A method, device and medium for energy consumption prediction of a rubber production plant

Abstract

The embodiment of the application discloses a kind of rubber production factory energy consumption prediction method, equipment and medium, belong to energy consumption prediction technical field, solve the problem of rubber production factory energy consumption monitoring lag.To be measured rubber production factory thermal imaging data is fused to obtain the space heat data corresponding to the rubber production factory to be measured;The historical energy consumption data set is divided and compared, and the abnormal historical data is screened out to determine the reference energy consumption prediction data based on the historical data after screening;According to the building distribution data in the rubber production factory to be measured, the equipment distribution data in the rubber production factory to be measured and the space heat data, the energy consumption influence relationship diagram is constructed;Based on the preset energy consumption influence factor database and the energy consumption influence relationship diagram, the final energy consumption influence factor is determined;The reference energy consumption prediction data is adjusted based on the final energy consumption influence factor to obtain the final energy consumption prediction data.

Description

Technical Field

[0001] This application relates to the field of energy consumption prediction technology, and in particular to a method, equipment and medium for predicting energy consumption in a rubber production plant. Background Technology

[0002] With the acceleration of industrialization and the continuous expansion of production scale, the energy consumption of rubber production plants has been increasing year by year. This not only puts significant pressure on the environment but also greatly increases operating costs. Energy consumption forecasting for rubber production plants covers various energy sources used in the production process, and the use and consumption of these energy sources are directly related to the plant's operational efficiency and environmental impact. With the continuous advancement of modern information technology, Internet of Things (IoT) technology, and artificial intelligence (AI) technology, factory automation control systems and energy consumption monitoring systems have been widely applied in the rubber production field.

[0003] Although factory automation control systems and energy consumption monitoring systems have achieved significant energy-saving effects in rubber production plants, some challenges remain in practical applications. On the one hand, the complexity and diversity of energy data make it difficult to ensure its accuracy and reliability. On the other hand, because energy consumption in rubber production plants is influenced by various factors, adjustments are often made only when energy consumption reaches warning thresholds. This can lead to some production equipment operating near or at its limits, or frequent downtime for maintenance affecting equipment health, thus negatively impacting the stable operation of rubber production equipment. Summary of the Invention

[0004] This application provides a method, equipment, and medium for predicting energy consumption in a rubber production plant, which addresses the following technical problem: the energy consumption of a rubber production plant is only adjusted when it is in an alarm state, which affects the health of the equipment and consequently the stable operation of the energy-consuming equipment in the rubber production plant.

[0005] The embodiments of this application adopt the following technical solutions:

[0006] This application provides a method for predicting energy consumption in a rubber production plant. The method includes: acquiring thermal imaging data of the rubber production plant to be tested, and fusing the thermal imaging data to obtain spatial thermal data corresponding to the rubber production plant; acquiring historical energy consumption datasets corresponding to the rubber production plant to be tested, dividing and comparing the historical energy consumption datasets based on their data distribution, and filtering out abnormal historical data to determine reference energy consumption prediction data; constructing an energy consumption influence relationship diagram based on building distribution data, equipment distribution data, and spatial thermal data within the rubber production plant to be tested; determining energy consumption influence factors among different factors based on a pre-set energy consumption influence factor database and the energy consumption influence relationship diagram; processing the influence of the energy consumption influence factors based on the energy consumption influence relationship diagram and the energy consumption influence factors among different factors to determine the final energy consumption influence factor; and adjusting the reference energy consumption prediction data based on the final energy consumption influence factor to obtain the final energy consumption prediction data.

[0007] This application embodiment integrates thermal imaging data into a comprehensive dataset through fusion processing, improving data quality and reliability. Using filtered and organized historical data to determine reference energy consumption prediction data more accurately reflects the energy consumption characteristics and patterns of rubber production plants. Constructing an energy consumption impact diagram clearly shows the correlation between factors such as building distribution, equipment distribution, and heat transfer in the rubber production plant and energy consumption. By identifying the energy consumption impact factors among different factors, the acquired rubber production plant data can be processed through overlay and other methods, making the correlation between different influencing factors more accurate. By adjusting the reference energy consumption prediction data to take into account the impact of different factors, a more accurate and realistic final energy consumption prediction data is obtained, improving the accuracy of energy consumption prediction for rubber production plants.

[0008] In one implementation of this application, thermal imaging data of a rubber production plant under test is acquired, and the thermal imaging data of the rubber production plant under test is fused to obtain spatial thermal data corresponding to the rubber production plant under test. Specifically, this includes: acquiring a BIM model corresponding to the rubber production plant under test, and acquiring a thermal imaging image corresponding to the rubber production plant under test; determining a fusion mode; in the case of direct fusion, determining the geometric features of the rubber production plant under test in the BIM model, and determining the temperature change features in the thermal imaging image; matching the geometric features and the temperature change features to determine the spatial correspondence between the BIM model and the thermal imaging image; and based on the spatial correspondence, fusing the BIM model and the thermal imaging image into the same coordinate system.

[0009] In one implementation of this application, after determining the fusion mode, the method further includes: if the fusion mode is indirect fusion, determining an indirect model in the BIM model, and based on the geometric features of the indirect model, performing a first interval cropping on the thermal imaging image to obtain a first indirect image; matching the indirect model with the first indirect image, and performing a second interval cropping on the first indirect image based on the matching relationship to obtain a second indirect image; and fusing the indirect model and the second indirect image into the same coordinate system.

[0010] In one implementation of this application, a historical energy consumption dataset corresponding to the rubber production plant to be tested is obtained. Based on the data distribution of the historical energy consumption dataset, the historical energy consumption dataset is divided and compared to filter out abnormal historical data. Specifically, this includes: dividing the historical energy consumption dataset to obtain multiple energy consumption subsets and multiple composite subsets; identifying the normal energy consumption subsets and abnormal energy consumption subsets within the energy consumption subsets; calculating the similarity between the multiple composite subsets and the energy consumption subsets respectively, so as to classify the composite subsets that meet the similarity threshold into the normal energy consumption subsets or the abnormal energy consumption subsets; identifying the remaining datasets in the composite subsets; and based on the remaining datasets and the abnormal energy consumption subsets, identifying abnormal data in the historical energy consumption dataset and filtering out the abnormal historical data.

[0011] In one implementation of this application, reference energy consumption prediction data is determined based on filtered historical data. Specifically, this includes: obtaining personnel and equipment data corresponding to the rubber production plant under test based on the filtered historical data; determining historical personnel distribution data within the rubber production plant under test based on the personnel data, and determining equipment distribution data within the rubber production plant under test based on the equipment data; determining energy consumption prediction data to be adjusted based on spatial thermal data, historical personnel distribution data, equipment distribution data, and a pre-set energy consumption prediction model; obtaining current personnel distribution data corresponding to the rubber production plant under test based on the operating time information of the production equipment, and predicting the personnel distribution change rate within a specific future time period based on the difference between the current personnel distribution data and historical personnel distribution data; determining the energy consumption change rate of the rubber production plant equipment corresponding to different equipment distribution points based on the pre-set energy consumption change prediction model and the personnel distribution change rate; and adjusting the energy consumption prediction data to be adjusted for different equipment distribution points within a specific future time period based on the energy consumption change rate to obtain reference energy consumption prediction data. The reference energy consumption prediction data includes electricity consumption data and water consumption data.

[0012] In one implementation of this application, an energy consumption impact diagram is constructed based on building distribution data, equipment distribution data, and spatial thermal data within the rubber production plant to be tested. Specifically, this includes: determining building location data within the rubber production plant to be tested based on building distribution data, and determining a first node based on the building location data; determining a second node based on equipment distribution data within the rubber production plant to be tested; determining heat transfer data of the rubber production plant to be tested based on equipment distribution data and spatial thermal data; and constructing an energy consumption impact diagram based on the impact relationship between the first node and the second node, and the impact relationship between the heat transfer data of the rubber production plant to be tested and the second node.

[0013] In one implementation of this application, the energy consumption impact factors are processed based on the energy consumption impact relationship diagram and the energy consumption impact factors between different factors to determine the final energy consumption impact factor. Specifically, this includes: determining the direction of the connecting line based on the energy consumption impact relationship diagram, and determining the first energy consumption impact factor in a preset energy consumption impact factor database based on the node data corresponding to both ends of the connecting line; obtaining the first node corresponding to each second node, and determining the node impact degree based on the corresponding first node, and determining the second energy consumption impact factor based on the node impact degree; wherein, the node impact degree includes superimposed impact degree and repulsive impact degree; and determining the final energy consumption impact factor based on the preset energy consumption impact factor function, the first energy consumption impact factor, and the second energy consumption impact factor.

[0014] In one implementation of this application, based on a preset energy consumption impact factor function, specifically including:

[0015]

[0016] Where F is the final energy consumption impact factor corresponding to the current second node; i is the label of the first energy consumption impact factor corresponding to the current second node; p is the total number of first energy consumption impact factors corresponding to the current second node; W is the weight of the first energy consumption impact factor; Q is the first energy consumption impact factor; n is the current first node; m is the total number of first nodes that have an influence on the current second node; Z is the second energy consumption impact factor corresponding to the current first node; and T is the weight of the second energy consumption impact factor.

[0017] This application provides an energy consumption prediction device for a rubber production plant, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to: acquire thermal imaging data of the rubber production plant to be tested, and perform fusion processing on the thermal imaging data of the rubber production plant to be tested to obtain spatial thermal data corresponding to the rubber production plant to be tested; acquire historical energy consumption datasets corresponding to the rubber production plant to be tested, and divide and compare the historical energy consumption datasets based on the data distribution of the historical energy consumption datasets. Abnormal historical data is filtered out to determine reference energy consumption prediction data. An energy consumption impact diagram is constructed based on building distribution data, equipment distribution data, and spatial thermal data within the rubber production plant to be tested. Based on a pre-set energy consumption impact factor database and the energy consumption impact diagram, energy consumption impact factors among different factors are determined. Based on the energy consumption impact diagram and the energy consumption impact factors among different factors, the impact degree of the energy consumption impact factors is processed to determine the final energy consumption impact factors. The reference energy consumption prediction data is then adjusted based on the final energy consumption impact factors to obtain the final energy consumption prediction data.

[0018] This application provides a non-volatile computer storage medium storing computer-executable instructions. These instructions are configured to: acquire thermal imaging data of a rubber production plant under test, and perform fusion processing on the thermal imaging data to obtain spatial thermal data corresponding to the rubber production plant under test; acquire historical energy consumption datasets corresponding to the rubber production plant under test, and based on the data distribution of the historical energy consumption datasets, divide and compare the historical energy consumption datasets, filtering out abnormal historical data to determine reference energy consumption prediction data based on the filtered historical data; construct an energy consumption influence relationship diagram based on building distribution data, equipment distribution data, and spatial thermal data within the rubber production plant under test; determine energy consumption influence factors among different factors based on a pre-set energy consumption influence factor database and the energy consumption influence relationship diagram; process the influence of the energy consumption influence factors based on the energy consumption influence relationship diagram and the energy consumption influence factors among different factors to determine the final energy consumption influence factor; and adjust the reference energy consumption prediction data based on the final energy consumption influence factor to obtain the final energy consumption prediction data.

[0019] The above-mentioned technical solutions adopted in this application embodiment can achieve the following beneficial effects: This application embodiment integrates thermal imaging data into a comprehensive dataset through fusion processing, improving the quality and reliability of the data. Using filtered and organized historical data to determine reference energy consumption prediction data can more accurately reflect the energy consumption characteristics and patterns of rubber production plants. Constructing an energy consumption impact diagram clearly shows the correlation between factors such as personnel distribution, equipment distribution, and heat transfer in the rubber production plant and energy consumption. By determining the energy consumption impact factors among different factors, the acquired rubber production plant data can be processed through overlay and other methods, making the correlation between different influencing factors more accurate. By adjusting the reference energy consumption prediction data, the influence of different factors is taken into consideration, thereby obtaining more accurate and realistic final energy consumption prediction data, improving the accuracy of energy consumption prediction for rubber production plants. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0021] Figure 1 A flowchart of a method for predicting energy consumption in a rubber production plant, provided in an embodiment of this application;

[0022] Figure 2 This is a schematic diagram of the structure of an energy consumption prediction device for a rubber production plant, provided as an embodiment of this application.

[0023] Figure label:

[0024] 200: Energy consumption prediction equipment for rubber production plants; 201: Processor; 202: Memory. Detailed Implementation

[0025] This application provides a method, equipment, and medium for predicting energy consumption in a rubber production plant.

[0026] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0027] The technical solutions proposed in the embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0028] Figure 1 A flowchart of a method for predicting energy consumption in a rubber production plant, as provided in this application embodiment, is shown below. Figure 1 As shown, the method for predicting energy consumption in rubber production plants includes the following steps:

[0029] Step 101: Obtain thermal imaging data of the rubber production plant to be tested, and perform fusion processing on the thermal imaging data of the rubber production plant to be tested to obtain the spatial thermal data corresponding to the rubber production plant to be tested.

[0030] In one embodiment of this application, a BIM model of the rubber production plant to be tested and a corresponding thermal imaging image of the rubber production plant to be tested are acquired. A fusion mode is determined. If the fusion mode is direct fusion, the geometric features of the rubber production plant to be tested are determined in the BIM model, and the temperature change features are determined in the thermal imaging image. The geometric features and temperature change features are matched to determine the spatial correspondence between the BIM model and the thermal imaging image. Based on the spatial correspondence, the BIM model and the thermal imaging image are fused into the same coordinate system.

[0031] Specifically, firstly, a BIM (Building Information Modeling) model of the rubber production plant to be tested is acquired. Simultaneously, thermal images of the plant are captured using an infrared thermal imager. The BIM model contains the geometric, physical, and functional characteristics of the rubber production plant, while the thermal images display the temperature distribution on the plant's surfaces. In direct fusion mode, the geometric features of the rubber production plant, such as walls, windows, doors, and roofs, can be identified within the BIM model. These features have clearly defined dimensions and locations. In the thermal images, temperature variations are represented by color changes in different areas. Areas with higher temperatures appear as brighter colors, while areas with lower temperatures appear as darker colors. By identifying these temperature variation features, their corresponding locations and extents are determined.

[0032] Furthermore, key geometric features, such as door and window locations and wall corners, are extracted from the BIM model, and temperature change features are extracted from the thermal imaging images. Considering factors such as the location, shape, and size of the features, the geometric features in the BIM model are matched with the temperature change features in the thermal imaging images. Through feature matching, the spatial correspondence between the BIM model and the thermal imaging images is determined. If the coordinate systems of the BIM model and the thermal imaging images are different, coordinate system transformation is required to make them have the same coordinate system.

[0033] Specifically, based on information such as the location, shape, and size of features, image registration or feature point matching algorithms are used for initial matching. Based on the initial matching results, optimization algorithms, such as least squares, are used to optimize the matching results, eliminating erroneous matching points and improving matching accuracy. Based on the optimized matching results, each element in the BIM model, such as rooms, doors, windows, and equipment, is associated with the corresponding temperature region in the thermal imaging image.

[0034] In one embodiment of this application, when the fusion mode is indirect fusion, an indirect model is determined in the BIM model. Based on the geometric features of the indirect model, a first interval is cropped from the thermal imaging image to obtain a first indirect image. The indirect model is matched with the first indirect image, and a second interval is cropped from the first indirect image based on the matching relationship to obtain a second indirect image. The indirect model and the second indirect image are then fused into the same coordinate system.

[0035] Specifically, one or more indirect models are determined based on specific needs or analysis objectives. These indirect models can be a part, a functional area, or a specific geometric feature in the BIM model. When the fusion mode is indirect fusion, the region in the thermal imaging image corresponding to the geometric features of the indirect model is identified. Based on the boundaries or feature points of the indirect model, the thermal imaging image is cropped or masked to obtain the first indirect image. For example, if the indirect model is a workshop in the rubber production plant to be tested, the corresponding region is found in the thermal imaging image based on the workshop's geometric dimensions and location information, and extracted as the first indirect image.

[0036] Furthermore, using image registration or feature point matching algorithms, the geometric features of the indirect model are mapped to the features in the first indirect image. Based on the matching results, the first indirect image is adjusted as necessary, such as rotated, scaled, or translated, to ensure a perfect match with the indirect model. Further, image processing techniques are used to crop a region from the first indirect image that meets certain conditions, resulting in a second indirect image. For example, if only a specific area within the workshop (such as near a window) is of interest, the boundary of that area can be defined and cropped from the first indirect image to obtain the second indirect image. Finally, the indirect model and the second indirect image are fused into the same coordinate system and displayed on a map.

[0037] Step 102: Obtain the historical energy consumption dataset corresponding to the rubber production plant to be tested. Based on the data distribution of the historical energy consumption dataset, divide and compare the historical energy consumption dataset, filter out abnormal historical data, and determine the reference energy consumption prediction data based on the filtered historical data.

[0038] In one embodiment of this application, the historical energy consumption dataset is divided into multiple energy consumption subsets and multiple composite subsets. Normal energy consumption subsets and abnormal energy consumption subsets are identified within the energy consumption subsets. The multiple composite subsets are then compared with the energy consumption subsets to calculate similarity, and composite subsets meeting a similarity threshold are assigned to either the normal energy consumption subset or the abnormal energy consumption subset. The remaining datasets within the composite subsets are then identified. Based on the remaining datasets and the abnormal energy consumption subsets, abnormal data in the historical energy consumption dataset is identified and filtered out.

[0039] Specifically, the historical energy consumption dataset is logically divided into multiple energy consumption subsets based on factors such as time, equipment type, and geographical location. Simultaneously, composite subsets are identified based on other energy consumption-related factors within the historical dataset, such as environmental parameters and equipment status. For example, an energy consumption subset could be a subset of energy consumption data divided by month from January to December, and a composite subset could be a dataset containing composite information such as temperature, humidity, and equipment operating time.

[0040] Furthermore, statistical methods, such as mean, standard deviation, and median, are used to identify normal and abnormal energy consumption from the energy consumption subset. For example, for each month's energy consumption subset, its mean and standard deviation are calculated. If a data point's value exceeds the mean plus twice the standard deviation, it is considered abnormal. The composite subset is then compared to the energy consumption subset using methods such as cosine similarity. Based on the similarity calculation results, the composite subset is classified into normal or abnormal energy consumption subsets. For example, for each composite subset, its similarity to the normal energy consumption subset is calculated. If the similarity is higher than a preset threshold, the composite subset is classified as normal; otherwise, it is classified as abnormal.

[0041] Furthermore, after classifying the composite class subsets, the remaining composite class subsets are obtained. Anomalous data in the historical energy consumption dataset is determined based on the association between these remaining datasets and the anomalous energy consumption subsets. If a remaining dataset is highly correlated with multiple anomalous energy consumption subsets, it is considered anomalous data, thus filtering out anomalous data from the historical energy consumption dataset.

[0042] In one embodiment of this application, personnel and equipment data corresponding to the rubber production plant under test are obtained based on filtered historical data. Historical personnel distribution data within the rubber production plant under test is determined based on the personnel data, and equipment distribution data is determined based on the equipment data. Adjustable energy consumption prediction data is determined based on spatial thermal data, historical personnel distribution data, equipment distribution data, and a pre-set energy consumption prediction model. Current personnel distribution data corresponding to the rubber production plant under test is obtained based on the operating time information of the production equipment. The rate of change in personnel distribution within a specific future time period is predicted based on the difference between the current and historical personnel distribution data. The energy consumption change rate of the rubber production plant equipment corresponding to different equipment distribution points is determined based on the pre-set energy consumption change prediction model and the rate of change in personnel distribution. Based on the energy consumption change rate, the adjustable energy consumption prediction data corresponding to different equipment distribution points within a specific future time period is adjusted to obtain reference energy consumption prediction data; wherein, the reference energy consumption prediction data includes electricity consumption data and water consumption data.

[0043] Specifically, valid and relevant personnel and equipment data are selected from the historical database of the rubber production plant under test. This includes detailed data on the daily number of employees, job distribution, and equipment operating time and power over the past year. Based on the selected personnel data, the flow and distribution patterns of personnel in different time periods (such as weekdays, weekends, and holidays) are analyzed to form historical personnel distribution data. Simultaneously, based on equipment data, the spatial distribution and operating status of equipment are determined. For example, during weekdays, from 8:00 AM to 10:00 AM, employees are mainly concentrated in production line A and the office area, while equipment is mainly distributed in production line A and the warehouse. Combining spatial thermal data, such as environmental parameters like temperature and humidity, historical personnel distribution data, equipment distribution data, and the established energy consumption prediction model, the energy consumption of the plant in different time periods is predicted, and the energy consumption prediction data that needs adjustment is determined.

[0044] Furthermore, this application embodiment includes an energy consumption prediction model. The training process of this model is as follows: spatial thermal data samples, historical personnel distribution data samples, and equipment distribution samples are used as inputs, and the factory energy consumption data corresponding to the input samples is used as the output to train a pre-set neural network model to obtain the energy consumption prediction model. The spatial thermal data, personnel distribution data, and equipment data corresponding to the rubber production plant to be tested are input into the energy consumption prediction model to determine the energy consumption prediction data to be adjusted. For example, using the pre-set energy consumption prediction model, it can be predicted that during the period from 8:00 AM to 10:00 AM on weekdays, due to the concentration of employees and equipment, the factory's electricity and water consumption will increase significantly.

[0045] Furthermore, based on production equipment runtime information and current personnel distribution data, and compared with historical personnel distribution data, the rate of change in personnel distribution within a specific future time period can be predicted. For example, production equipment runtime information indicates that production line A needs to be stopped for maintenance during the current time period, while production line B begins operation. Current data shows that employees are starting to shift from production line A to production line B. Machine learning models can be used to predict that the number of employees on production line B will increase within the next week, while the number on production line A will decrease accordingly. Using a pre-set energy consumption change prediction model combined with the predicted rate of change in personnel distribution, the energy consumption change rate corresponding to different equipment distribution points can be calculated; that is, the energy consumption for production line A will decrease, while the energy consumption for production line B will increase.

[0046] The training process of the pre-set energy consumption change prediction model is as follows: Equipment information samples and personnel distribution change rate samples from different distribution points are used as inputs, and the energy consumption change rate samples corresponding to each distribution point are used as outputs to train the pre-set neural network model to obtain the pre-set energy consumption change prediction model. By inputting the current personnel distribution change rate and equipment information from the distribution points into the pre-set energy consumption change prediction model, the energy consumption change rate of the rubber production plant equipment corresponding to different distribution points can be obtained.

[0047] Furthermore, the obtained energy consumption forecast data to be adjusted is modified based on the energy consumption change rate to obtain more accurate reference energy consumption forecast data, including electricity and water consumption. Specifically, by multiplying the predicted energy consumption change rate with the current energy consumption value, the reference energy consumption forecast value corresponding to each equipment distribution point in the future time period can be obtained. For example, if the personnel density at a certain equipment distribution point decreases, the energy consumption at that equipment distribution point will decrease; the greater the decrease in personnel density, the greater the decrease in energy consumption.

[0048] Step 103: Based on the building distribution data, equipment distribution data and spatial thermal data within the rubber production plant to be tested, construct an energy consumption impact diagram.

[0049] In one embodiment of this application, building location data within the rubber production plant to be tested is determined based on building distribution data, and a first node is determined based on the building location data. A second node is determined based on equipment distribution data within the rubber production plant to be tested. Furthermore, heat transfer data of the rubber production plant to be tested is determined based on the equipment distribution data and spatial thermal data within the rubber production plant to be tested. An energy consumption impact diagram is constructed based on the influence relationship between the first node and the second node, and the influence relationship between the heat transfer data of the rubber production plant to be tested and the second node.

[0050] Specifically, based on the spatial thermal data of the rubber production plant under test, the exact locations of each building within the plant and the corresponding thermal data for each location can be determined. This location data typically includes information such as the building's coordinates, area, and height. Based on the collected data, the specific locations of each building within the plant and their spatial relationships are determined. Multiple first nodes are then identified based on the building locations. Assume the rubber production plant under test has three main production workshops, workshop A, workshop B, and workshop C, and one office. Through building distribution data, the exact locations of these buildings can be determined, and each workshop and the office are designated as first nodes.

[0051] Furthermore, based on equipment distribution data from the rubber production plant, the specific location and energy consumption characteristics of each piece of equipment are determined. Then, based on factors such as the equipment's energy consumption level and its contribution to overall energy consumption, several key pieces of equipment are selected as second nodes. For example, in workshop A, there are multiple injection molding machines and vulcanizing machines. Using equipment distribution data, the locations of these machines can be determined, and each injection molding machine and vulcanizing machine can be designated as a second node.

[0052] Furthermore, based on the equipment distribution data and spatial thermal data within the rubber production plant under test, the heat transfer situation inside the plant is analyzed. This includes the heat generated by the equipment and the conduction and barrier effects of the building structure on heat. Specifically, the equipment types and data corresponding to different workshops are obtained, the heat generated in each workshop is acquired, and the heat transfer coefficient is determined based on the building structure of each workshop, such as wall height and wall material type. Through calculation and analysis, the heat transfer data of the rubber production plant can be obtained. For example, in workshop A, the injection molding machine generates a large amount of heat during operation. Simultaneously, the building structure of the workshop, such as walls and roof, has a certain influence on heat conduction. Through calculation and analysis, the heat transfer data of workshop A, such as heat transfer rate and temperature distribution, can be obtained.

[0053] Furthermore, based on the influence relationship between the first and second nodes—that is, the energy consumption influence relationship between different building areas and equipment—and the influence relationship between the heat transfer data of the rubber production plant and the second node—that is, the heating effect of the heat generated by the equipment on the surrounding environment—an energy consumption influence relationship diagram is constructed to show the energy transfer and influence paths between each node. For example, workshop A (the first node) and the injection molding machine (the second node) can be connected with arrows, indicating that the energy consumption of the injection molding machine has a direct impact on the energy consumption of workshop A. Simultaneously, the heat generated by the injection molding machine can be connected with the heat transfer data of workshop A with arrows, indicating that the heat generated by the injection molding machine has a direct impact on the temperature distribution of workshop A.

[0054] Step 104: Based on the pre-set energy consumption impact factor database and the energy consumption impact relationship diagram, determine the energy consumption impact factors among different factors.

[0055] In one embodiment of this application, a preset energy consumption influencing factor database is provided. This database includes different energy consumption influencing factors, as well as energy consumption influencing factors corresponding to each of these factors. Based on the energy consumption influencing relationship diagram in this embodiment, different factors with influencing relationships are identified, thereby determining the energy consumption influencing factors between different factors in the preset energy consumption influencing factor database.

[0056] Step 105: Based on the energy consumption impact relationship diagram and the energy consumption impact factors between different factors, process the impact degree of the energy consumption impact factors to determine the final energy consumption impact factors.

[0057] In one embodiment of this application, the direction of the connecting line is determined based on the energy consumption impact diagram, and a first energy consumption impact factor is determined in a preset energy consumption impact factor database based on the node data corresponding to both ends of the connecting line. The first node corresponding to each second node is obtained, and the node influence degree is determined based on the corresponding first node. A second energy consumption impact factor is then determined based on the node influence degree; wherein, the node influence degree includes superimposed influence degree and repulsive influence degree. The final energy consumption impact factor is determined based on the preset energy consumption impact factor function, the first energy consumption impact factor, and the second energy consumption impact factor.

[0058] Specifically, based on the energy consumption impact diagram, the direction of the connecting lines between each node is determined. Based on the node data at both ends of the connecting lines, the first energy consumption impact factor is searched and determined in the database. The first nodes corresponding to each second node are obtained, and the node impact degree is determined based on the corresponding first nodes. The node impact degree includes superimposed impact degree and repulsive impact degree. Superimposed impact degree indicates that an increase in the energy consumption of one node will lead to an increase in the energy consumption of another node; repulsive impact degree indicates that an increase in the energy consumption of one node will lead to a decrease in the energy consumption of another node. Multiple first nodes corresponding to each second node are determined, and the energy consumption impact factor of each determined first node for that first node is obtained. If it is a superimposed impact degree, the multiple energy consumption impact factors corresponding to that first node are superimposed; if there is a repulsive impact degree, the energy consumption impact factor corresponding to the repulsive impact degree is subtracted.

[0059] In one embodiment of this application, the preset energy consumption impact factor function specifically includes:

[0060]

[0061] Where F is the final energy consumption impact factor corresponding to the current second node; i is the label of the first energy consumption impact factor corresponding to the current second node; p is the total number of first energy consumption impact factors corresponding to the current second node; W is the weight of the first energy consumption impact factor; Q is the first energy consumption impact factor; n is the current first node; m is the total number of first nodes that have an influence on the current second node; Z is the second energy consumption impact factor corresponding to the current first node; and T is the weight of the second energy consumption impact factor.

[0062] Furthermore, based on the preset energy consumption impact factor function, the first energy consumption impact factor, and the second energy consumption impact factor, the final energy consumption impact factor is determined.

[0063] Step 106: Adjust the reference energy consumption prediction data based on the final energy consumption impact factor to obtain the final energy consumption prediction data.

[0064] Specifically, based on the obtained final energy consumption impact factor, the energy consumption of multiple second nodes in the rubber production plant under test is recalculated. For example, the reference energy consumption prediction data corresponding to each second node can be multiplied with the corresponding final energy consumption impact factor to adjust the reference energy consumption prediction data and obtain the final energy consumption prediction data.

[0065] Figure 2 This is a schematic diagram of the structure of an energy consumption prediction device for a rubber production plant, provided as an embodiment of this application. Figure 2 As shown, a rubber production plant energy consumption prediction device 200 includes: at least one processor 201; and a memory 202 communicatively connected to the at least one processor 201. The memory 202 stores instructions executable by the at least one processor 201. These instructions, when executed by the at least one processor 201, enable the at least one processor 201 to: acquire thermal imaging data of the rubber production plant under test, and perform fusion processing on the thermal imaging data to obtain spatial thermal data corresponding to the rubber production plant under test; acquire historical energy consumption datasets corresponding to the rubber production plant under test, and, based on the data distribution of the historical energy consumption datasets, process the historical energy consumption data... The data set is divided and compared to filter out abnormal historical data, and reference energy consumption prediction data is determined based on the filtered historical data. An energy consumption impact diagram is constructed based on the building distribution data, equipment distribution data, and spatial thermal data within the rubber production plant to be tested. Based on a pre-set energy consumption impact factor database and the energy consumption impact diagram, energy consumption impact factors among different factors are determined. Based on the energy consumption impact diagram and the energy consumption impact factors among different factors, the impact degree of the energy consumption impact factors is processed to determine the final energy consumption impact factors. The reference energy consumption prediction data is adjusted based on the final energy consumption impact factors to obtain the final energy consumption prediction data.

[0066] This application provides a non-volatile computer storage medium storing computer-executable instructions. These instructions are configured to: acquire thermal imaging data of a rubber production plant under test, and perform fusion processing on the thermal imaging data to obtain spatial thermal data corresponding to the rubber production plant under test; acquire historical energy consumption datasets corresponding to the rubber production plant under test, and based on the data distribution of the historical energy consumption datasets, divide and compare the historical energy consumption datasets, filtering out abnormal historical data to determine reference energy consumption prediction data based on the filtered historical data; construct an energy consumption influence relationship diagram based on building distribution data, equipment distribution data, and spatial thermal data within the rubber production plant under test; determine energy consumption influence factors among different factors based on a pre-set energy consumption influence factor database and the energy consumption influence relationship diagram; process the influence of the energy consumption influence factors based on the energy consumption influence relationship diagram and the energy consumption influence factors among different factors to determine the final energy consumption influence factor; and adjust the reference energy consumption prediction data based on the final energy consumption influence factor to obtain the final energy consumption prediction data.

[0067] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0068] The above descriptions are merely embodiments of this application and are not intended to limit the scope of this application. For those skilled in the art, various modifications and variations can be made to the embodiments of this application. These modifications or substitutions do not cause the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions in the embodiments of this application.

Claims

1. A method of predicting energy consumption of a rubber production plant, characterized by, The method comprises: acquiring thermal imaging data of a rubber production factory to be tested, and performing fusion processing on the thermal imaging data of the rubber production factory to be tested to obtain spatial thermal data corresponding to the rubber production factory to be tested; acquiring a historical energy consumption data set corresponding to the rubber production factory to be tested, dividing and comparing the historical energy consumption data set based on data distribution of the historical energy consumption data set, screening out abnormal historical data, and determining reference energy consumption prediction data based on the screened historical data; constructing an energy consumption influence relationship graph based on building distribution data in the rubber production factory to be tested, equipment distribution data in the rubber production factory to be tested, and the spatial thermal data; determining energy consumption influence factors between different factors based on a preconfigured energy consumption influence factor database and the energy consumption influence relationship graph; performing influence degree processing on the energy consumption influence factors based on the energy consumption influence relationship graph and the energy consumption influence factors between different factors to determine final energy consumption influence factors; adjusting the reference energy consumption prediction data based on the final energy consumption influence factors to obtain final energy consumption prediction data.

2. The rubber production plant energy consumption prediction method according to claim 1, characterized in that, The method comprises: acquiring a BIM model corresponding to the rubber production factory to be tested, and acquiring a thermal imaging image corresponding to the rubber production factory to be tested; determining a fusion mode, determining geometric features of the rubber production factory to be tested in the BIM model, and determining temperature change features in the thermal imaging image in a case where the fusion mode is direct fusion; matching the geometric features and the temperature change features to determine a spatial correspondence relationship between the BIM model and the thermal imaging image; fusing the BIM model and the thermal imaging image into the same coordinate system based on the spatial correspondence relationship.

3. The rubber production plant energy consumption prediction method according to claim 2, characterized in that, After the fusion mode is determined, the method further comprises: in a case where the fusion mode is indirect fusion, determining an indirect model in the BIM model, performing first interval cutting on the thermal imaging image based on geometric features of the indirect model to obtain a first indirect image; the indirect model is a certain part, a certain functional area, or a certain geometric feature area determined in the BIM model based on a requirement or an analysis target; matching the indirect model and the first indirect image, performing second interval cutting on the first indirect image based on a matching relationship to obtain a second indirect image; fusing the indirect model and the second indirect image into the same coordinate system.

4. The rubber production plant energy consumption prediction method according to claim 1, characterized in that, The method comprises: dividing the historical energy consumption data set to obtain a plurality of energy consumption sub-data sets and a plurality of composite class sub-data sets; The normal energy consumption sub-data set and the abnormal energy consumption sub-data set are determined from the energy consumption sub-data set, similarity calculation is performed between the plurality of composite class sub-data sets and the energy consumption sub-data set, and the composite class sub-data set meeting the similarity threshold is divided into the normal energy consumption sub-data set or the abnormal energy consumption sub-data set; The remaining data set is determined from the composite class sub-data set, and the abnormal data is determined from the remaining data set and the abnormal energy consumption sub-data set, and the abnormal data is screened out; the composite class sub-data set is a data set containing temperature, humidity, and device running time.

5. The rubber production plant energy consumption prediction method according to claim 1, characterized in that, The reference energy consumption prediction data is determined based on the screened historical data, and specifically includes: Based on the screened historical data, the personnel data and the equipment data corresponding to the rubber production factory to be tested are obtained; Based on the personnel data, the historical personnel distribution data in the rubber production factory to be tested is determined, and based on the equipment data, the equipment distribution data in the rubber production factory to be tested is determined; Based on the space heat data, the historical personnel distribution data, the equipment distribution data, and the preset energy consumption prediction model, the to-be-adjusted energy consumption prediction data is determined; Based on the production equipment running time period information, the current personnel distribution data corresponding to the rubber production factory to be tested is obtained, and based on the difference between the current personnel distribution data and the historical personnel distribution data, the personnel distribution change rate in the future time period is predicted; Based on the preset energy consumption prediction model and the personnel distribution change rate, the energy consumption change rate of the rubber production factory equipment corresponding to different equipment distribution points is determined; Based on the energy consumption change rate, the to-be-adjusted energy consumption prediction data corresponding to different equipment distribution points in the future time period is adjusted to obtain the reference energy consumption prediction data; The reference energy consumption prediction data includes power consumption data and water consumption data.

6. The rubber production plant energy consumption prediction method according to claim 1, characterized in that, The energy consumption influence relationship diagram is constructed according to the building distribution data in the rubber production factory to be tested, the equipment distribution data in the rubber production factory to be tested, and the space heat data, and specifically includes: Based on the building distribution data, the building position data in the rubber production factory to be tested is determined, and the first node is determined according to the building position data; Based on the equipment distribution data in the rubber production factory to be tested, the second node is determined; And based on the equipment distribution data in the rubber production factory to be tested and the space heat data, the heat transfer data of the rubber production factory to be tested is determined; The energy consumption influence relationship diagram is constructed based on the influence relationship between the first node and the second node, and the influence relationship between the heat transfer data of the rubber production factory to be tested and the second node.

7. The method for predicting energy consumption in a rubber production plant according to claim 6, characterized in that, The energy consumption influence factor between the energy consumption influence relationship diagram and the different factors is processed to determine the final energy consumption influence factor, and specifically includes: Based on the energy consumption influence relationship diagram, a connection line direction is determined, and based on corresponding node data at two ends of the connection line, a first energy consumption influence factor is determined in a preset energy consumption influence factor database; Corresponding first nodes of each second node are acquired, and based on the corresponding first nodes, a node influence degree is determined, and based on the node influence degree, a second energy consumption influence factor is determined; wherein the node influence degree includes superposition influence degree and repulsion influence degree; wherein the superposition influence degree indicates that energy consumption increase of one node leads to energy consumption increase of another node, and the repulsion influence degree indicates that energy consumption increase of one node leads to energy consumption decrease of another node; Based on a preset energy consumption influence factor function, the first energy consumption influence factor, and the second energy consumption influence factor, a final energy consumption influence factor is determined.

8. The rubber production plant energy consumption prediction method according to claim 7, characterized in that, The preset energy consumption influence factor function specifically includes: ； wherein, F is a final energy consumption impact factor corresponding to the current second node; i is an index of a first energy consumption impact factor corresponding to the current second node; p is a total number of first energy consumption impact factors corresponding to the current second node; W is a weight corresponding to the first energy consumption impact factor; Q is the first energy consumption impact factor; n is a current first node; m is a total number of first nodes having an influence degree with the current second node; Z is a second energy consumption impact factor corresponding to the current first node; T is a weight corresponding to the second energy consumption impact factor.

9. A rubber production plant energy consumption prediction device characterized by, The device includes a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, the device is triggered to execute the method of any one of claims 1-8.

10. A non-transitory computer storage medium storing computer-executable instructions, the computer-executable instructions comprising instructions for: receiving a request to access a file; determining whether the file is stored in a cache; and in response to determining that the file is stored in the cache, providing access to the file from the cache. The computer executable instructions can execute the method of any one of claims 1-8.

Images