A method for intelligent monitoring of unit consumption of a gypsum board production line
By establishing a correlation topology between auxiliary materials and energy consumption points in the gypsum board production line, locating monitoring points and constructing anomaly factor topology, the problem of lack of real-time monitoring and anomaly location in existing technologies is solved, and efficient data monitoring and anomaly investigation are achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- TAICANG BEIXIN BUILDING MATERIALS CO LTD
- Filing Date
- 2022-11-18
- Publication Date
- 2026-06-23
AI Technical Summary
The existing gypsum board production lines lack real-time monitoring and guidance, making it difficult to fully locate production anomalies, resulting in untimely data statistics and cumbersome anomaly detection.
By establishing a correlation topology between auxiliary materials and energy consumption points, the monitoring points are located using node criticality, and an anomaly factor topology is constructed to monitor and locate anomalies in real time.
It enables real-time data monitoring of gypsum board production lines, reduces secondary construction costs, simplifies anomaly troubleshooting, and improves data processing efficiency and accuracy.
Smart Images
Figure CN116382137B_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of gypsum board production, in particular to a single consumption intelligent monitoring method for a gypsum board production line. BACKGROUND
[0002] Gypsum board production mainly uses various auxiliary materials, water, electricity and gas, etc. The various auxiliary materials include desulfurized gypsum, face paper and various low-value consumables. The consumption of various materials and energy constitutes the single consumption of each square meter of gypsum board. Real-time supervision of single consumption can effectively save costs and scientifically achieve "increasing efficiency and reducing costs".
[0003] Currently, the production data of the gypsum board production line are filled in by the production line foremen. The consumption of raw materials, total electricity, gas (coal) and water in the month are manually counted. Although the automatic electric control system of each section can count the consumption in a day, it cannot collect and summarize information and automatically form a report. Therefore, the statistics are not timely, and effective real-time supervision and guidance of production cannot be achieved. In addition, it is difficult to fully locate the production abnormal points, and the abnormal finding process is complicated. SUMMARY
[0004] The purpose of the present application is to provide a single consumption intelligent monitoring method for a gypsum board production line to solve the technical problem that the prior art cannot form effective real-time supervision and guidance of production and cannot fully locate the production abnormal points.
[0005] To solve the above technical problems, the present application specifically provides the following technical solutions:
[0006] A single consumption intelligent monitoring method for a gypsum board production line, comprising the following steps:
[0007] Step S1, obtaining all auxiliary material use points and energy consumption use points in the gypsum board production line, and sequentially associating and topologically establishing each auxiliary material use point and each energy consumption use point to obtain a point association topology of each auxiliary material and a point association topology of each energy consumption. In the point association topology, the monitoring points of each auxiliary material and the monitoring points of each energy consumption are located by node keyness. The energy consumption use point represents the position of the device using energy consumption, and the auxiliary material use point represents the position of the device using auxiliary materials.
[0008] Step S2, extracting all devices having topological association with each monitoring point from the device association topology formed by the devices in the gypsum board production line, and constructing an abnormal factor topology from each monitoring point and all devices having topological association with each monitoring point.
[0009] Step S3, monitoring the auxiliary material usage amount features and the energy consumption amount features at the monitoring points of each auxiliary material usage point and each energy consumption usage point in real time, and locating the auxiliary material usage abnormal points and the energy consumption usage abnormal points in the abnormal factor topology according to the auxiliary material usage amount features and the energy consumption amount features.
[0010] As a preferred scheme of the present application, the establishment of the point position correlation topology of each auxiliary material comprises:
[0011] obtaining the material usage amount time sequence of each usage point of each auxiliary material within a time sequence as a material usage amount time sequence feature, and sequentially calculating the point position similarity between any two usage points of each auxiliary material based on the material usage amount time sequence feature, and the calculation formula of the point position similarity of the auxiliary material is:
[0012]
[0013] In the formula, I ijk is the point position similarity of the jth usage point and the kth usage point of the ith auxiliary material, S ij is the material usage amount time sequence feature of the jth usage point of the ith auxiliary material, S ik is the material usage amount time sequence feature of the kth usage point of the ith auxiliary material, ||S ij -S ik || 2 is the Euclidean distance of S ij and S ik , i, j, k are measurement numbers;
[0014] using each usage point of each auxiliary material as a topology node, setting a topology edge between the two topology nodes corresponding to the two usage points with a point position similarity higher than a similarity threshold value for topology connection, and using the point position similarity corresponding to the two topology edges as the edge weight of the topology edge, so as to construct the point position correlation topology of each auxiliary material by the topology node, the topology edge and the edge weight.
[0015] As a preferred scheme of the present application, the establishment of the point position correlation topology of each energy consumption comprises:
[0016] obtaining the energy consumption amount time sequence of each usage point of each energy consumption within a time sequence as an energy consumption amount time sequence feature, and sequentially calculating the point position similarity between any two usage points of each energy consumption based on the energy consumption amount time sequence feature, and the calculation formula of the point position similarity of the energy consumption is:
[0017]
[0018] In the formula, J ijkR represents the similarity between the j-th and k-th usage points of the i-th energy consumption type. ij Let R be the time-series characteristic of energy consumption at the j-th usage point of the i-th type of energy consumption. ik Let ||R| represent the time-series characteristics of energy consumption at the k-th usage point of the i-th type of energy consumption. ij -R ik || 2 For R ij and R ik The Euclidean distance, where i, j, k are quantifiers;
[0019] Each usage point of each energy consumption is used as a topology node. A topology edge is set between the two topology nodes corresponding to two usage points whose similarity is higher than the similarity threshold to form a topology connection. The similarity between the two topology edges is used as the edge weight of the topology edge. The point association topology of each energy consumption is constructed by the topology nodes, topology edges and edge weights.
[0020] As a preferred embodiment of the present invention, the location of monitoring points for each auxiliary material includes:
[0021] Community analysis is performed on the location association topology of each auxiliary material to divide each topology node in the location association topology into multiple location communities;
[0022] In each location community, the intra-node degree of each topological node is calculated as the node criticality of each topological node, and the usage points corresponding to the topological nodes whose node criticality is higher than the criticality threshold are used as the monitoring points for each type of auxiliary material.
[0023] As a preferred embodiment of the present invention, the location of monitoring points for each type of energy consumption includes:
[0024] Community analysis is performed on the location association topology for each type of energy consumption to divide each topology node in the location association topology into multiple location communities;
[0025] In each location community, the intra-node degree of each topology node is calculated as the node criticality of each topology node, and the usage points corresponding to the topology nodes whose node criticality is higher than the criticality threshold are used as monitoring points for each type of energy consumption.
[0026] As a preferred embodiment of the present invention, when a point community contains only one topological node, the intra-degree of the topological node in the point community is 1.
[0027] As a preferred embodiment of the present invention, the construction of the device association topology includes:
[0028] The operating condition time sequence of each piece of equipment in the gypsum board production line within a certain time period is obtained as the operating condition time sequence feature. Based on the operating condition time sequence feature, the operating condition similarity between any two pieces of equipment is calculated sequentially. The formula for calculating the operating condition similarity is as follows:
[0029]
[0030] In the formula, K rl L represents the similarity of operating conditions between the r-th device and the l-th device. r Let L be the operating time sequence characteristics of the r-th device. l Let L be the operating time sequence characteristics of the l-th device. r -L l || 2 For L r and L l The Euclidean distance, where r and l are quantifiers;
[0031] Each device is treated as a topology node. Topology edges are set between the two topology nodes corresponding to two devices with working conditions connected to form a topology connection. The similarity of the points corresponding to the two topology edges is used as the edge weight of the topology edge. The device association topology is constructed from the topology nodes, topology edges and edge weights.
[0032] In the equipment association topology, the equipment corresponding to the monitoring point of each auxiliary material and all equipment with topological connection relationship with the equipment corresponding to the monitoring point are identified. In the equipment association topology, the equipment corresponding to the monitoring point of each energy consumption and all equipment with topological connection relationship with the equipment corresponding to the monitoring point are identified. This is to extract all equipment with topological association with each monitoring point in each auxiliary material or each energy consumption.
[0033] As a preferred embodiment of the present invention, the construction of an anomaly factor topology by connecting each monitoring point and all devices with topological association to each monitoring point includes:
[0034] The abnormal factor topology of each auxiliary material is obtained by constructing a topology using the topological edges and edge weights in the topology of each monitoring point and all devices with topological association with each monitoring point in each auxiliary material.
[0035] For each type of energy consumption, the topology of abnormal factors is obtained by constructing a topology of the devices corresponding to each monitoring point and all devices with topological association with each monitoring point, using the topological edges and edge weights in the device association topology.
[0036] As a preferred embodiment of the present invention, the step of locating anomalous points in auxiliary material usage and energy consumption in the anomalous factor topology based on the auxiliary material usage characteristics and energy consumption characteristics includes:
[0037] The usage characteristics of each auxiliary material at each monitoring point are compared with the standard values of the usage characteristics of each monitoring point. Among them, the monitoring points where the usage characteristics of auxiliary materials are inconsistent with the standard values of the usage characteristics of auxiliary materials, as well as all equipment locations in the abnormal factor topology corresponding to the monitoring points, are regarded as abnormal points in the usage of each auxiliary material.
[0038] The energy consumption characteristics of each monitoring point for each type of energy consumption are compared with the standard values of energy consumption characteristics of each monitoring point. Among them, the monitoring points where the energy consumption characteristics are inconsistent with the standard values of energy consumption characteristics, as well as all device locations in the abnormal factor topology corresponding to the monitoring points, are regarded as abnormal points for each type of energy consumption.
[0039] As a preferred embodiment of the present invention, the time-series characteristics of material usage, energy consumption, and operating conditions are all subjected to time-series normalization processing.
[0040] Compared with the prior art, the present invention has the following advantages:
[0041] This invention establishes a point-to-point topology for each type of auxiliary material and a point-to-point topology for each type of energy consumption. Within these topologies, monitoring points for each auxiliary material and each type of energy consumption are located based on node criticality. An anomaly factor topology is constructed by linking each monitoring point with all devices that have topological relationships to that monitoring point. Based on the characteristics of auxiliary material usage and energy consumption, anomaly points in auxiliary material usage and energy consumption are located within this anomaly factor topology. This allows for real-time capture of effective data on water, electricity, gas usage, and auxiliary material consumption within the existing electrical control system, reducing secondary construction costs. The collected data is anomaly identified, abnormal data is automatically filtered out, and usage anomaly points are located through topological relationships, making anomaly detection simple and fast. Attached Figure Description
[0042] To more clearly illustrate the embodiments of the present invention or the technical solutions in 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 merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.
[0043] Figure 1 A flowchart illustrating the intelligent monitoring method for unit consumption in a gypsum board production line provided in this embodiment of the invention. Detailed Implementation
[0044] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0045] like Figure 1 As shown, this invention provides a method for intelligent monitoring of unit consumption in a gypsum board production line, comprising the following steps:
[0046] Step S1: Obtain all auxiliary material usage points and energy consumption points in the gypsum board production line, and sequentially establish the association topology for each auxiliary material usage point and each energy consumption point to obtain the point association topology for each auxiliary material and each energy consumption point. In the point association topology, locate the monitoring points for each auxiliary material and each energy consumption point by node criticality. Energy consumption points are represented by the location of the equipment that uses energy, including electricity points, water points, and gas points. Auxiliary material usage points are represented by the location of the equipment that uses auxiliary materials, including desulfurized gypsum, facing paper, and various low-value consumables.
[0047] The establishment of the point-to-point topology for each type of auxiliary material includes:
[0048] Obtain the time series sequence of material usage at each usage location of each auxiliary material over a certain period of time as the time series feature of material usage. Based on the time series feature of material usage, calculate the location similarity between any two usage locations of each auxiliary material. The formula for calculating the location similarity of auxiliary materials is as follows:
[0049]
[0050] In the formula, I ijk S represents the similarity between the j-th and k-th usage points of the i-th auxiliary material. ij S represents the time-series characteristic of material usage at the j-th usage point of the i-th auxiliary material. ik Let S be the time series characteristic of the material usage at the k-th usage point of the i-th auxiliary material. ij -S ik || 2 For S ij and S ik The Euclidean distance, where i, j, k are quantifiers;
[0051] Each usage point of each auxiliary material is used as a topological node. Topological edges are set between the two topological nodes corresponding to two usage points with a similarity higher than the similarity threshold to form a topological connection. This ensures that only strong correlations between usage points are retained in the topology, improving the representativeness of the monitoring points. This makes it feasible to reduce the dimensionality of real-time monitoring of all usage points to real-time monitoring of the monitoring points. Furthermore, the similarity between the points corresponding to two topological edges is used as the edge weight of the topological edge. The point association topology of each auxiliary material is constructed from the topological nodes, topological edges, and edge weights.
[0052] The establishment of the point-to-point topology for each type of energy consumption includes:
[0053] Obtain the time series sequence of energy consumption at each usage location for each type of energy consumption over a certain period of time as the time series feature of energy consumption. Based on the time series feature of energy consumption, calculate the location similarity between any two usage locations for each type of energy consumption. The formula for calculating the location similarity of energy consumption is as follows:
[0054]
[0055] In the formula, J ijk R represents the similarity between the j-th and k-th usage points of the i-th energy consumption type. ij Let R be the time-series characteristic of energy consumption at the j-th usage point of the i-th type of energy consumption. ik Let ||R| represent the time-series characteristics of energy consumption at the k-th usage point of the i-th type of energy consumption. ij -R ik || 2 For R ij and R ik The Euclidean distance, where i, j, k are quantifiers;
[0056] Each usage point for each type of energy consumption is treated as a topological node. Topological edges are set between the two topological nodes corresponding to two usage points with a similarity higher than the similarity threshold to form a topological connection. This ensures that only strong correlations between usage points are retained in the topology, improving the representativeness of the monitoring points. This makes it feasible to reduce the dimensionality of real-time monitoring of all usage points to real-time monitoring of the monitoring points. Furthermore, the similarity between the points corresponding to two topological edges is used as the edge weight of the topological edge. The topological nodes, topological edges, and edge weights are used to construct the point association topology for each type of energy consumption.
[0057] By constructing the point-to-point correlation topology for each auxiliary material and each energy consumption level, strong correlations can be established between various usage points for each auxiliary material. For example, point 1 is downstream of point 2's production, and point 2 depends on point 1's production process or results. If point 1 experiences an anomaly in auxiliary material or energy consumption, point 2 will also experience anomalies in auxiliary material or energy consumption due to the influence of point 1. Therefore, when monitoring points, only point 1 can be monitored, and the data results for point 2 can be obtained based on the strong correlation between point 1 and point 2. Thus, the point-to-point correlation topology will reflect various strong correlations between points. Subsequent analysis of the point-to-point correlation topology will further demonstrate this. Group analysis is used to identify topological nodes with node criticality in the point-location association topology as monitoring points. This reduces the dimensionality of real-time monitoring of all usage points to real-time monitoring of specific monitoring points. Reducing the number of monitoring points decreases the amount of real-time data processing. At the same time, the representativeness of the monitoring points to the remaining usage points (the association relationships presented in the point-location association topology) allows for accurate acquisition of the data characteristics of the remaining usage points based on the data characteristics of the monitoring points. Therefore, by locating monitoring points through the point-location association topology, real-time monitoring of auxiliary material usage data or energy consumption data can be achieved, reducing the dimensionality of real-time data concurrency while ensuring the accuracy of acquiring abnormal data.
[0058] Step S2: Extract all equipment with topological association with each monitoring point from the equipment association topology formed by each piece of equipment in the gypsum board production line, and construct an anomaly factor topology by combining each monitoring point and all equipment with topological association with each monitoring point.
[0059] The location of monitoring points for each type of auxiliary material includes:
[0060] Community analysis is performed on the location association topology of each auxiliary material to divide each topology node in the location association topology into multiple location communities;
[0061] In each location community, the intra-node degree of each topological node is calculated as the node criticality of each topological node, and the usage points corresponding to the topological nodes whose node criticality is higher than the criticality threshold are used as the monitoring points for each type of auxiliary material.
[0062] The location of monitoring points for each type of energy consumption includes:
[0063] Community analysis is performed on the location association topology for each type of energy consumption to divide each topology node in the location association topology into multiple location communities;
[0064] In each location community, the intra-node degree of each topology node is calculated as the node criticality of each topology node, and the usage points corresponding to the topology nodes whose node criticality is higher than the criticality threshold are used as monitoring points for each type of energy consumption.
[0065] If a point community contains only one topology node, then the in-degree of that topology node in the point community is 1.
[0066] The construction of the device association topology includes:
[0067] The operating condition time sequence of each piece of equipment in the gypsum board production line within a certain time period is obtained as the operating condition time sequence feature. Based on the operating condition time sequence feature, the operating condition similarity between any two pieces of equipment is calculated sequentially. The formula for calculating the operating condition similarity is as follows:
[0068]
[0069] In the formula, K rl L represents the similarity of operating conditions between the r-th device and the l-th device. r Let L be the operating time sequence characteristics of the r-th device. l Let L be the operating time sequence characteristics of the l-th device. r -L l || 2 For L r and L l The Euclidean distance, where r and l are quantifiers;
[0070] Each device is treated as a topology node. Topology edges are set between the two topology nodes corresponding to two devices with working conditions connected to form a topology connection. The similarity of the points corresponding to the two topology edges is used as the edge weight of the topology edge. The device association topology is constructed from the topology nodes, topology edges and edge weights.
[0071] In the equipment association topology, the equipment corresponding to the monitoring point of each auxiliary material and all equipment with topological connection relationship with the equipment corresponding to the monitoring point are identified. In the equipment association topology, the equipment corresponding to the monitoring point of each energy consumption and all equipment with topological connection relationship with the equipment corresponding to the monitoring point are identified. This is to extract all equipment with topological association with each monitoring point in each auxiliary material or each energy consumption.
[0072] An anomaly topology is constructed by assigning each monitoring point and all devices with topological relationships to each monitoring point, including:
[0073] The abnormal factor topology of each auxiliary material is obtained by constructing a topology using the topological edges and edge weights in the topology of each monitoring point and all devices with topological association with each monitoring point in each auxiliary material.
[0074] For each type of energy consumption, the topology of abnormal factors is obtained by constructing a topology of the devices corresponding to each monitoring point and all devices with topological association with each monitoring point, using the topological edges and edge weights in the device association topology.
[0075] Based on the equipment association topology, an abnormal factor topology is constructed by fusion of monitoring points. This allows for accurate mapping of data anomalies at monitoring points into the equipment topology, i.e., locating the various devices in the production line that caused the data anomalies at the monitoring points. This enables the location of abnormal devices without having to check the operating conditions of each device individually to determine the device causing the anomaly.
[0076] Step S3: Monitor the usage characteristics of each auxiliary material and the energy consumption characteristics at each monitoring point in real time. Based on the usage characteristics of the auxiliary materials and the energy consumption characteristics, locate the abnormal points of auxiliary material usage and energy consumption in the abnormal factor topology.
[0077] Based on the characteristics of auxiliary material usage and energy consumption, abnormal points in auxiliary material usage and energy consumption are located in the abnormal factor topology, including:
[0078] The usage characteristics of each auxiliary material at each monitoring point are compared with the standard values of the usage characteristics of each monitoring point. Among them, the monitoring points where the usage characteristics of auxiliary materials are inconsistent with the standard values of the usage characteristics of auxiliary materials, as well as all equipment locations in the abnormal factor topology corresponding to the monitoring points, are regarded as abnormal points in the usage of each auxiliary material.
[0079] The energy consumption characteristics of each monitoring point for each type of energy consumption are compared with the standard values of energy consumption characteristics of each monitoring point. Among them, the monitoring points where the energy consumption characteristics are inconsistent with the standard values of energy consumption characteristics, as well as all device locations in the abnormal factor topology corresponding to the monitoring points, are regarded as abnormal points for each type of energy consumption.
[0080] The time-series characteristics of material usage, energy consumption, and operating conditions were all normalized.
[0081] After automatically filtering out abnormal data, the valid data can be analyzed to generate line charts, bar charts, pie charts, etc. Finally, the data are generated into reports. The reports have the function of querying data for a specific day, a specific time period of a specific day, a specific month, or a specific year. They also have an over-limit reminder function and can be queried through an APP terminal (including mobile devices such as mobile phones and computer terminals). On the implementation side, data acquisition from field instruments is mainly carried out through the MODBUS protocol. The acquired data is transmitted wirelessly to reduce field wiring, lower construction difficulty, and increase feasibility.
[0082] This invention establishes a point-to-point topology for each type of auxiliary material and a point-to-point topology for each type of energy consumption. Within this topology, monitoring points for each auxiliary material and each type of energy consumption are located based on node criticality. An anomaly factor topology is constructed by linking each monitoring point with all devices that have topological relationships to that monitoring point. Based on the characteristics of auxiliary material usage and energy consumption, anomaly points in auxiliary material usage and energy consumption are located within this anomaly factor topology. This allows for real-time capture of effective data on water, electricity, gas usage, and auxiliary material consumption within the existing electrical control system, reducing secondary construction costs. The collected data is anomaly identified, abnormal data is automatically filtered out, and usage anomaly points are located through topological relationships, making anomaly investigation simple and fast.
[0083] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.
Claims
1. A method for intelligent monitoring of unit consumption in a gypsum board production line, characterized in that: Includes the following steps: Step S1: Obtain all auxiliary material usage points and energy consumption points in the gypsum board production line, and sequentially establish the association topology for each auxiliary material usage point and each energy consumption point to obtain the point association topology for each auxiliary material and the point association topology for each energy consumption point. In the point association topology for each auxiliary material and the point association topology for each energy consumption point, locate the monitoring points for each auxiliary material and the monitoring points for each energy consumption point by node criticality. The energy consumption point represents the location of the equipment that uses energy, and the auxiliary material usage point represents the location of the equipment that uses auxiliary materials. Step S2: Extract all equipment with topological association with each monitoring point from the equipment association topology formed by each piece of equipment in the gypsum board production line, and construct an anomaly factor topology by combining each monitoring point and all equipment with topological association with each monitoring point. Step S3: Monitor the usage characteristics of each auxiliary material and the energy consumption characteristics at each auxiliary material usage point and energy consumption point in real time. Based on the usage characteristics of the auxiliary materials and the energy consumption characteristics, locate the abnormal points of auxiliary material usage and energy consumption in the abnormal factor topology. The establishment of the point-to-point topology for each type of auxiliary material includes: The time series sequence of material usage at each usage location of each auxiliary material over a certain period of time is obtained as the time series feature of material usage. Based on the time series feature of material usage, the location similarity between any two usage locations of each auxiliary material is calculated sequentially. The formula for calculating the location similarity of auxiliary materials is as follows: ； In the formula, I ijk S represents the similarity between the j-th and k-th usage points of the i-th auxiliary material. ij S represents the time-series characteristic of material usage at the j-th usage point of the i-th auxiliary material. ik Let S be the time series characteristic of the material usage at the k-th usage point of the i-th auxiliary material. ij -S ik || 2 For S ij and S ik The Euclidean distance, where i, j, k are quantifiers; Each usage point of each auxiliary material is used as a topological node. Topological edges are set between the two topological nodes corresponding to two usage points with a similarity higher than the similarity threshold to form a topological connection. The similarity between the two topological edges is used as the edge weight of the topological edge. The point-location association topology of each auxiliary material is constructed by topological nodes, topological edges and edge weights.
2. The intelligent monitoring method for unit consumption of a gypsum board production line according to claim 1, characterized in that: The establishment of the point-to-point topology for each type of energy consumption includes: Obtain the time series sequence of energy consumption at each usage location for each type of energy consumption over a certain period of time as the time series feature of energy consumption. Based on the time series feature of energy consumption, calculate the location similarity between any two usage locations for each type of energy consumption. The formula for calculating the location similarity of energy consumption is as follows: ； In the formula, J ijk R represents the similarity between the j-th and k-th usage points of the i-th energy consumption type. ij Let R be the time-series characteristic of energy consumption at the j-th usage point of the i-th type of energy consumption. ik Let ||R| represent the time-series characteristics of energy consumption at the k-th usage point of the i-th type of energy consumption. ij -R ik || 2 For R ij and R ik The Euclidean distance, where i, j, k are quantifiers; Each usage point of each energy consumption is used as a topology node. A topology edge is set between the two topology nodes corresponding to two usage points whose similarity is higher than the similarity threshold to form a topology connection. The similarity between the two topology edges is used as the edge weight of the topology edge. The point association topology of each energy consumption is constructed by the topology nodes, topology edges and edge weights.
3. The intelligent monitoring method for unit consumption of a gypsum board production line according to claim 2, characterized in that: The location of monitoring points for each type of auxiliary material includes: Community analysis is performed on the location association topology of each auxiliary material to divide each topology node in the location association topology into multiple location communities; In each location community, the intra-node degree of each topological node is calculated as the node criticality of each topological node, and the usage points corresponding to the topological nodes whose node criticality is higher than the criticality threshold are used as the monitoring points for each type of auxiliary material.
4. The intelligent monitoring method for unit consumption of a gypsum board production line according to claim 3, characterized in that: The location of monitoring points for each type of energy consumption includes: Community analysis is performed on the location association topology for each type of energy consumption to divide each topology node in the location association topology into multiple location communities; In each location community, the intra-node degree of each topology node is calculated as the node criticality of each topology node, and the usage points corresponding to the topology nodes whose node criticality is higher than the criticality threshold are used as monitoring points for each type of energy consumption.
5. The intelligent monitoring method for unit consumption of a gypsum board production line according to claim 4, characterized in that: If a point community contains only one topology node, then the in-degree of that topology node in the point community is 1.
6. The intelligent monitoring method for unit consumption of a gypsum board production line according to claim 5, characterized in that: The construction of the device association topology includes: The operating condition time sequence of each piece of equipment in the gypsum board production line within a certain time period is obtained as the operating condition time sequence feature. Based on the operating condition time sequence feature, the operating condition similarity between any two pieces of equipment is calculated sequentially. The formula for calculating the operating condition similarity is as follows: ； In the formula, K rl L represents the similarity of operating conditions between the r-th device and the l-th device. r Let L be the operating time sequence characteristics of the r-th device. l Let L be the operating time sequence characteristics of the l-th device. r -L l || 2 For L r and L l The Euclidean distance, where r and l are quantifiers; Each device is treated as a topology node. Topology edges are set between the two topology nodes corresponding to two devices with working conditions connected to form a topology connection. The similarity of the points corresponding to the two topology edges is used as the edge weight of the topology edge. The device association topology is constructed from the topology nodes, topology edges and edge weights. In the equipment association topology, the equipment corresponding to the monitoring point of each auxiliary material and all equipment with topological connection relationship with the equipment corresponding to the monitoring point are identified. In the equipment association topology, the equipment corresponding to the monitoring point of each energy consumption and all equipment with topological connection relationship with the equipment corresponding to the monitoring point are identified. This is to extract all equipment with topological association with each monitoring point in each auxiliary material or each energy consumption.
7. The intelligent monitoring method for unit consumption of a gypsum board production line according to claim 6, characterized in that: The process of constructing an anomaly topology for each monitoring point and all devices with topological associations to each monitoring point includes: The abnormal factor topology of each auxiliary material is obtained by constructing a topology using the topological edges and edge weights in the topology of each monitoring point and all devices with topological association with each monitoring point in each auxiliary material. For each type of energy consumption, the topology of abnormal factors is obtained by constructing a topology of the devices corresponding to each monitoring point and all devices with topological association with each monitoring point, using the topological edges and edge weights in the device association topology.
8. The intelligent monitoring method for unit consumption of a gypsum board production line according to claim 7, characterized in that, The step of locating anomalous points in auxiliary material usage and energy consumption based on the characteristics of auxiliary material usage and energy consumption in the anomalous factor topology includes: The usage characteristics of each auxiliary material at each monitoring point are compared with the standard values of the usage characteristics of each monitoring point. Among them, the monitoring points where the usage characteristics of auxiliary materials are inconsistent with the standard values of the usage characteristics of auxiliary materials, as well as all equipment locations in the abnormal factor topology corresponding to the monitoring points, are regarded as abnormal points in the usage of each auxiliary material. The energy consumption characteristics of each monitoring point for each type of energy consumption are compared with the standard values of energy consumption characteristics of each monitoring point. Among them, the monitoring points where the energy consumption characteristics are inconsistent with the standard values of energy consumption characteristics, as well as all device locations in the abnormal factor topology corresponding to the monitoring points, are regarded as abnormal points for each type of energy consumption.
9. The intelligent monitoring method for unit consumption of a gypsum board production line according to claim 8, characterized in that, The time-series characteristics of material usage, energy consumption, and operating conditions are all subjected to time-series normalization.