GIS equipment production and manufacturing method based on digital twin workshop

Abstract

The invention discloses a GIS equipment production and manufacturing method based on a digital twin workshop, and belongs to the technical field of power equipment. Real-time synchronization of physical and digital workshops is realized by constructing a digital twin workshop and applying three-dimensional modeling software, a sensor, a particle filter algorithm and the like. A genetic algorithm and a particle swarm optimization algorithm are utilized to optimize production plans and scheduling, and efficiency is improved by considering factors in multiple aspects and by means of a simulation technology. Real-time monitoring, abnormity early warning and quality prediction are performed on the production process by means of digital twinning and machine learning algorithms, and the process is adjusted in time. And production resource management and optimization are carried out by using an RFID technology, a path optimization algorithm and the like, so that the cost is reduced. The problems of information islanding, low efficiency, difficulty in quality control and the like in traditional GIS equipment production and manufacturing are effectively solved, the production efficiency can be improved, the cost can be reduced, the product quality can be improved, intelligent production can be realized, and the method has important application value and market prospect.

Description

Technical Field

[0001] This invention relates to the field of power equipment technology, and in particular to a method for manufacturing GIS equipment based on a digital twin workshop. Background Technology

[0002] In the current power equipment manufacturing industry, GIS (Gas Insulated Switchgear) is a key piece of equipment for ensuring the safe and stable operation of power systems, and its manufacturing technology has attracted much attention. Traditional GIS equipment manufacturing methods have revealed many insurmountable problems in long-term practice, severely restricting the further development of the industry.

[0003] The phenomenon of information silos is severe: In traditional production models, data and information generated at each stage of production, from raw material procurement and component processing to equipment assembly and quality inspection, are isolated from each other, lacking effective data sharing mechanisms and collaborative platforms. Each department operates like an information silo, working independently, leading to untimely and inaccurate information transmission during production, and hindering efficient decision-making based on unified data. For example, the production department cannot obtain real-time material arrival information from the procurement department, easily causing production interruptions; the quality inspection department cannot promptly report quality problems to the production department, resulting in an increased defect rate.

[0004] Low production efficiency: Production planning often relies on experience, failing to fully utilize data for scientific analysis and neglecting to comprehensively consider factors such as equipment's actual processing capacity, material supply cycles, and personnel skill levels and workloads. This leads to a disconnect between production plans and actual production conditions, resulting in inflexible and untimely production scheduling. Consequently, equipment often remains idle for extended periods or tasks accumulate excessively, extending the overall production cycle and hindering the improvement of production efficiency.

[0005] Difficulties in quality control: Traditional production processes rely mainly on post-production inspection to control product quality, lacking real-time monitoring and quality prediction methods. Quality inspection is only conducted after production is complete. Once quality problems are discovered, significant time and resources are often wasted, making it difficult to promptly identify and resolve potential quality issues and ensure product quality stability and consistency.

[0006] With the rise of Industry 4.0 and smart manufacturing, digital twin technology, as an emerging digital technology, offers new ideas and methods for solving the problems of traditional production models. By constructing virtual digital models corresponding to physical entities, digital twin technology enables real-time mapping, monitoring, analysis, and optimization of physical entities, potentially breaking through the bottlenecks of traditional production models and driving the development of GIS equipment manufacturing towards intelligence and efficiency. Summary of the Invention

[0007] The purpose of this invention is to provide a GIS equipment manufacturing method based on a digital twin workshop, so as to improve production efficiency, reduce production costs, improve product quality, and realize intelligent production.

[0008] To achieve the above objectives, the present invention provides a method for manufacturing GIS equipment based on a digital twin workshop, comprising the following steps:

[0009] S1. Constructing a digital twin workshop: Using 3D modeling software to construct an accurate digital model of the physical entity of the workshop, employing lightweight technology to improve rendering efficiency, and combining environmental parameters collected by sensors to construct an environmental model; collecting equipment, material, and environmental data through various sensors, RFID technology, and cameras; using particle filtering algorithms to fuse the data and establish a mapping function to achieve real-time synchronization between the physical and digital workshops;

[0010] S2. Production Planning and Scheduling Optimization: Based on orders and workshop capacity, the production plan is optimized using a genetic algorithm, taking into account various constraints and using simulation technology to analyze indicators; based on real-time data, the particle swarm optimization algorithm is used for dynamic scheduling, taking into account uncertainties, and the optimal solution is selected through simulation verification using digital twin technology;

[0011] S3. Production process monitoring and prediction: Real-time display of workshop production status using digital twin technology, and construction of interface using data visualization; construction of equipment failure prediction model based on historical data and support vector machine, and prediction of failure by combining digital twin and stress analysis; construction of quality prediction model using long short-term memory network, and adjustment of process based on key parameters.

[0012] S4. Production Resource Management and Optimization: Utilize RFID technology and path optimization algorithms to manage materials and set up an inventory early warning mechanism; develop preventative maintenance plans based on equipment operation data and optimize maintenance schemes using digital twin technology; and establish energy consumption models using linear programming algorithms to optimize energy management.

[0013] Preferably, in step S1, a lightweight modeling technique based on a triangle mesh simplification algorithm is used to remove details that have little impact on the rendering effect, reduce model complexity, and improve rendering efficiency. The specific steps are as follows:

[0014] Before performing lightweighting, the original model is first divided into a triangular mesh; the model surface is discretized into a large number of triangular patches, each of which is defined by three vertices;

[0015] Assuming the geometry of the original model is M, after triangular meshing, the resulting triangular mesh set is T = {t1, t2, ..., t...} n}, where t iThis represents the i-th triangular facet;

[0016] The triangular mesh simplification algorithm removes details with minimal impact on rendering by performing edge folding operations. The criterion used is the cost function of edge folding, the expression of which is shown below:

[0017] Assume the edge to be folded is e = (v i ,v j ), where v i and v j They are the two vertices of the edge;

[0018] The cost is calculated based on the distance from a vertex to its neighborhood plane. Let vertex v be an example of a vertex. i The equation of the neighborhood plane is ax + by + cz + d = 0; then the vertex Similarly, calculate vertex v j Distance d to its neighboring plane j The cost function is C(e) = d i +d j ;

[0019] Where a, b, and c are the components of the plane normal vector, and d is the plane offset; x i y i z i It is vertex v i Coordinate values ​​in three-dimensional space;

[0020] By calculating the cost function value of all edges, the edge with the lowest cost is selected for the folding operation.

[0021] Preferably, in step S1, the specific expression for fusing multi-source heterogeneous data using the particle filter algorithm is as follows:

[0022] The system state equation is set as follows

[0023] x k =f(x) k-1 ,u k-1 )+w k-1 ;

[0024] Among them, x k Let x represent the system state at time k. k-1 U represents the system state at time k-1. k-1 w represents the control input at time k-1. k-1 Let f represent the process noise, and let f represent the transition relationship between the state at time k-1 and the control input to the state at time k.

[0025] The observation equation is

[0026] z k =h(x k)+v k ;

[0027] Among them, z k Let x represent the observation at time k. k Let v be the system state at time k. k represents observation noise, and h represents the mapping relationship from the system state to the observed value;

[0028] A more accurate system state estimate can be obtained by updating the particle weights and resampling.

[0029] Preferably, in step S2, based on order demand and workshop production capacity, a genetic algorithm is used to optimize the production plan, with the objective function being:

[0030] minf(x) = w1*T + w2*C;

[0031] Where T is the production cycle, C is the production cost, and w1 and w2 are weighting coefficients;

[0032] Simulation technology is used to establish a production system simulation model to simulate the production process under different production plans and analyze relevant indicators to optimize the production plan.

[0033] Preferably, in step S2, dynamic scheduling optimization is performed using a particle swarm optimization algorithm based on real-time production data and workshop status, with the objective function being:

[0034] maxf(x) = w3*U + w4*E;

[0035] Where U is the equipment utilization rate, E is the production efficiency, and w3 and w4 are weighting coefficients;

[0036] Digital twin technology is used to simulate, verify, and optimize production scheduling schemes. The relevant indicators of different schemes are compared and analyzed to select the optimal scheme.

[0037] Preferably, in step S3, real-time monitoring utilizes digital twin technology to map the workshop production status into a digital model in real time, enabling real-time monitoring of equipment operating status, material flow, product quality information, etc.; a visual interface for workshop production status is constructed, and data visualization technology is adopted to display the operating time and downtime of different equipment through bar charts and to display the changing trends of product quality indicators through line charts.

[0038] Preferably, in step S3, the anomaly warning is based on historical data and the support vector machine (SVM) algorithm to construct a device fault prediction model;

[0039] Assume the training sample set is (x i ,y i ), i = 1, 2, ..., n, where x iGiven the input feature vector, y i Category labels;

[0040] SVM separates normal samples from faulty samples by finding an optimal classification hyperplane. In the prediction phase, new equipment operation data is used as input feature vectors and fed into the trained SVM model to determine whether the equipment has a risk of failure.

[0041] Digital twin technology is used to simulate and analyze the operating status of equipment. By establishing a digital model of the equipment, the operating conditions of the equipment under different working conditions are simulated, and the time and location of equipment failures can be predicted.

[0042] Preferably, in step S3, the quality prediction utilizes a Long Short-Term Memory (LSTM) network within a Recurrent Neural Network (RNN) to construct a product quality prediction model. By introducing memory units and gating mechanisms, the long-term dependency problem in time-series data can be effectively addressed. Based on key parameters in the production process, product quality indicators are predicted using the LSTM model.

[0043] Preferably, in step S4, RFID technology is used to achieve real-time tracking and management of material information, and material waiting time is reduced by optimizing the delivery route during the material delivery process.

[0044] Based on equipment operation data and predictive models, a preventive maintenance plan is developed, and the remaining service life of the equipment is predicted by analyzing the equipment operation data.

[0045] Digital twin technology is used to simulate and optimize equipment maintenance processes, modeling different maintenance schemes in a digital model. The optimal maintenance scheme is then selected to improve maintenance efficiency.

[0046] By using data analysis and linear programming algorithms, we can optimize workshop energy consumption. By establishing an energy consumption model, we can analyze the energy consumption of different equipment and different production processes, identify energy waste, and formulate energy-saving measures.

[0047] Therefore, the GIS equipment manufacturing method based on the above-described structure of the present invention has the following beneficial effects:

[0048] (1) This invention utilizes genetic algorithms and particle swarm optimization to optimize production planning and scheduling, fully considering factors such as order demand, equipment capacity, material supply, and personnel allocation, making production planning more reasonable and production scheduling more flexible and efficient. Simulation technology is used to verify and optimize different solutions, effectively reducing production cycles and improving equipment utilization.

[0049] (2) This invention utilizes digital twin technology and machine learning algorithms to achieve real-time monitoring and early warning of anomalies in the production process, enabling timely detection and resolution of production problems and avoiding production interruptions and delays. Through a quality prediction model, product quality issues can be predicted in advance, allowing for timely adjustments to production processes and ensuring continuous and efficient production.

[0050] (3) This invention utilizes RFID technology to achieve precise management of materials and optimize delivery routes, reducing material waiting time and inventory backlog, and lowering material waste costs. Through energy management optimization algorithms, it analyzes the energy consumption of different equipment and production processes, and takes energy-saving measures to reduce energy consumption costs.

[0051] (4) By constructing a digital model that is synchronized with the physical workshop in real time through digital twin technology, and combining artificial intelligence algorithms to analyze and make decisions on production data, intelligent monitoring, optimization and management of the production process can be realized.

[0052] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. Attached Figure Description

[0053] Figure 1 This is a schematic flowchart of a GIS equipment manufacturing method based on a digital twin workshop according to the present invention. Detailed Implementation

[0054] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0055] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0056] Example

[0057] like Figure 1 As shown, this invention provides a method for manufacturing GIS equipment based on a digital twin workshop, comprising the following steps:

[0058] S1. Constructing a digital twin workshop: Using 3D modeling software to construct an accurate digital model of the physical entity of the workshop, employing lightweight technology to improve rendering efficiency, and combining environmental parameters collected by sensors to construct an environmental model; collecting equipment, material, and environmental data through various sensors, RFID technology, and cameras; using particle filtering algorithms to fuse the data and establish a mapping function to achieve real-time synchronization between the physical and digital workshops;

[0059] S2. Production Planning and Scheduling Optimization: Based on orders and workshop capacity, the production plan is optimized using a genetic algorithm, taking into account various constraints and using simulation technology to analyze indicators; based on real-time data, the particle swarm optimization algorithm is used for dynamic scheduling, taking into account uncertainties, and the optimal solution is selected through simulation verification using digital twin technology;

[0060] S3. Production process monitoring and prediction: Real-time display of workshop production status using digital twin technology, and construction of interface using data visualization; construction of equipment failure prediction model based on historical data and support vector machine, and prediction of failure by combining digital twin and stress analysis; construction of quality prediction model using long short-term memory network, and adjustment of process based on key parameters.

[0061] S4. Production Resource Management and Optimization: Utilize RFID technology and path optimization algorithms to manage materials and set up an inventory early warning mechanism; develop preventative maintenance plans based on equipment operation data and optimize maintenance schemes using digital twin technology; and establish energy consumption models using linear programming algorithms to optimize energy management.

[0062] Preferably, in step S1, a lightweight modeling technique based on a triangle mesh simplification algorithm is used to remove details that have little impact on the rendering effect, reduce model complexity, and improve rendering efficiency. The specific steps are as follows:

[0063] Before performing lightweighting, the original model is first divided into a triangular mesh; the model surface is discretized into a large number of triangular patches, each of which is defined by three vertices;

[0064] Assuming the geometry of the original model is M, after triangular meshing, the resulting triangular mesh set is T = {t1, t2, ..., t...} n}, where t i This represents the i-th triangular facet;

[0065] The triangular mesh simplification algorithm removes details with minimal impact on rendering by performing edge folding operations. The criterion used is the cost function of edge folding, the expression of which is shown below:

[0066] Assume the edge to be folded is e = (v i ,v j ), where v i and vj They are the two vertices of the edge;

[0067] The cost is calculated based on the distance from a vertex to its neighborhood plane. Let vertex v be an example of a vertex. i The equation of the neighborhood plane is ax + by + cz + d = 0; then the vertex Similarly, calculate vertex v j Distance d to its neighboring plane j The cost function is C(e) = d i +d j ;

[0068] Where a, b, and c are the components of the plane normal vector, and d is the plane offset; x i y i z i It is vertex v i Coordinate values ​​in three-dimensional space;

[0069] By calculating the cost function value of all edges, the edge with the lowest cost is selected for the folding operation.

[0070] Preferably, in step S1, the specific expression for fusing multi-source heterogeneous data using the particle filter algorithm is as follows:

[0071] The system state equation is set as follows

[0072] x k =f(x) k-1 ,u k-1 )+w k-1 ;

[0073] Among them, x k Let x represent the system state at time k. k-1 U represents the system state at time k-1. k-1 w represents the control input at time k-1. k-1 Let f represent the process noise, and let f represent the transition relationship between the state at time k-1 and the control input to the state at time k.

[0074] The observation equation is

[0075] z k =h(x k )+v k ;

[0076] Among them, z k Let x represent the observation at time k. k Let v be the system state at time k. k represents observation noise, and h represents the mapping relationship from the system state to the observed value;

[0077] A more accurate system state estimate can be obtained by updating the particle weights and resampling.

[0078] Preferably, in step S2, based on order demand and workshop production capacity, a genetic algorithm is used to optimize the production plan, with the objective function being:

[0079] minf(x) = w1*T + w2*C;

[0080] Where T is the production cycle, C is the production cost, and w1 and w2 are weighting coefficients;

[0081] Simulation technology is used to establish a production system simulation model to simulate the production process under different production plans and analyze relevant indicators to optimize the production plan.

[0082] Preferably, in step S2, dynamic scheduling optimization is performed using a particle swarm optimization algorithm based on real-time production data and workshop status, with the objective function being:

[0083] maxf(x) = w3*U + w4*E;

[0084] Where U is the equipment utilization rate, E is the production efficiency, and w3 and w4 are weighting coefficients;

[0085] Digital twin technology is used to simulate, verify, and optimize production scheduling schemes. The relevant indicators of different schemes are compared and analyzed to select the optimal scheme.

[0086] Preferably, in step S3, real-time monitoring utilizes digital twin technology to map the workshop production status into a digital model in real time, enabling real-time monitoring of equipment operating status, material flow, product quality information, etc.; a visual interface for workshop production status is constructed, and data visualization technology is adopted to display the operating time and downtime of different equipment through bar charts and to display the changing trends of product quality indicators through line charts.

[0087] Preferably, in step S3, the anomaly warning is based on historical data and the support vector machine (SVM) algorithm to construct a device fault prediction model;

[0088] Assume the training sample set is (x i ,y i ), i = 1, 2, ..., n, where x i Given the input feature vector, y i Category labels;

[0089] SVM separates normal samples from faulty samples by finding an optimal classification hyperplane. In the prediction phase, new equipment operation data is used as input feature vectors and fed into the trained SVM model to determine whether the equipment has a risk of failure.

[0090] Digital twin technology is used to simulate and analyze the operating status of equipment. By establishing a digital model of the equipment, the operating conditions of the equipment under different working conditions are simulated, and the time and location of equipment failures can be predicted.

[0091] Preferably, in step S3, the quality prediction utilizes a Long Short-Term Memory (LSTM) network within a Recurrent Neural Network (RNN) to construct a product quality prediction model. By introducing memory units and gating mechanisms, the long-term dependency problem in time-series data can be effectively addressed. Based on key parameters in the production process, product quality indicators are predicted using the LSTM model.

[0092] Preferably, in step S4, RFID technology is used to achieve real-time tracking and management of material information, and material waiting time is reduced by optimizing the delivery route during the material delivery process.

[0093] Based on equipment operation data and predictive models, a preventive maintenance plan is developed, and the remaining service life of the equipment is predicted by analyzing the equipment operation data.

[0094] Digital twin technology is used to simulate and optimize equipment maintenance processes, modeling different maintenance schemes in a digital model. The optimal maintenance scheme is then selected to improve maintenance efficiency.

[0095] By using data analysis and linear programming algorithms, we can optimize workshop energy consumption. By establishing an energy consumption model, we can analyze the energy consumption of different equipment and different production processes, identify energy waste, and formulate energy-saving measures.

[0096] Therefore, the present invention adopts the above-mentioned method for manufacturing GIS equipment based on a digital twin workshop, which can effectively improve production efficiency, reduce production costs, improve product quality, and realize the intelligent manufacturing of GIS equipment, and has important application value and market prospects.

[0097] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for manufacturing GIS equipment based on a digital twin workshop, characterized in that, Includes the following steps: S1. Constructing a digital twin workshop: Using 3D modeling software to construct an accurate digital model of the physical entity of the workshop, employing lightweight technology to improve rendering efficiency, and combining environmental parameters collected by sensors to construct an environmental model; collecting equipment, material, and environmental data through various sensors, RFID technology, and cameras; using particle filtering algorithms to fuse the data and establish a mapping function to achieve real-time synchronization between the physical and digital workshops; S2. Production Planning and Scheduling Optimization: Based on orders and workshop capacity, the production plan is optimized using a genetic algorithm, taking into account various constraints and using simulation technology to analyze indicators; based on real-time data, the particle swarm optimization algorithm is used for dynamic scheduling, taking into account uncertainties, and the optimal solution is selected through simulation verification using digital twin technology; S3. Production process monitoring and prediction: Real-time display of workshop production status using digital twin technology, and construction of interface using data visualization; construction of equipment failure prediction model based on historical data and support vector machine, and prediction of failure by combining digital twin and stress analysis; construction of quality prediction model using long short-term memory network, and adjustment of process based on key parameters. S4. Production Resource Management and Optimization: Utilize RFID technology and path optimization algorithms to manage materials and set up an inventory early warning mechanism; Develop preventative maintenance plans based on equipment operating data and optimize maintenance schemes using digital twin technology; Energy consumption models can be established using linear programming algorithms to optimize energy management.

2. The GIS equipment manufacturing method based on a digital twin workshop according to claim 1, characterized in that: In step S1, lightweight modeling techniques are used to improve rendering efficiency. The specific steps are as follows: Before performing lightweighting, the original model is first divided into a triangular mesh; the model surface is discretized into a large number of triangular patches, each of which is defined by three vertices; Assuming the geometry of the original model is M, after triangular meshing, the resulting triangular mesh set is T = {t1, t2, ..., t...} n }, where t i This represents the i-th triangular facet; The triangular mesh simplification algorithm removes details with minimal impact on rendering by performing edge folding operations. The criterion used is the cost function of edge folding, the expression of which is shown below: Assume the edge to be folded is e = (v i ,v j ), where v i and v j They are the two vertices of the edge; The cost is calculated based on the distance from a vertex to its neighborhood plane. Let vertex v be an example of a vertex. i The equation of the neighborhood plane is ax + by + cz + d = 0; then the vertex Similarly, calculate vertex v j Distance d to its neighboring plane j The cost function is C(e) = d i +d j ; Where a, b, and c are the components of the plane normal vector, and d is the plane offset; x i y i z i It is vertex v i Coordinate values ​​in three-dimensional space; By calculating the cost function value of all edges, the edge with the lowest cost is selected for the folding operation.

3. The GIS equipment manufacturing method based on a digital twin workshop according to claim 1, characterized in that: In step S1, the specific expression for fusing multi-source heterogeneous data using the particle filter algorithm is as follows: The system state equation is set as follows x k =f(x k-1 ,u k-1 )+w k-1 ; Among them, x k Let x represent the system state at time k. k-1 U represents the system state at time k-1. k-1 w represents the control input at time k-1. k-1 Let f represent the process noise, and let f represent the transition relationship between the state at time k-1 and the control input to the state at time k. The observation equation is z k =h(x k )+v k ; Among them, z k Let x represent the observation at time k. k Let v be the system state at time k. k represents observation noise, and h represents the mapping relationship from the system state to the observed value; A more accurate system state estimate can be obtained by updating the particle weights and resampling.

4. The GIS equipment manufacturing method based on a digital twin workshop according to claim 1, characterized in that: In step S2, based on order demand and workshop production capacity, a genetic algorithm is used to optimize the production plan, with the objective function being: minf(x) = w1*T + w2*C; Where T is the production cycle, C is the production cost, and w1 and w2 are weighting coefficients; Simulation technology is used to establish a production system simulation model to simulate the production process under different production plans and analyze relevant indicators to optimize the production plan.

5. A method for manufacturing GIS equipment based on a digital twin workshop according to claim 1, characterized in that: In step S2, based on real-time production data and workshop status, dynamic scheduling optimization is performed using the particle swarm optimization algorithm, with the objective function being: maxf(x) = w3*U + w4*E; Where U is the equipment utilization rate, E is the production efficiency, and w3 and w4 are weighting coefficients; Digital twin technology is used to simulate, verify, and optimize production scheduling schemes. The relevant indicators of different schemes are compared and analyzed to select the optimal scheme.

6. The GIS equipment manufacturing method based on a digital twin workshop according to claim 1, characterized in that: In step S3, real-time monitoring utilizes digital twin technology to map the workshop production status into a digital model in real time, enabling real-time monitoring of equipment operating status, material flow, product quality information, etc.; a visual interface for workshop production status is constructed, and data visualization technology is used to display the operating time and downtime of different equipment through bar charts, and to display the changing trends of product quality indicators through line charts.

7. A method for manufacturing GIS equipment based on a digital twin workshop according to claim 1, characterized in that: In step S3, the anomaly warning is based on historical data and the support vector machine (SVM) algorithm to construct a device fault prediction model; Assume the training sample set is (x i ,y i ), i = 1, 2, ..., n, where x i Given the input feature vector, y i Category labels; SVM separates normal samples from faulty samples by finding an optimal classification hyperplane. In the prediction phase, new equipment operation data is used as input feature vectors and fed into the trained SVM model to determine whether the equipment has a risk of failure. Digital twin technology is used to simulate and analyze the operating status of equipment. By establishing a digital model of the equipment, the operating conditions of the equipment under different working conditions are simulated, and the time and location of equipment failures can be predicted.

8. A method for manufacturing GIS equipment based on a digital twin workshop according to claim 1, characterized in that: In step S3, the quality prediction utilizes the Long Short-Term Memory (LSTM) network in the recurrent neural network (RNN) to construct a product quality prediction model. By introducing memory units and gating mechanisms, the long-term dependency problem in time series data can be effectively handled. Based on key parameters in the production process, the LSTM model predicts product quality indicators.

9. A method for manufacturing GIS equipment based on a digital twin workshop according to claim 1, characterized in that: In step S4, RFID technology is used to achieve real-time tracking and management of material information. During the material delivery process, the waiting time for materials is reduced by optimizing the delivery route. Based on equipment operation data and predictive models, a preventive maintenance plan is developed, and the remaining service life of the equipment is predicted by analyzing the equipment operation data. Digital twin technology is used to simulate and optimize the equipment maintenance process. Different maintenance schemes are simulated in the digital model, and the optimal maintenance scheme is selected to improve maintenance efficiency. By using data analysis and linear programming algorithms, we can optimize workshop energy consumption. By establishing an energy consumption model, we can analyze the energy consumption of different equipment and different production processes, identify energy waste, and formulate energy-saving measures.

Images