization matrix A list of high-energy consumption areas is determined, and through the list of high-energy consumption areas and the energy consumption data set, a flow tracking algorithm is used to analyze the energy flow path guided by the gradient direction to obtain the position coordinates of the abnormal loss point, wherein the list of high-energy consumption areas includes multiple high-energy consumption areas, and the abnormal loss point is the gradient mutation point in the gradient direction; a second generation module is used to determine the energy consumption propagation influence range of the abnormal loss point according to the position coordinates and the energy consumption gradient quantization matrix, and calculate the power allocation scheme of the multiple production areas through a dynamic programming algorithm based on the energy consumption propagation influence range and the real-time energy consumption change data of the multiple production areas, and generate the power upper limit adjustment data of the intelligent mold factory according to the multiple power allocation schemes; an update module is used to update the power allocation of the multiple production areas through a real-time control algorithm according to the power upper limit adjustment data to obtain an optimized energy consumption distribution state.

[0007] According to another aspect of an embodiment of the present application, a computer-readable storage medium is further provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned energy consumption zoning monitoring method for a smart mold factory when running.

[0008] According to another aspect of an embodiment of the present application, an electronic device is also provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the energy consumption zoning monitoring method for a smart mold factory through the computer program.

[0009] In an embodiment of the present application, energy consumption data of multiple production areas of an intelligent mold factory are obtained through a sensor network to generate an energy consumption data set containing time series and spatial coordinates; the energy consumption data set is spatially clustered by a spatial clustering algorithm to obtain the energy consumption spatial distribution characteristics of these multiple production areas; the energy consumption difference and physical distance of two adjacent production areas are calculated in turn according to these energy consumption spatial distribution characteristics, and an energy consumption gradient quantization matrix containing gradient values, spatial coordinate pairs and gradient directions is generated according to the calculation results, wherein the gradient value is the ratio of the energy consumption difference to the physical distance; then a list of high energy consumption areas is determined according to the energy consumption gradient quantization matrix, and the energy flow path is analyzed by a flow tracking algorithm through the high energy consumption area list and the energy consumption data set to obtain the location coordinates of the abnormal loss point, wherein the high energy consumption area list includes multiple high energy consumption areas, and the abnormal loss point is a gradient mutation point; the energy consumption propagation of the abnormal loss point is determined according to the location coordinates and the energy consumption gradient quantization matrix The impact range is determined, and the power allocation scheme is calculated through a dynamic programming algorithm based on the energy consumption propagation impact range and the real-time energy consumption change data of the production area. Then, the power upper limit adjustment data of the intelligent mold factory is generated according to multiple power allocation schemes; finally, according to the power upper limit adjustment data, the power allocation of these multiple production areas is updated through a real-time control algorithm to obtain the optimized energy consumption distribution state; using the above scheme, the present application realizes the precise analysis and optimal regulation of the energy consumption of the intelligent mold factory, effectively identifies abnormal loss points, and reasonably allocates power to each area to achieve the purpose of reducing overall energy consumption, providing technical support for energy conservation and emission reduction in the mold manufacturing industry; thereby solving the problem in related technologies that the current energy consumption monitoring of mold factories mostly relies on data collection and total energy consumption statistics of a single device, and cannot effectively capture the energy consumption distribution characteristics of processing equipment in different areas, resulting in high-energy consumption areas being difficult to accurately locate and insufficient energy flow analysis, thereby limiting the depth and real-time performance of energy consumption optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 This is a hardware structure block diagram of a computer terminal for an optional energy consumption zoning monitoring method for an intelligent mold factory according to an embodiment of the present application; Figure 2 This is a flow chart of an optional energy consumption zoning monitoring method for an intelligent mold factory according to an embodiment of the present application; Figure 3 This is a structural block diagram of an energy consumption zoning monitoring system for an intelligent mold factory according to an embodiment of the present application. DETAILED DESCRIPTION

[0011] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this application.

[0012] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in a sequence other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0013] The method embodiments provided in the embodiments of the present application can be executed on a computer terminal or a similar computing system. Taking running on a computer terminal as an example, Figure 1 This is a hardware structure block diagram of a computer terminal for an energy consumption zoning monitoring method for an intelligent mold factory according to an embodiment of the present application. Figure 1 As shown, the computer terminal may include one or more ( Figure 1 Only one is shown) a processor 102 (the processor 102 may include but is not limited to a microprocessor MCU or a programmable logic device FPGA and other processing systems) and a memory 104 for storing data. In an exemplary embodiment, the computer terminal may also include a transmission device 106 for communication functions and an input and output device 108. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the above-mentioned computer terminal. For example, the computer terminal may also include Figure 1 More or fewer components than shown, or with Figure 1 Equivalent functions or comparisons shown Figure 1 Shown are different configurations with more functionality.

[0014] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the energy consumption zoning monitoring method for the intelligent mold factory in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, implementing the above-mentioned method. The memory 104 may include a high-speed random access memory and may also include a non-volatile memory, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include a memory remotely located relative to the processor 102, and these remote memories may be connected to a secure text via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0015] Transmission system 106 is configured to receive or transmit data via a network. A specific example of such a network may include a wireless network provided by a computer terminal's communications provider. In one embodiment, transmission system 106 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet.

[0016] In related technologies, energy consumption data of mold processing equipment is usually collected in the form of time series, but lacks correlation analysis in the spatial dimension, making it difficult for factories to grasp the degree of dispersion of energy consumption in different production areas. As a result, it becomes difficult to identify high-energy consumption areas and energy flows, and factories cannot accurately determine which areas or equipment are the main sources of energy consumption anomalies. Furthermore, due to the lack of quantitative analysis of energy consumption gradients between adjacent areas, abnormal loss points in the energy transfer process are difficult to be discovered in a timely manner, affecting the accuracy of energy consumption optimization. On a deeper level, existing systems usually adopt static power allocation strategies, which make it difficult to dynamically adjust the power upper limit of each partition according to real-time energy consumption changes, resulting in waste of resources or reduced production efficiency.

[0017] Therefore, how to accurately identify high-energy consumption areas and abnormal loss points through spatial energy consumption data analysis, and dynamically optimize the power distribution of each partition, has become a key issue that needs to be urgently addressed in the energy consumption management of smart mold factories.

[0018] In order to solve the above problems, this embodiment provides an energy consumption zoning monitoring method for an intelligent mold factory, which is applied to a computer terminal. Figure 2 This is a flow chart of an optional energy consumption zoning monitoring method for a smart mold factory according to an embodiment of the present application, the process comprising the following steps: Step S202: acquiring energy consumption data of multiple production areas of the smart mold factory through a sensor network, and generating an energy consumption dataset of the smart mold factory including time series and spatial coordinates; Step S204, performing spatial clustering processing on the energy consumption dataset using a spatial clustering algorithm to obtain energy consumption spatial distribution characteristics of the multiple production areas; Step S206, sequentially calculating the energy consumption difference and the physical distance between two adjacent production areas in the multiple production areas based on the multiple energy consumption spatial distribution characteristics, and generating an energy consumption gradient quantization matrix including gradient values, spatial coordinate pairs, and gradient directions based on the multiple sets of energy consumption differences and physical distances, wherein the gradient value is a ratio of the energy consumption difference to the physical distance; Step S208: determining a list of high energy consumption areas based on the energy consumption gradient quantization matrix, and using the list of high energy consumption areas and the energy consumption dataset, employing a flow tracing algorithm to analyze the energy flow path guided by the gradient direction, to obtain the position coordinates of abnormal energy consumption points, wherein the list of high energy consumption areas includes multiple high energy consumption areas, and the abnormal energy consumption points are gradient mutation points in the gradient direction; Step S210: determining the energy consumption propagation impact range of the abnormal loss point based on the location coordinates and the energy consumption gradient quantization matrix, calculating power allocation plans for the multiple production areas using a dynamic programming algorithm based on the energy consumption propagation impact range and the real-time energy consumption change data of the multiple production areas, and generating power upper limit adjustment data for the smart mold factory based on the multiple power allocation plans; Step S212: updating the power distribution of the plurality of production areas through a real-time control algorithm according to the power upper limit adjustment data to obtain an optimized energy consumption distribution state.

[0019] Through the above steps, the energy consumption data of multiple production areas of the smart mold factory are obtained through the sensor network to generate an energy consumption data set containing time series and spatial coordinates; the energy consumption data set is spatially clustered by the spatial clustering algorithm to obtain the energy consumption spatial distribution characteristics of these multiple production areas; the energy consumption difference and physical distance of two adjacent production areas are calculated in turn according to these energy consumption spatial distribution characteristics, and an energy consumption gradient quantization matrix containing gradient values, spatial coordinate pairs and gradient directions is generated according to the calculation results, wherein the gradient value is the ratio of the energy consumption difference to the physical distance; then a list of high energy consumption areas is determined according to the energy consumption gradient quantization matrix, and the energy flow path is analyzed by the flow tracking algorithm through the high energy consumption area list and the energy consumption data set to obtain the position coordinates of the abnormal loss point, wherein the high energy consumption area list includes multiple high energy consumption areas, and the abnormal loss point is the gradient mutation point; the energy consumption propagation impact of the abnormal loss point is determined according to the position coordinates and the energy consumption gradient quantization matrix The impact range is determined, and the power allocation scheme is calculated through a dynamic programming algorithm based on the energy consumption propagation impact range and the real-time energy consumption change data of the production area. Then, the power upper limit adjustment data of the intelligent mold factory is generated according to multiple power allocation schemes; finally, according to the power upper limit adjustment data, the power allocation of these multiple production areas is updated through a real-time control algorithm to obtain the optimized energy consumption distribution state; using the above scheme, the present application realizes the precise analysis and optimal regulation of the energy consumption of the intelligent mold factory, effectively identifies abnormal loss points, and reasonably allocates power to each area to achieve the purpose of reducing overall energy consumption, providing technical support for energy conservation and emission reduction in the mold manufacturing industry; thereby solving the problem in related technologies that the current energy consumption monitoring of mold factories mostly relies on the data collection and total energy consumption statistics of a single device, and cannot effectively capture the energy consumption distribution characteristics of processing equipment in different areas, resulting in difficulty in accurately locating high-energy consumption areas and insufficient analysis of energy flow, thereby limiting the depth and real-time performance of energy consumption optimization.

[0020] Optionally, the energy consumption dataset is spatially clustered by a spatial clustering algorithm to obtain the spatial distribution characteristics of energy consumption of the multiple production areas, including: spatially clustering the energy consumption dataset according to the spatial coordinates by the spatial clustering algorithm to obtain a regional division dataset; calculating the energy consumption average value and standard deviation of each of the multiple production areas according to the regional division dataset, and generating characteristic vectors of the multiple production areas according to the energy consumption average value and standard deviation, wherein the characteristic vector is used to indicate the spatial distribution pattern of energy consumption data of the production area; when there is a first production area among the multiple production areas whose energy consumption average value is greater than a first preset threshold, the first production area is marked as an energy consumption abnormality area through an abnormal area identifier to obtain the energy consumption spatial distribution characteristics of the multiple production areas, wherein the energy consumption spatial distribution characteristics include the characteristic vector and the abnormal area identifier.

[0021] First, a spatial clustering algorithm (such as K-means or DBSCAN) is used to cluster the energy consumption dataset according to spatial coordinates. Next, the mean and standard deviation of energy consumption are calculated for each region. The mean represents the average level of energy consumption within the region, while the standard deviation reflects the fluctuation in energy consumption within the region. These two statistical indicators form a feature vector, which describes the spatial distribution of energy consumption data for a specific production area. Next, the system aggregates the energy consumption readings from all sensors within each region to calculate the average energy consumption level for that region. After calculating the feature vectors for all production regions, the system analyzes these data to determine whether any region's average energy consumption significantly exceeds a preset first threshold. If such a region is found, namely the first production region, it is marked as an "abnormal energy consumption region" using an abnormal region identifier. Ultimately, through this process, the system generates a dataset of energy consumption spatial distribution characteristics, including feature vectors and abnormal region identifiers.

[0022] Optionally, the energy consumption difference and physical distance between two adjacent production areas in the multiple production areas are calculated in sequence according to the multiple energy consumption spatial distribution characteristics, and an energy consumption gradient quantization matrix containing gradient values, spatial coordinate pairs and gradient directions is generated according to the multiple groups of energy consumption differences and the physical distances, including: using a spatial interpolation algorithm to fill missing values ​​for the multiple energy consumption spatial distribution characteristics according to the spatial coordinates of the energy consumption data set to fill the energy consumption data of the second production area, and obtain multiple standardized energy consumption spatial distribution characteristics, wherein the intelligent mold factory includes the multiple production areas and the second production area; according to the multiple standardized energy consumption spatial distribution characteristics The energy consumption average value is used to calculate the energy consumption difference and the physical distance between two adjacent production areas in the multiple production areas and the second production area in turn; the spatial coordinate pair is determined according to the neighborhood relationship between the multiple production areas and the second production area, and a data table containing the energy consumption difference, the physical distance and the spatial coordinate pair is obtained; when the energy consumption difference and the physical distance in the data table are both non-zero values, the gradient value is determined by calculating the ratio of the energy consumption difference to the physical distance, and the gradient direction is determined according to the gradient value through a gradient calculation method; the gradient value, the spatial coordinate pair and the gradient direction are converted into the energy consumption gradient quantization matrix through a matrix generation algorithm.

[0023] During the energy consumption data collection process, some areas may be missing data. This may be due to equipment failure, network issues, or sensor failure. To ensure comprehensive and accurate analysis, this embodiment proposes using a spatial interpolation algorithm to fill in missing values. Spatial interpolation estimates the energy consumption values ​​of surrounding areas to fill in the gaps. For example, if energy consumption data for the second production area is missing, the system will use interpolation methods (such as inverse distance weighting or kriging) based on the energy consumption values ​​of its neighboring areas to fill in the missing data, thereby obtaining a complete and standardized set of energy consumption spatial distribution features. After filling in and standardizing, the system calculates the energy consumption difference and physical distance between adjacent production areas. The energy consumption difference reflects the difference in energy consumption levels between the two areas, while the physical distance is the actual spatial separation between the two areas. This calculation is crucial for understanding how energy flows in space, helping to identify high and low energy consumption points and potential losses along the energy transmission path. The system then iterates through all production areas and, for each pair of adjacent areas, calculates the difference in their average energy consumption. For example, area A is adjacent to area B, and their average energy consumption is 100kWh and 80kWh respectively, so the difference is 20kWh. At the same time, the system measures the physical distance between the adjacent areas. For example, the distance between area A and area B is 20 meters. Next, based on the calculated energy consumption difference and physical distance, the system generates an energy consumption gradient quantization matrix. The gradient value is calculated by the ratio of the energy consumption difference to the physical distance. It represents the rate of change of energy consumption per unit distance, and thus reflects the intensity of energy flow. The gradient direction indicates the direction of energy flow and is the result of the combined analysis of the gradient value and the spatial coordinates.

[0024] Finally, the gradient value, spatial coordinate pair (such as the coordinates of A and B), and gradient direction are stored in a matrix to form an energy gradient quantization matrix.

[0025] Optionally, a list of high-energy consumption areas is determined based on the energy consumption gradient quantization matrix, including: determining whether the gradient values ​​of the multiple production areas exceed a second preset threshold, and determining the production areas whose gradient values ​​exceed the second preset threshold as high-energy consumption areas, to obtain a preliminary list of high-energy consumption areas; performing regional division on the multiple high-energy consumption areas in the preliminary list of high-energy consumption areas through the spatial clustering algorithm, to obtain multiple clustered high-energy consumption areas; obtaining the regional boundaries and coordinate ranges of the multiple clustered high-energy consumption areas, to obtain an optimized list of high-energy consumption areas; mapping the spatial coordinate pairs and gradient directions of the multiple high-energy consumption areas to the optimized list of high-energy consumption areas, to obtain the list of high-energy consumption areas.

[0026] First, from the energy consumption gradient quantification matrix, the system will check the gradient value of each production area one by one to see if it exceeds the second preset threshold value set in advance. If the gradient value of a certain area does exceed the second preset threshold value, then this area will be marked as a "high energy consumption area" and added to the preliminary list of high energy consumption areas. This list contains all production areas that are initially identified as having excessive energy consumption, providing a starting point for further analysis and optimization. After the preliminary list of high energy consumption areas is generated, a spatial clustering algorithm is used to process these areas. The clustering algorithm can further integrate the high energy consumption areas in the preliminary list into several large high energy consumption area clusters based on the geographical location of the production areas and the similarity of the gradient direction. Each high energy consumption area cluster has its own clear boundaries and coordinate range.

[0027] Finally, the system maps the spatial coordinate pairs and gradient directions of each high-energy consumption area cluster back to the original factory layout, generating a final, optimized list of high-energy consumption areas. This list not only clearly identifies the high-energy consumption areas, but also provides their precise locations and impact areas, as well as their gradient directions.

[0028] Optionally, a flow tracing algorithm is used to analyze the energy flow path guided by the gradient direction through the list of high-energy consumption areas and the energy consumption data set to obtain the location coordinates of the abnormal loss point, including: obtaining the spatial coordinates and gradient directions of the multiple production areas from the energy consumption data set and the list of high-energy consumption areas, and using the flow tracing algorithm to calculate the energy flow path based on multiple groups of the spatial coordinates and the gradient directions to obtain a path data set, wherein the path data set includes path nodes and the gradient directions; sequentially determining whether the difference between the first gradient directions of multiple path nodes in the path data set and the second gradient directions of adjacent path nodes of the multiple path nodes exceeds a third preset threshold; determining the path nodes whose difference exceeds the third preset threshold as the gradient mutation points, and generating a preliminary abnormal loss point set based on the gradient mutation points and the spatial coordinates of the gradient mutation points; using the spatial clustering algorithm to divide the multiple gradient mutation points into regions based on the spatial coordinates of the multiple gradient mutation points in the preliminary abnormal loss point set, and obtaining the region boundaries and coordinate ranges after division, and determining the region boundaries and the coordinate ranges as the location coordinates of the abnormal loss point.

[0029] The process for determining the location coordinates of abnormal energy loss points involves first extracting the spatial coordinates and gradient directions of each production area from the energy consumption dataset and the list of high-energy consumption areas. The spatial coordinates are used for positioning, while the gradient directions indicate the direction of energy flow and serve as the starting point for tracing energy flow paths. A flow-tracking algorithm then uses the spatial coordinates and gradient directions to calculate the energy flow path from low-energy consumption areas to high-energy consumption areas. Within the path dataset, the system compares the gradient directions of adjacent nodes along the path to determine whether the difference exceeds a third preset threshold. This threshold identifies abnormal changes in energy flow, which may indicate a sudden energy leak or loss at a specific point. If the gradient direction difference between adjacent nodes is significant, exceeding the third preset threshold, the system labels the node as a "gradient mutation point" and records its spatial coordinates, forming a preliminary set of abnormal energy loss points. These mutation points may be key locations of energy loss. The system then applies a spatial clustering algorithm to this preliminary set of abnormal energy loss points, grouping closely located gradient mutation points and dividing them into regions. By obtaining the boundaries and coordinate ranges of each divided region, the precise location of the abnormal energy loss point is ultimately determined.

[0030] Optionally, the energy consumption propagation influence range of the abnormal loss point is determined according to the position coordinates and the energy consumption gradient quantization matrix, including: performing regional segmentation on the energy consumption gradient quantization matrix through a grid division algorithm and the position coordinates to generate multiple matrix partition data containing the abnormal loss point, and determining the energy consumption distribution characteristics of the abnormal loss point according to the matrix partition data; analyzing the gradient direction strength of multiple third production areas through a gradient calculation algorithm according to the matrix partition data, wherein the multiple third production areas correspond one-to-one to the multiple matrix partition data, and the gradient direction strength is used to indicate the energy change rate per unit distance; when it is determined that the gradient direction strength of a fourth production area among the multiple third production areas is greater than a fourth preset threshold, generating an initial flow path corresponding to the fourth production area through matrix operation, and obtaining gradient direction data; determining the overlapping area according to the gradient direction data and the initial flow path through a path tracing algorithm, and performing a comprehensive analysis of the overlapping area and the energy consumption distribution of the overlapping area through data fusion technology to generate spatial overlapping feature data, and determining the propagation boundary according to the spatial overlapping feature data; performing boundary segmentation on the propagation boundary through a boundary division algorithm to obtain boundary data, and determining the energy consumption propagation influence range according to the boundary data.

[0031] First, a gridding algorithm is used to partition the energy gradient quantization matrix, generating multiple matrix partitions. Each partition contains energy gradient data and a specific coordinate range, facilitating local analysis of the energy consumption distribution at the abnormal loss point. By associating the coordinates of the abnormal loss point, the system generates multiple matrix partitions containing information about the abnormal loss point. This data includes not only the energy gradient information for the abnormal loss point itself, but also relevant data for the surrounding area. For the multiple third production areas corresponding to the matrix partition data, a gradient calculation algorithm is used to analyze their gradient directional strength. The gradient directional strength reflects the speed of energy change per unit distance and is a key indicator for determining the strength of energy flow. If the gradient directional strength of a fourth production area exceeds a fourth preset threshold, this indicates that energy flow in that area is abnormally strong and may be affected by the abnormal loss point. For the fourth production area with abnormal gradient directional strength, the system uses matrix operations to generate an initial flow path, i.e., the path along which energy diffuses from the abnormal loss point to the surrounding area.

[0032] Next, a path tracing algorithm is used, combined with gradient direction data and the initial flow path, to identify overlapping areas affected by abnormal loss points. A comprehensive analysis of the energy consumption distribution in the overlapping areas is conducted, and data fusion technology is used to integrate data from different partitions to understand the energy propagation pattern from a more comprehensive perspective. The generated spatial overlapping feature data contains detailed energy consumption information of the overlapping areas, including but not limited to energy density, distribution uniformity, etc. The propagation boundary is then determined based on the spatial overlapping feature data, that is, the maximum range of energy impact of the abnormal loss point. The system can segment and define possible propagation boundaries through a boundary partitioning algorithm to ensure the accuracy of the boundary data. Finally, based on the boundary data, the system clearly marks the energy consumption propagation impact range of the abnormal loss point, that is, the set of all affected areas.

[0033] Optionally, the power allocation schemes for the multiple production areas are calculated through a dynamic programming algorithm based on the energy consumption propagation influence range and the real-time energy consumption change data of the multiple production areas, and the power upper limit adjustment data of the smart mold factory are generated based on the multiple power allocation schemes, including: determining the energy consumption propagation influence range and loss weight of the abnormal loss point to obtain the energy consumption change data of the multiple production areas; calculating the multiple power allocation schemes based on the energy consumption change data and the loss weight through the dynamic programming algorithm; generating the power upper limit adjustment values ​​for the multiple production areas based on the power allocation scheme and the preset adjustment frequency; and determining the power upper limit adjustment data based on the comparison results of the multiple real-time energy consumption change data and the multiple power upper limit adjustment values.

[0034] First, the system determines the actual impact of the abnormal power loss point on surrounding areas. This involves analyzing the gradient-directed energy flow paths and spatial overlap to determine which areas are most affected by the abnormal power loss point. Based on the severity of the abnormal power loss point and the energy consumption changes in the affected areas, the system calculates a power loss weight. This weight reflects the magnitude of the abnormal power loss point's impact on the factory's total energy consumption. Input data includes real-time energy consumption changes for each production area and the power loss weight of the abnormal power loss point. Based on this data, the algorithm iteratively adjusts the power allocation for each area until it finds a solution that maintains production efficiency while optimizing overall energy use. During the calculation process, the algorithm considers the factory's power supply capacity and the load conditions of each area to ensure optimal power distribution. Once the optimal power allocation solution is calculated, the system generates power ceiling adjustments for each production area based on this solution and a preset adjustment frequency.

[0035] Finally, the system monitors each zone's actual energy consumption in real time and compares this data with the calculated power cap adjustment. If a zone's actual energy consumption approaches or exceeds its adjusted value, the system determines that the power cap for that zone requires further adjustment to accommodate the new energy consumption distribution or to compensate for any abnormal consumption points.

[0036] In an optional embodiment, the present application provides an optional energy consumption zoning monitoring method for a smart mold factory, which mainly includes: S201, acquiring energy consumption data of each production area of ​​the mold factory through a sensor network, and generating an energy consumption dataset including time series and spatial coordinates; S202, based on the energy consumption data set, a spatial clustering algorithm is used to divide the production areas to obtain the energy consumption spatial distribution characteristics of each area; Energy consumption data for each production area is obtained from the sensor network. A preset sampling frequency is used to record the time series using timestamps. Combined with sensor location information, an original data set containing spatial coordinates is generated to obtain the original energy consumption data set. A K-means clustering algorithm is used for the original energy consumption data set to cluster the production areas according to their spatial coordinates, generating clustering results containing groupings of each area to obtain the regional division data set. Based on the regional division data set, the average energy consumption and standard deviation of each area are calculated to generate a feature vector describing the spatial distribution of energy consumption in each area to obtain the distribution feature data set. If the average energy consumption of a region in the distribution feature data set exceeds a preset threshold, the region is marked as an energy consumption abnormality region, and a final distribution feature data set containing the abnormal area identifier is generated to obtain the final energy consumption distribution data set.

[0037] For example, in a mold factory, energy consumption data is collected through a sensor network, ensuring that the data covers all production areas and has temporal and spatial characteristics. The sensor network is deployed in 10 production areas of the factory, with a total of 80 sensors, and a sampling frequency of once every 5 minutes. Each data point is recorded with a timestamp, such as 2025-04-30 09:00:00, and the sensor location coordinates, such as (x2, y2, z2), to form the raw energy consumption dataset. The data is in kilowatt-hours; for example, a sensor records 45 kWh at 09:00:00. This collection method facilitates capturing time series and spatially distributed energy consumption changes, providing a comprehensive data foundation for subsequent analysis.

[0038] In one possible implementation, the K-means clustering algorithm is used to partition the production areas based on the original energy consumption dataset. K-means clustering calculates the Euclidean distance between sensor coordinates, grouping spatially close sensors into the same cluster. Assuming the default number of clusters, K, is 4, clustering will divide the 80 sensors into four regional groups. For example, Region A contains 20 sensors, and Region B contains 25 sensors. The clustering results generate a regional partitioning dataset, recording the cluster label for each sensor, such as sensor S1 belonging to Region A. This partitioning reflects the energy consumption distribution characteristics of the factory's spatial layout and facilitates regional energy consumption management.

[0039] S203, calculating the energy consumption difference and physical distance between adjacent regions based on the energy consumption spatial distribution characteristics, and generating an energy consumption gradient quantization matrix including gradient values, spatial coordinate pairs, and gradient directions; Energy consumption data and spatial coordinates are obtained from the data grid, and the energy consumption data of unmeasured areas are filled in using a spatial interpolation algorithm to obtain a standardized energy consumption data set. Based on the standardized energy consumption data set, the energy consumption difference and physical distance between adjacent areas are calculated, and the coordinate pairs are determined through the neighborhood relationship to obtain a data table containing the difference, distance, and coordinate pairs. If the energy consumption difference and physical distance in the data table are both non-zero, the gradient value is calculated and the gradient direction is determined to obtain a gradient value and direction set. Using a matrix generation algorithm, the gradient value, the spatial coordinate pair, and the gradient direction are stored as a quantization matrix to obtain an energy consumption gradient quantization matrix.

[0040] For example, when obtaining energy consumption data and spatial coordinates from a data grid, assume a sensor network is deployed throughout a large factory, covering key points in the production area. Each sensor records energy consumption data, such as hourly electricity consumption, along with spatial coordinates, such as x, y, and z values ​​in a three-dimensional coordinate system. Sensors may not be deployed in certain areas due to equipment limitations, resulting in missing data. To address this issue, spatial interpolation algorithms, such as the inverse distance weighted method (IDW), can be used. In principle, this algorithm estimates the energy consumption of unmeasured points based on the energy consumption and distance of known points. For example, if an unmeasured point is 2 meters, 3 meters, and 5 meters away from three known points, with energy consumption values ​​of 100 kWh, 120 kWh, and 80 kWh, respectively, the energy consumption value at that point can be estimated by taking a weighted average. This method effectively fills in data gaps, generating a standardized energy consumption data set and providing a complete data foundation for subsequent analysis.

[0041] In one possible implementation, based on a standardized set of energy consumption data, the energy consumption difference and physical distance between adjacent areas are calculated to establish a neighborhood relationship. Assume that the factory is divided into 10 areas, each with a certain center coordinate. The energy consumption difference between adjacent areas is obtained by comparing the energy consumption values ​​of the center points. For example, if the energy consumption of area A is 150 kWh and that of area B is 130 kWh, the difference is 20 kWh. The physical distance is calculated using coordinates. For example, if the distance between the center points of the two areas is 10 meters, the neighborhood relationship can be determined by setting a distance threshold, such as only considering pairs of areas with a distance less than 15 meters. This method can clearly define spatial correlation, facilitating further analysis of energy consumption changes between areas.

[0042] S204, if the gradient value of a certain area in the energy consumption gradient quantization matrix exceeds a preset threshold, the area is determined to be a high energy consumption area by combining the spatial coordinate pair and the gradient direction, and a high energy consumption area list is generated; The gradient value and the corresponding spatial coordinate pair of each region are obtained from the energy consumption gradient quantization matrix, and the gradient direction is determined by a gradient calculation method to obtain a first data set including the gradient value, the spatial coordinate pair and the gradient direction. For the first data set, if the gradient value of a certain region exceeds a preset threshold, a threshold judgment is performed based on the gradient direction and the spatial coordinate pair to determine the high-energy consumption region and generate a preliminary high-energy consumption region set. Based on the preliminary high-energy consumption region set, a spatial clustering algorithm is used to divide the high-energy consumption region into regions, and the boundaries and coordinate ranges of the divided regions are obtained to obtain an optimized high-energy consumption region set. Through the optimized high-energy consumption region set, the spatial coordinate pair and the gradient direction of each high-energy consumption region are mapped to a region list, and a data mapping method is used to generate a final high-energy consumption region list.

[0043] For example, obtaining the gradient value and spatial coordinate pair for each region from the energy gradient quantization matrix involves parsing the matrix structure to extract key information. The energy gradient quantization matrix is ​​typically stored in a grid format, with rows and columns corresponding to spatial coordinates, and cells storing gradient values ​​and directions.

[0044] In one possible implementation, assuming that the gradient value of a certain area is 3.2 and the coordinates are (4, 6), by comparing the energy consumption of eight adjacent areas, the direction is determined to be northeast, and a first data set is generated, including the gradient value 3.2, the coordinates (4, 6), and the direction northeast.

[0045] Preferably, a spatial clustering algorithm is used to divide the preliminary high energy consumption area set into regions. Spatial clustering algorithms such as DBSCAN group the regions based on coordinate distance and gradient direction.

[0046] For example, coordinates (4,6) and (5,7) are both marked as high-energy consumption areas. Since the distance between them is less than the preset radius of 0.5 and their directions are similar, they are classified into the same cluster. After division, the boundaries and coordinate ranges of each cluster are obtained. For example, the boundary of a cluster is the rectangular area [(4,6), (6,8)].

[0047] In one embodiment, 10 high energy consumption areas are clustered into 3 clusters to generate an optimized set of high energy consumption areas, thereby improving the consistency of area division.

[0048] It can be understood that in order to generate the final high-energy consumption area list through the optimized high-energy consumption area set, it is necessary to map the spatial coordinate pairs and gradient directions to a structured list.

[0049] For example, a cluster contains coordinates (4,6) and (5,7), both in the northeast direction, which is mapped to the list item: {region 1: coordinates [(4,6), (5,7)], direction: northeast}.

[0050] S205, using the high energy consumption area list and the energy consumption data set, a flow tracking algorithm is used to analyze the energy flow path guided by the gradient direction, and the coordinates of the abnormal loss points where the path is interrupted or the gradient direction suddenly changes are obtained; The spatial coordinates and gradient direction of each area are obtained from the energy consumption dataset and the list of high-energy consumption areas. A flow tracing algorithm is used to calculate the energy flow path, resulting in a path dataset containing path nodes and the gradient directions. For this path dataset, if the difference between the gradient direction of a path node and that of an adjacent node exceeds a preset threshold, the node is identified as a gradient mutation point. The spatial coordinates of the mutation point are obtained to obtain a preliminary set of abnormal energy loss points. Based on this preliminary set of abnormal energy loss points, a spatial clustering algorithm is used to partition the abnormal energy loss points into regions. The boundaries and coordinate ranges of the partitioned regions are then determined to obtain an optimized set of abnormal energy loss points.

[0051] In one possible implementation, suppose a commercial district has 100 zones. A dataset records the coordinates and direction of each zone, such as zone A with coordinates (5, 7) pointing southeast and zone B with coordinates (6, 8) pointing east. The flow-tracking algorithm analyzes the gradient direction to trace the energy flow path. In principle, the algorithm starts from the starting zone and connects adjacent zones one by one based on the gradient direction to form a path.

[0052] For example, starting from coordinate (5,7), heading southeast, connect to (6,8), and then according to the direction east of (6,8), connect to (7,8), forming the path [(5,7)→(6,8)→(7,8)], and generating a path dataset, which includes node coordinates and directions.

[0053] Specifically, to determine the gradient mutation point in a path dataset, we need to compare the gradient direction differences between adjacent nodes. Assuming a preset threshold of 45 degrees, if the direction of node (6,8) in a path is east and the direction of the next node (7,8) is north, and the difference of 90 degrees exceeds the threshold, then (7,8) is marked as a mutation point.

[0054] In one embodiment, 15 mutation points are identified in 100 paths to generate a preliminary abnormal loss point set, with their coordinates recorded as (7, 8). The mutation points usually reflect abnormal interruptions in energy consumption flow, which may be caused by equipment failure or pipeline blockage.

[0055] It should be noted that the threshold selection is based on historical data analysis to ensure that significant anomalies are screened out.

[0056] Preferably, a spatial clustering algorithm is used to divide the preliminary abnormal loss point set into regions. The DBSCAN algorithm groups the points based on coordinate distance and direction similarity.

[0057] For example, the distance between mutation points (7,8) and (8,9) is less than the preset radius 0.5, and the direction difference is less than 30 degrees, so they are classified into the same cluster.

[0058] In one embodiment, 15 mutation points are clustered into four clusters. The boundaries of each cluster are obtained, such as a rectangular region [(7, 8), (9, 10)], to generate an optimized set of abnormal loss points. Clustering improves the spatial coherence of abnormal points, making it easier to locate problem areas.

[0059] S206, analyzing the spatial overlap characteristics of the gradient direction and the flow path based on the position coordinates of the abnormal loss point and the energy consumption gradient quantization matrix, and determining the energy consumption propagation impact range of the abnormal loss point; The spatial positioning data is obtained from the position coordinates of the abnormal loss point through coordinate mapping, and the grid division algorithm is used to perform regional segmentation on the energy consumption gradient quantization matrix to generate matrix partition data containing the position of the abnormal point, and the distribution characteristics of the abnormal point in the matrix are determined. Based on the matrix partition data, the gradient calculation algorithm is used to perform directional analysis on the energy consumption gradient quantization matrix. If the intensity of the gradient direction exceeds the preset threshold, the initial flow path is generated through matrix operation to obtain gradient direction data. The overlapping area data is obtained from the gradient direction data and the initial flow path through the path tracing algorithm. The overlapping area and energy consumption distribution are comprehensively analyzed using data fusion technology to generate spatial overlapping feature data and determine the propagation boundary of the overlapping area. Based on the spatial overlapping feature data, the boundary division algorithm is used to perform regional segmentation on the propagation boundary. The spatial consistency of the boundary data is verified through matrix operation to generate boundary data of the energy consumption propagation influence range and determine the propagation range of the abnormal loss point.

[0060] For example, in an energy consumption monitoring scenario, coordinate mapping is used to convert the spatial coordinates of abnormal consumption points into spatial positioning data. For example, suppose an industrial park is equipped with energy consumption sensors. The coordinates of the abnormal consumption point are (x1, y1, z1). Through coordinate mapping, these coordinates can be converted into positioning data in a three-dimensional grid, such as (10, 15, 5), facilitating subsequent analysis. This mapping can be based on the actual geographic information system of the scene to ensure location accuracy.

[0061] In one possible implementation, a grid partitioning algorithm partitions the energy gradient quantization matrix into regions. Assume the matrix is ​​a 100×100 two-dimensional grid, with each grid point recording the energy gradient value. The grid partitioning algorithm can divide the matrix into 10×10 subregions, generating matrix partition data.

[0062] For example, the outlier point is located in subregion (3,4). Its distribution characteristics show that the energy consumption gradient in this area is higher than that of the surrounding area, suggesting the possibility of equipment failure. The partition data provides a spatial basis for subsequent path analysis.

[0063] Specifically, a gradient calculation algorithm is used to analyze the directionality of matrix partitioned data. Assuming the gradient strength within subregion (3,4) is 50, exceeding a preset threshold of 30, matrix operations are used to generate an initial flow path. This path might show energy consumption spreading from (3,4) to (4,5), indicating that the outlier is affecting neighboring areas. This directional analysis helps identify the propagation trend of energy consumption anomalies.

[0064] Preferably, the path tracing algorithm extracts overlapping area data from the gradient direction data and the initial flow direction path.

[0065] For example, path tracing revealed that subregions (4,5) and (5,5) both appear in multiple flow paths, indicating overlapping areas. Data fusion technology further combined energy consumption distribution to generate spatial overlap signature data, showing that energy consumption in the overlapping area is 20% higher than the average, suggesting that the anomalous propagation is concentrated there. The propagation boundary can be preliminarily determined to be an area centered at (4,5) with a radius of two grid cells.

[0066] For example, the boundary partitioning algorithm segments the propagation boundary into regions. Assuming the set of overlapping boundary points is {(4,5), (5,5), (4,6)}, the algorithm generates a boundary range of (3-5, 4-6). Matrix operations verify the consistency of the boundary data, ensuring that all boundary points are located within the high-energy consumption area. The resulting boundary data indicates that the propagation range of the abnormal loss point covers approximately nine grid cells, clearly defining the area affected by the anomaly and facilitating precise investigation.

[0067] S207, based on the energy consumption propagation impact range of the abnormal loss point and the real-time energy consumption change data, a dynamic programming algorithm is used to calculate the power allocation plan for each area and generate a power upper limit adjustment table; Real-time energy consumption data and abnormal loss points for each area are acquired from data acquisition equipment. Combined with preset zone division rules, the propagation range and loss weight of these abnormal loss points are determined to obtain energy consumption change data for each area. A dynamic programming algorithm is used to calculate the power allocation plan for each area based on the energy consumption change data and loss weights. If the energy consumption change in a particular area exceeds a preset threshold, the allocation priority of that area is adjusted to obtain a preliminary power allocation table. Based on this power allocation table and the preset adjustment frequency, a power upper limit adjustment value is generated for each area. The final power upper limit adjustment table is determined by comparing the real-time energy consumption data with these upper limit adjustment values.

[0068] For example, in an industrial park energy consumption management scenario, data collection equipment is typically smart sensors deployed in various areas to collect real-time energy consumption data and abnormal energy consumption points. Assume the park is divided into 50 areas, each with sensors installed to record hourly energy consumption. For example, area A1 consumes 200 kWh. Sensors detect abnormal energy consumption points. For example, a device in area A2 experiences a sudden increase in energy consumption to 300 kWh due to a malfunction, exceeding the normal range. Zoning rules can be based on geographic location or functional zoning. For example, the park can be divided into 5×10 grids, with A2 located in grid (2,3). By analyzing the energy consumption data of the abnormal point and combining it with the grid division, it can be preliminarily determined that its propagation range covers the adjacent grids (2,2) and (2,4). The loss weight is 1.5, indicating a significant impact on the surrounding areas. After generating energy consumption change data, it shows that energy consumption in area A2 has increased by 50%.

[0069] In one possible implementation, a dynamic programming algorithm is used to optimize power allocation. Assuming the goal is to minimize overall energy consumption fluctuations, the algorithm takes energy consumption variation data and loss weights as input and calculates a power allocation plan for each zone.

[0070] For example, due to its high loss weight, area A2 is prioritized for additional power allocation to stabilize equipment operation. A preliminary allocation table is generated: 300 kilowatts allocated to A2 and 180 kilowatts to A1. The algorithm iteratively compares the total power to ensure that it remains within the park's upper limit of 10,000 kilowatts.

[0071] Specifically, if a region's energy consumption changes by more than a preset threshold, such as a 50% increase in A2 exceeding the 30% threshold, its allocation priority is adjusted. The dynamic programming algorithm reorders the priorities, raising A2's priority. The power allocation table is updated to allocate 320 kilowatts to A2 and 170 kilowatts to A1. This adjustment ensures stability in the outlier region.

[0072] S208 , using the power upper limit adjustment table and a real-time control algorithm to update the power allocation of each area, to obtain an optimized energy consumption distribution state.

[0073] The real-time energy consumption data and abnormal loss points of each area are obtained from the data acquisition equipment. Combined with the preset area division rules, the propagation range of the abnormal loss and the loss weight are determined to obtain the energy consumption change data of each area. A real-time control algorithm is used to calculate the power allocation plan for each area based on the energy consumption change data and the loss weight. If the energy consumption change data exceeds the preset threshold, the allocation priority of the area is adjusted first to obtain a preliminary power allocation table. Based on the preliminary power allocation table and the preset adjustment frequency, the power upper limit adjustment value of each area is generated. By comparing the real-time energy consumption data with the power upper limit adjustment value, the final power upper limit adjustment table is determined. Through the real-time control algorithm, the power allocation of each area is updated according to the final power upper limit adjustment table to obtain the optimized energy consumption distribution state.

[0074] For example, acquiring real-time power demand data for each area from sensors is a core component of energy management systems. Sensors are typically deployed on electrical equipment in each area, such as motors in industrial plants and air conditioning systems in office buildings. The collected power demand data reflects the power consumption status of the equipment at a specific moment. Recording the data collection moment with a timestamp ensures data timeliness and traceability.

[0075] For example, an industrial park has three areas, A, B, and C. Sensors collect data every five minutes, recording the power demand at 10:00 AM as 500 kW in area A, 300 kW in area B, and 200 kW in area C. The data is timestamped to the second. This facilitates subsequent analysis of power demand fluctuations.

[0076] Preferably, a state feedback mechanism is used to obtain energy consumption distribution data for the adjusted power allocation plan, and the energy consumption distribution status is recorded with a timestamp to further optimize system operation. This state feedback mechanism typically involves real-time communication between sensors and the control center. For example, after the adjusted allocation plan is implemented, the sensor feedback indicates that the actual power in area A is 590kW, area B is 245kW, and area C is 55kW, with a timestamp of 10:05. The energy consumption distribution status records this data, forming a dynamic energy consumption curve to facilitate analysis of the adjustment effect. This mechanism supports the system's refined management of power allocation through continuous monitoring and data recording.

[0077] Through the description of the above embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the existing technology, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0078] Figure 3 is a structural block diagram of an energy consumption zoning monitoring system for an intelligent mold factory according to an embodiment of the present application; Figure 3 Shown, including: An acquisition module 31 is configured to acquire energy consumption data of multiple production areas of the smart mold factory through a sensor network, and generate an energy consumption dataset of the smart mold factory including time series and spatial coordinates; A spatial clustering module 32 is configured to perform spatial clustering processing on the energy consumption dataset using a spatial clustering algorithm to obtain energy consumption spatial distribution characteristics of the multiple production areas; A first generating module 33 is configured to sequentially calculate the energy consumption difference and the physical distance between two adjacent production areas in the multiple production areas based on the multiple energy consumption spatial distribution characteristics, and generate an energy consumption gradient quantization matrix including gradient values, spatial coordinate pairs, and gradient directions based on the multiple sets of energy consumption differences and physical distances, wherein the gradient value is a ratio of the energy consumption difference to the physical distance; a determination module 34 configured to determine a list of high energy consumption areas based on the energy consumption gradient quantization matrix, and to analyze the energy flow path guided by the gradient direction using a flow tracing algorithm based on the list of high energy consumption areas and the energy consumption dataset to obtain the position coordinates of abnormal energy consumption points, wherein the list of high energy consumption areas includes multiple high energy consumption areas, and the abnormal energy consumption points are gradient mutation points in the gradient direction; A second generating module 35 is configured to determine the energy consumption propagation influence range of the abnormal loss point according to the position coordinates and the energy consumption gradient quantization matrix, calculate power allocation schemes for the multiple production areas using a dynamic programming algorithm based on the energy consumption propagation influence range and the real-time energy consumption change data of the multiple production areas, and generate power upper limit adjustment data for the smart mold factory based on the multiple power allocation schemes; The updating module 36 is configured to update the power distribution of the plurality of production areas according to the power upper limit adjustment data through a real-time control algorithm to obtain an optimized energy consumption distribution state.

[0079] By means of the above-mentioned device, the energy consumption data of multiple production areas of the intelligent mold factory are obtained through the sensor network to generate an energy consumption data set containing time series and spatial coordinates; the energy consumption data set is spatially clustered by the spatial clustering algorithm to obtain the energy consumption spatial distribution characteristics of these multiple production areas; the energy consumption difference and physical distance of two adjacent production areas are calculated in turn according to these energy consumption spatial distribution characteristics, and an energy consumption gradient quantization matrix containing gradient values, spatial coordinate pairs and gradient directions is generated according to the calculation results, wherein the gradient value is the ratio of the energy consumption difference to the physical distance; then a list of high-energy consumption areas is determined according to the energy consumption gradient quantization matrix, and the energy flow path is analyzed by the flow tracking algorithm through the high-energy consumption area list and the energy consumption data set to obtain the position coordinates of the abnormal loss point, wherein the high-energy consumption area list includes multiple high-energy consumption areas, and the abnormal loss point is the gradient mutation point; the energy consumption propagation impact of the abnormal loss point is determined according to the position coordinates and the energy consumption gradient quantization matrix The impact range is determined, and the power allocation scheme is calculated through a dynamic programming algorithm based on the energy consumption propagation impact range and the real-time energy consumption change data of the production area. Then, the power upper limit adjustment data of the intelligent mold factory is generated according to multiple power allocation schemes; finally, according to the power upper limit adjustment data, the power allocation of these multiple production areas is updated through a real-time control algorithm to obtain the optimized energy consumption distribution state; using the above scheme, the present application realizes the precise analysis and optimal regulation of the energy consumption of the intelligent mold factory, effectively identifies abnormal loss points, and reasonably allocates power to each area to achieve the purpose of reducing overall energy consumption, providing technical support for energy conservation and emission reduction in the mold manufacturing industry; thereby solving the problem in related technologies that the current energy consumption monitoring of mold factories mostly relies on the data collection and total energy consumption statistics of a single device, and cannot effectively capture the energy consumption distribution characteristics of processing equipment in different areas, resulting in difficulty in accurately locating high-energy consumption areas and insufficient analysis of energy flow, thereby limiting the depth and real-time performance of energy consumption optimization.

[0080] In an exemplary embodiment, the spatial clustering module 32 is also used to perform spatial clustering processing on the energy consumption data set according to the spatial coordinates through the spatial clustering algorithm to obtain a regional division data set; calculate the energy consumption average value and standard deviation of each of the multiple production areas based on the regional division data set, and generate characteristic vectors of the multiple production areas based on the energy consumption average value and standard deviation, wherein the characteristic vector is used to indicate the spatial distribution law of the energy consumption data of the production area; when there is a first production area among the multiple production areas whose energy consumption average value is greater than a first preset threshold, the first production area is marked as an energy consumption abnormality area through an abnormal area identifier to obtain the energy consumption spatial distribution characteristics of the multiple production areas, wherein the energy consumption spatial distribution characteristics include the characteristic vector and the abnormal area identifier.

[0081] In an exemplary embodiment, the first generation module 33 is further used to perform missing value filling processing on the multiple energy consumption spatial distribution characteristics using a spatial interpolation algorithm according to the spatial coordinates of the energy consumption data set to fill the energy consumption data of the second production area and obtain multiple standardized energy consumption spatial distribution characteristics, wherein the intelligent mold factory includes the multiple production areas and the second production area; the energy consumption difference and physical distance of two adjacent production areas in the multiple production areas and the second production area are calculated in turn according to the average energy consumption in the multiple standardized energy consumption spatial distribution characteristics; the spatial coordinate pair is determined according to the neighborhood relationship between the multiple production areas and the second production area to obtain a data table containing the energy consumption difference, the physical distance and the spatial coordinate pair; when the energy consumption difference and the physical distance in the data table are both non-zero values, the gradient value is determined by calculating the ratio of the energy consumption difference and the physical distance, and the gradient direction is determined by the gradient calculation method according to the gradient value; the gradient value, the spatial coordinate pair and the gradient direction are converted into the energy consumption gradient quantization matrix through a matrix generation algorithm.

[0082] Optionally, the determination module 34 is further used to determine whether the gradient values ​​of the multiple production areas exceed a second preset threshold, and determine the production areas whose gradient values ​​exceed the second preset threshold as high-energy consumption areas to obtain a preliminary high-energy consumption area list; divide the multiple high-energy consumption areas in the preliminary high-energy consumption area list into regions through the spatial clustering algorithm to obtain multiple clustered high-energy consumption areas; obtain the regional boundaries and coordinate ranges of the multiple clustered high-energy consumption areas to obtain an optimized high-energy consumption area list; map the spatial coordinate pairs and gradient directions of the multiple high-energy consumption areas to the optimized high-energy consumption area list to obtain the high-energy consumption area list.

[0083] In an exemplary embodiment, the above-mentioned determination module 34 is also used to obtain the spatial coordinates and gradient directions of the multiple production areas from the energy consumption data set and the high-energy consumption area list, and use the flow tracking algorithm to calculate the energy flow path based on multiple groups of the spatial coordinates and the gradient directions to obtain a path data set, wherein the path data set includes path nodes and the gradient directions; sequentially determine whether the difference between the first gradient direction of multiple path nodes in the path data set and the second gradient direction of adjacent path nodes of the multiple path nodes exceeds a third preset threshold; determine the path node whose difference exceeds the third preset threshold as the gradient mutation point, and generate a preliminary abnormal loss point set based on the gradient mutation point and the spatial coordinates of the gradient mutation point; use the spatial clustering algorithm to divide the multiple gradient mutation points into regions based on the spatial coordinates of the multiple gradient mutation points in the preliminary abnormal loss point set, and obtain the regional boundaries and coordinate ranges after division, and determine the regional boundaries and the coordinate ranges as the position coordinates of the abnormal loss points.

[0084] Optionally, the second generation module 35 is used to perform regional segmentation on the energy consumption gradient quantization matrix through a grid partitioning algorithm and the position coordinates, generate multiple matrix partitioning data containing abnormal loss points, and determine the energy consumption distribution characteristics of the abnormal loss points based on the matrix partitioning data; analyze the gradient direction strength of multiple third production areas through a gradient calculation algorithm based on the matrix partitioning data, wherein the multiple third production areas correspond one-to-one to the multiple matrix partitioning data, and the gradient direction strength is used to indicate the energy change rate per unit distance; when it is determined that the gradient direction strength of a fourth production area among the multiple third production areas is greater than a fourth preset threshold, generate an initial flow path corresponding to the fourth production area through matrix operation, and obtain gradient direction data; determine the overlapping area based on the gradient direction data and the initial flow path through a path tracking algorithm, and perform a comprehensive analysis of the overlapping area and the energy consumption distribution of the overlapping area through data fusion technology to generate spatial overlapping feature data, and determine the propagation boundary based on the spatial overlapping feature data; perform boundary segmentation on the propagation boundary through a boundary partitioning algorithm to obtain boundary data, and determine the energy consumption propagation influence range based on the boundary data.

[0085] In an exemplary embodiment, the above-mentioned second generation module 35 is also used to determine the energy consumption propagation impact range and loss weight of the abnormal loss point, and obtain the energy consumption change data of the multiple production areas; calculate the multiple power allocation schemes according to the energy consumption change data and the loss weight through the dynamic programming algorithm; generate the power upper limit adjustment values ​​of the multiple production areas according to the power allocation scheme and the preset adjustment frequency; determine the power upper limit adjustment data based on the comparison results of multiple real-time energy consumption change data and multiple power upper limit adjustment values.

[0086] An embodiment of the present application further provides a storage medium, which includes a stored program, wherein the program executes any of the above methods when it is run.

[0087] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk, or an optical disk.

[0088] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail here.

[0089] An embodiment of the present application further provides an electronic device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0090] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0091] Optionally, specific examples in this embodiment may refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be described in detail here.

[0092] Obviously, those skilled in the art should understand that the modules or steps of the present application described above can be implemented using a general-purpose computing system, they can be concentrated on a single computing system, or distributed across a network composed of multiple computing systems. Alternatively, they can be implemented using program code executable by the computing system, so that they can be stored in a storage system and executed by the computing system. In some cases, the steps shown or described can be performed in a different order than herein, or they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0093] The above description is merely a preferred embodiment of the present application and is not intended to limit the present application. Various modifications and variations are possible for those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application shall be included within the scope of protection of the present application.

Claims

1. A method for monitoring energy consumption by zones in a smart mold factory, characterized in that: include: Acquire energy consumption data of multiple production areas of the smart mold factory through a sensor network, and generate an energy consumption dataset of the smart mold factory including time series and spatial coordinates; Performing spatial clustering processing on the energy consumption dataset using a spatial clustering algorithm to obtain energy consumption spatial distribution characteristics of the multiple production areas; sequentially calculating the energy consumption difference and the physical distance between two adjacent production areas in the multiple production areas according to the multiple energy consumption spatial distribution characteristics, and generating an energy consumption gradient quantization matrix including gradient values, spatial coordinate pairs, and gradient directions according to the multiple sets of the energy consumption difference and the physical distance, wherein the gradient value is a ratio of the energy consumption difference to the physical distance; Determine a list of high energy consumption areas according to the energy consumption gradient quantization matrix, and use the high energy consumption area list and the energy consumption dataset to analyze the energy flow path guided by the gradient direction using a flow tracking algorithm to obtain the position coordinates of abnormal energy consumption points, wherein the high energy consumption area list includes multiple high energy consumption areas, and the abnormal energy consumption points are gradient mutation points in the gradient direction; Determine the energy consumption propagation impact range of the abnormal loss point according to the position coordinates and the energy consumption gradient quantization matrix, calculate power allocation plans for the multiple production areas through a dynamic programming algorithm based on the energy consumption propagation impact range and the real-time energy consumption change data of the multiple production areas, and generate power upper limit adjustment data for the smart mold factory based on the multiple power allocation plans; According to the power upper limit adjustment data, the power distribution of the multiple production areas is updated through a real-time control algorithm to obtain an optimized energy consumption distribution state.

2. The energy consumption zoning monitoring method for an intelligent mold factory according to claim 1 is characterized in that: The energy consumption dataset is subjected to spatial clustering processing by a spatial clustering algorithm to obtain the energy consumption spatial distribution characteristics of the multiple production areas, including: Performing spatial clustering processing on the energy consumption dataset according to the spatial coordinates using the spatial clustering algorithm to obtain a regional division dataset; Calculating an average energy consumption value and a standard deviation of each of the multiple production areas based on the regional division data set, and generating feature vectors of the multiple production areas based on the average energy consumption value and the standard deviation, wherein the feature vectors are used to indicate a spatial distribution pattern of energy consumption data of the production areas; When there is a first production area among the multiple production areas whose average energy consumption value is greater than a first preset threshold, the first production area is marked as an energy consumption abnormal area through an abnormal area identification, and the energy consumption spatial distribution characteristics of the multiple production areas are obtained, wherein the energy consumption spatial distribution characteristics include the feature vector and the abnormal area identification.

3. The energy consumption zoning monitoring method for an intelligent mold factory according to claim 2 is characterized in that: The method includes: sequentially calculating the energy consumption difference and the physical distance between two adjacent production areas in the multiple production areas according to the multiple energy consumption spatial distribution characteristics, and generating an energy consumption gradient quantization matrix including gradient values, spatial coordinate pairs, and gradient directions according to multiple groups of the energy consumption difference and the physical distances. The method includes: Performing missing value filling processing on the plurality of energy consumption spatial distribution features using a spatial interpolation algorithm according to the spatial coordinates of the energy consumption dataset to fill the energy consumption data of the second production area, thereby obtaining a plurality of standardized energy consumption spatial distribution features, wherein the smart mold factory includes the plurality of production areas and the second production area; Calculating the energy consumption difference and the physical distance between two adjacent production areas in the multiple production areas and the second production area in sequence according to the average energy consumption values ​​in the multiple standardized energy consumption spatial distribution characteristics; Determine the spatial coordinate pairs according to the neighborhood relationship between the multiple production areas and the second production area, and obtain a data table including the energy consumption difference, the physical distance, and the spatial coordinate pairs; When both the energy consumption difference and the physical distance in the data table are non-zero values, determining the gradient value by calculating a ratio of the energy consumption difference to the physical distance, and determining the gradient direction according to the gradient value using a gradient calculation method; The gradient value, the spatial coordinate pair, and the gradient direction are converted into the energy consumption gradient quantization matrix through a matrix generation algorithm.

4. The energy consumption zoning monitoring method for an intelligent mold factory according to claim 1 is characterized in that: Determining a high energy consumption area list according to the energy consumption gradient quantization matrix includes: determining whether the gradient values ​​of the plurality of production areas exceed a second preset threshold, and determining the production areas whose gradient values ​​exceed the second preset threshold as high-energy consumption areas, to obtain a preliminary high-energy consumption area list; Dividing the multiple high-energy consumption areas in the preliminary high-energy consumption area list by the spatial clustering algorithm to obtain multiple clustered high-energy consumption areas; Obtaining the region boundaries and coordinate ranges of the plurality of clustered high-energy consumption regions to obtain an optimized high-energy consumption region list; The spatial coordinate pairs and gradient directions of the multiple high-energy consumption areas are mapped to the optimized high-energy consumption area list to obtain the high-energy consumption area list.

5. The energy consumption zoning monitoring method for an intelligent mold factory according to claim 4 is characterized in that: By using the high energy consumption area list and the energy consumption data set, a flow tracking algorithm is used to analyze the energy flow path guided by the gradient direction to obtain the position coordinates of the abnormal consumption point, including: Obtaining the spatial coordinates and gradient directions of the multiple production areas from the energy consumption dataset and the list of high-energy consumption areas, and using the flow tracing algorithm to calculate the energy flow path according to the multiple sets of spatial coordinates and the gradient directions to obtain a path dataset, wherein the path dataset includes path nodes and the gradient directions; determining in sequence whether a difference between first gradient directions of a plurality of path nodes in the path data set and second gradient directions of adjacent path nodes of the plurality of path nodes exceeds a third preset threshold; Determine the path node whose difference exceeds the third preset threshold as the gradient mutation point, and generate a preliminary abnormal loss point set according to the gradient mutation point and the spatial coordinates of the gradient mutation point; The spatial clustering algorithm is used to divide the multiple gradient mutation points into regions according to the spatial coordinates of the multiple gradient mutation points in the preliminary abnormal loss point set, and the regional boundaries and coordinate ranges after the division are obtained, and the regional boundaries and the coordinate ranges are determined as the position coordinates of the abnormal loss points.

6. The energy consumption zoning monitoring method for an intelligent mold factory according to claim 1 is characterized in that: Determining the energy consumption propagation influence range of the abnormal loss point according to the position coordinates and the energy consumption gradient quantization matrix includes: Performing regional segmentation on the energy consumption gradient quantization matrix using a grid partitioning algorithm and the position coordinates to generate a plurality of matrix partition data containing abnormal loss points, and determining energy consumption distribution characteristics of the abnormal loss points based on the matrix partition data; Analyzing the gradient direction strengths of the plurality of third production areas using a gradient calculation algorithm according to the matrix partition data, wherein the plurality of third production areas correspond one-to-one to the plurality of matrix partition data, and the gradient direction strength is used to indicate the rate of change of energy per unit distance; When it is determined that there is a fourth production area among the plurality of third production areas, and the gradient direction strength thereof is greater than a fourth preset threshold, generating an initial flow direction path corresponding to the fourth production area through matrix operation, and obtaining gradient direction data; Determining an overlapping area based on the gradient direction data and the initial flow path using a path tracing algorithm, performing a comprehensive analysis of the overlapping area and the energy consumption distribution of the overlapping area using a data fusion technique to generate spatial overlapping feature data, and determining a propagation boundary based on the spatial overlapping feature data; The propagation boundary is segmented by a boundary partitioning algorithm to obtain boundary data, and the energy consumption propagation influence range is determined according to the boundary data.

7. The energy consumption zoning monitoring method for an intelligent mold factory according to claim 6 is characterized in that: Calculating power allocation schemes for the multiple production areas using a dynamic programming algorithm based on the energy consumption propagation impact range and the real-time energy consumption change data of the multiple production areas, and generating power upper limit adjustment data for the smart mold factory based on the multiple power allocation schemes, including: Determine the energy consumption propagation impact range and loss weight of the abnormal loss point, and obtain energy consumption change data of the multiple production areas; Calculating a plurality of power allocation schemes according to the energy consumption change data and the loss weights by the dynamic programming algorithm; generating power upper limit adjustment values ​​for the plurality of production areas according to the power allocation scheme and a preset adjustment frequency; The power upper limit adjustment data is determined according to a comparison result of the plurality of the real-time energy consumption change data and the plurality of the power upper limit adjustment values.

8. An energy consumption zoning monitoring system for an intelligent mold factory, characterized in that: include: an acquisition module, configured to acquire energy consumption data of multiple production areas of the smart mold factory through a sensor network, and generate an energy consumption dataset of the smart mold factory including time series and spatial coordinates; A spatial clustering module is used to perform spatial clustering processing on the energy consumption dataset using a spatial clustering algorithm to obtain the energy consumption spatial distribution characteristics of the multiple production areas; A first generating module is configured to sequentially calculate the energy consumption difference and the physical distance between two adjacent production areas in the multiple production areas according to the multiple energy consumption spatial distribution characteristics, and generate an energy consumption gradient quantization matrix including a gradient value, a spatial coordinate pair, and a gradient direction according to the multiple sets of the energy consumption difference and the physical distance, wherein the gradient value is a ratio of the energy consumption difference to the physical distance; a determination module, configured to determine a list of high energy consumption areas based on the energy consumption gradient quantization matrix, and analyze the energy flow path guided by the gradient direction using a flow tracing algorithm based on the list of high energy consumption areas and the energy consumption dataset to obtain the position coordinates of abnormal energy consumption points, wherein the list of high energy consumption areas includes multiple high energy consumption areas, and the abnormal energy consumption points are gradient mutation points in the gradient direction; A second generation module is configured to determine the energy consumption propagation influence range of the abnormal loss point according to the position coordinates and the energy consumption gradient quantization matrix, calculate power allocation schemes for the multiple production areas through a dynamic programming algorithm based on the energy consumption propagation influence range and the real-time energy consumption change data of the multiple production areas, and generate power upper limit adjustment data for the smart mold factory based on the multiple power allocation schemes; An updating module is used to update the power distribution of the multiple production areas according to the power upper limit adjustment data through a real-time control algorithm to obtain an optimized energy consumption distribution state.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method described in any one of claims 1 to 7 when executed.

10. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 7 through the computer program.

Images