Intelligent factory production line energy efficiency data acquisition and analysis method and system

Abstract

The invention relates to the technical field of energy efficiency data acquisition and analysis oriented to factory management, in particular to an intelligent factory standardized production line energy efficiency data acquisition and analysis method and system, and the method comprises the following steps: a multi-source data perception interaction module captures a component pose and thermal parameters to construct a data set; the thermal energy efficiency dynamic analysis module calculates thermal demand and energy supply deviation based on a process stage; the logistics transmission power consumption evaluation module constructs a damping loss model and distinguishes effective work and abnormal loss, and the global energy efficiency optimization decision-making module executes an adaptive weighted fusion algorithm, solves an optimal working point and generates a closed-loop feedback energy-saving regulation and control strategy. The energy flow relation of production nodes is analyzed in real time through a high-frequency sensor and a dynamic thermodynamic model, hidden energy waste points are accurately positioned, process-level standardized energy efficiency indexes are formed, and energy consumption optimization, abnormity tracing and production line energy utilization rate improvement are achieved.

Description

Technical Field

[0001] This invention relates to the field of energy efficiency data acquisition and analysis technology for factory management, and in particular to a method and system for energy efficiency data acquisition and analysis suitable for standardized production lines in intelligent factories. Background Technology

[0002] Energy efficiency management and analysis technology involves the collection, monitoring, and in-depth analysis of energy consumption data in industrial production processes. Its aim is to achieve precise, traceable, and benchmarkable energy consumption management through digital means under factory-based, large-scale, and standardized production organization models, thereby optimizing energy allocation and reducing production costs. Traditional production line energy efficiency data collection and analysis systems rely on installing electricity or gas meters at the bus terminal to collect total energy consumption data for the entire production line and manually calculating the average unit consumption based on the quantity of finished products.

[0003] Traditional production line energy efficiency data acquisition and analysis systems focus on macro-level statistics of total energy consumption for the entire line. They lack spatiotemporal calibration and unified indicator standards for standardized process nodes, and also lack high-frequency correlation mapping between real-time physical conditions and instantaneous energy flow for independent production nodes. They cannot accurately identify hidden energy losses caused by uneven thermal distribution in independent kiln curing temperature zones or unloaded slippage of material transport equipment. This results in a serious lag in energy efficiency anomaly detection, leading to huge ineffective energy waste costs in factories under long-term high-load operation. This seriously restricts the substantial improvement of the level of refined and standardized management in the precast component production process. Summary of the Invention

[0004] The purpose of this invention is to address the problems of crude energy consumption statistics, difficulty in benchmarking node energy efficiency, and difficulty in timely locating abnormal energy consumption in factory management scenarios. It proposes an intelligent energy efficiency data acquisition and analysis method and system for factory production lines, which provides data and decision support for standardized energy efficiency assessment and intelligent closed-loop energy control while meeting product process quality constraints.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a method for energy efficiency data acquisition and analysis in an intelligent factory production line, comprising the following steps: The multi-source data perception and interaction module drives vision and sensing devices to capture the spatial pose coordinates of components, the thermal parameters of independent kiln temperature zones and the driving current of transmission equipment, and generates a full-element production operation status dataset based on a synchronous clock signal. The thermal efficiency dynamic analysis module identifies the heating, constant temperature and cooling process stages based on the spatial orientation coordinates of the components, inputs the full-element production operation status dataset into the three-dimensional thermodynamic distribution model for calculation, calculates the difference between the theoretical heat demand and the actual energy supply heat flux of each independent temperature zone, and generates a regional thermal efficiency deviation matrix. The logistics transmission power consumption assessment module constructs a mechanical transmission damping loss model based on the driving current of the transmission equipment and the spatial pose coordinates of the component. The mechanical transmission damping loss model is used to accurately distinguish between normal transmission work and abnormal slippage idling energy consumption, and to generate the logistics transmission mechanical energy efficiency ratio. The global energy efficiency optimization decision module maps the regional thermal efficiency deviation matrix and the energy efficiency ratio of logistics transmission machinery to a multi-dimensional energy efficiency evaluation space, executes an adaptive weighted fusion algorithm to solve for the optimal energy efficiency operating point of the production line, and generates a closed-loop feedback energy-saving control strategy in real time based on the dynamic energy efficiency benchmark range.

[0006] As a further aspect of the present invention, the process by which the multi-source data perception and interaction module generates the all-factor production operation status dataset specifically includes: A high-precision master clock synchronization signal is broadcast to the vision acquisition devices, thermocouple sensors and frequency converters distributed at various nodes of the production line to trigger the data synchronization sampling logic of multiple devices. Real-time acquisition of visual image streams containing the spatial pose coordinates of the component, temperature and pressure time-series data containing the thermal parameters of the independent kiln temperature zone, and electrical parameter waveforms containing the drive current of the transmission device. The multi-source heterogeneous data is time-aligned based on the timestamp of the synchronization clock signal, and the visual coordinates and physical sensor positions are spatially registered using spatial calibration parameters. The aligned data is then denoised and interpolated to generate the full-element production operation status dataset with a unified spatiotemporal label.

[0007] As a further aspect of the present invention, the process by which the thermal efficiency dynamic analysis module generates the regional thermal efficiency deviation matrix specifically includes: The spatial pose coordinates of the component are mapped to a preset digital twin topology map of the kiln to accurately determine the specific process temperature zone where the component is currently located and the corresponding heating, constant temperature or cooling state. The material thermal property parameters that match the current state are called, and combined with the environmental boundary conditions in the full-factor production operation status dataset, the three-dimensional thermodynamic distribution model is driven to perform unsteady-state heat transfer simulation to calculate the theoretical heat demand of each temperature zone. The heat flux supplied to the actual input temperature zone is calculated based on the thermal parameters of the independent kiln temperature zone, and then corrected by combining the combustion efficiency factor. The relative deviation between the theoretical heat demand and the actual heat flux supplied in each temperature zone is calculated one by one, and the regional thermal efficiency deviation matrix is ​​generated according to the physical arrangement order of the temperature zones.

[0008] As a further aspect of the present invention, the process by which the logistics transportation power consumption assessment module generates the logistics transportation mechanical energy efficiency ratio specifically includes: Spectral analysis is performed on the drive current of the transmission equipment to extract the load torque component, and the gravity component and running resistance of the transmission chain under the current path are calculated in combination with the spatial pose coordinates of the component. The mechanical transmission damping loss model is used to quantify the frictional loss work between the chain and the guide rail, as well as the energy consumption of abnormal slippage and idling caused by mechanical clearance, and to separate the effective transmission work for component displacement from the total input electrical energy. Obtain the total power consumption of the transmission system within the statistical period, and define the ratio of the effective transmission work to the total power consumption as the instantaneous energy efficiency index; The instantaneous energy efficiency indexes of multiple sampling periods are weighted and averaged to generate the energy efficiency ratio of the logistics transmission machinery, which can reflect the health of the transmission system.

[0009] As a further aspect of the present invention, the process by which the global energy efficiency optimization decision module generates the closed-loop feedback energy-saving control strategy specifically includes: The thermal deviation values ​​in the regional thermal efficiency deviation matrix and the mechanical energy efficiency ratio of the logistics transmission are normalized to eliminate dimensional differences and construct a system state feature vector. The state feature vector is projected onto the multidimensional energy efficiency evaluation space, and the Euclidean distance between the current operating point and the theoretical optimal region is calculated based on the preset energy efficiency reference surface. The adaptive weighted fusion algorithm is run to iteratively search for the combination of key control parameters that minimizes the Euclidean distance, while satisfying product process quality constraints. The optimal temperature setpoint and transmission speed command obtained from the solution are encoded into the closed-loop feedback energy-saving control strategy and sent to the underlying controller for execution.

[0010] As a further aspect of the present invention, the data denoising and interpolation process specifically includes: An adaptive median filtering algorithm is used to smooth out abrupt noise points in the thermal parameters of the independent kiln temperature zone, while preserving the true temperature fluctuation trend. Detect missing frames or abnormal jump segments in the spatial pose coordinate sequence of the component, and use cubic spline interpolation function to predict and fill in the missing pose data based on the motion vectors of the preceding and following frames; Wavelet transform threshold denoising is performed on the high-frequency noise component of the drive current of the transmission device to reconstruct the fundamental current signal that can truly reflect the load change. Verify the completeness and logical consistency of each data dimension after processing, and remove invalid samples that do not meet the confidence threshold.

[0011] As a further aspect of the present invention, the calculation formula for each element in the regional thermal efficiency deviation matrix is ​​as follows: ; in, Representing the The independent temperature zone in the first The deviation of thermal efficiency at any given time. This represents the theoretical heat demand obtained through model calculation. This represents the actual heat flux supplied, calculated based on monitoring parameters. This represents the preset energy conversion efficiency coefficient of the combustion system. This represents the standard deviation of the temperature range within the current time window. This represents the average process set temperature for this temperature range. The weighted penalty factor represents the impact of temperature stability on energy efficiency.

[0012] As a further aspect of the present invention, the formula for calculating the energy efficiency ratio of the logistics transmission machinery is as follows: ; in, This represents the calculated energy efficiency ratio. The value representing the traction force of the transmission chain. The real-time transmission linear velocity value representing the component. This represents the velocity-related frictional damping power dissipation value estimated based on the model. and These represent the real-time values ​​of the input line voltage and line current of the drive motor, respectively. Represents the power factor. This represents the time integration interval for the evaluation calculation.

[0013] As a further aspect of the present invention, the execution process of the adaptive weighted fusion algorithm specifically includes: Construct a multi-objective optimization function that includes the objectives of minimizing total system energy consumption and minimizing product quality fluctuations, and define the upper and lower limits of temperature and the transmission speed limit for each independent temperature zone as constraints. Based on the coupling relationship between energy efficiency and quality in historical operating data, the weighting coefficients of the regional thermal efficiency deviation matrix and the energy efficiency ratio of logistics transportation machinery in the optimization function are dynamically adjusted. An improved particle swarm optimization algorithm is adopted, using the current production line operating parameters as the initial population, and performing parallel search and iterative updates within the solution space defined by the constraints. When the rate of change of the population fitness value is lower than the preset convergence threshold, the parameter set corresponding to the global optimal particle position is output as the optimal energy efficiency operating point of the production line.

[0014] An intelligent factory production line energy efficiency data acquisition and analysis system is provided. The system is used to implement the aforementioned intelligent factory production line energy efficiency data acquisition and analysis method. The system includes: The multi-source data perception and interaction module is configured to drive vision and sensing devices to synchronously capture the spatial pose coordinates of components, the thermal parameters of independent kiln temperature zones, and the driving current of transmission equipment, and generate a full-element production operation status dataset. The thermal efficiency dynamic analysis module is configured to identify the process stage based on the spatial orientation coordinates of the component, calculate the theoretical heat demand using a three-dimensional thermodynamic distribution model and compare it with the actual energy supply heat flux to generate a regional thermal efficiency deviation matrix. The logistics transmission power consumption assessment module is configured to analyze the driving current of the transmission equipment and the spatial pose coordinates of the component using a mechanical transmission damping loss model, distinguish between effective work and abnormal loss, and generate the logistics transmission mechanical energy efficiency ratio. The global energy efficiency optimization decision module is configured to map energy efficiency indicators to a multi-dimensional energy efficiency evaluation space, use an adaptive weighted fusion algorithm to solve for the optimal operating point, and generate a closed-loop feedback energy-saving control strategy.

[0015] Compared with the prior art, the advantages and positive effects of the present invention are as follows: This invention constructs a full-link energy efficiency mapping mechanism based on spatiotemporal coupling, utilizing a high-frequency sensor array to capture the nonlinear relationship between the physical state of production nodes and energy flow data in real time. Combined with a dynamic thermodynamic distribution model and kinetic load curves, it accurately calculates the real-time energy efficiency deviation of each independent curing temperature zone and material transport section. This achieves millisecond-level precise location and quantitative assessment of hidden energy waste points throughout the production process, completely solving the problem that traditional extensive statistical methods cannot keenly perceive local energy efficiency bottlenecks. This significantly reduces ineffective energy consumption in the production process and significantly improves the overall energy utilization rate and operational stability of fully automated production lines. Simultaneously, this invention can form standardized energy efficiency indicators and dynamic benchmark ranges at the process node / temperature zone unit granularity, providing unified data support for workshop-level energy efficiency benchmarking, anomaly tracing, and closed-loop optimization, enhancing the visualization, controllability, and replicability of factory management. Attached Figure Description

[0016] Figure 1 This is the overall flowchart of the intelligent factory production line energy efficiency data acquisition and analysis method of the present invention; Figure 2 This is a flowchart illustrating the generation process of the full-element production operation status dataset for this invention. Figure 3 This is a flowchart illustrating the generation process of the regional thermal efficiency deviation matrix of this invention. Figure 4 This is a flowchart illustrating the process of generating the energy efficiency ratio of logistics transmission machinery according to the present invention. Figure 5 This is a flowchart illustrating the closed-loop feedback energy-saving control strategy generation process of the present invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the software-based technical solution is described in detail below with reference to system architecture diagrams and embodiments. It should be understood that the specific embodiments described herein are only for explaining the technical solutions of this invention and do not constitute a limitation on the scope of protection.

[0018] In the description of this invention, the system architecture relationships or data processing flows indicated by terms such as "layer," "module," "interface," "data flow," "client," and "server" are all defined based on the architecture diagram or flowchart corresponding to the embodiments. This way of describing is only used to clearly illustrate the logical relationships between the elements in the technical solution, and not to limit the physical deployment form. The term "multiple" includes two or more technical units, including but not limited to multiple data nodes, processing threads, service instances, or functional components and other scalable elements. The specific number is determined according to the actual business scenario and needs to be specifically specified.

[0019] Please see Figure 1 and Figure 2 The present invention provides a technical solution: A method for collecting and analyzing energy efficiency data in an intelligent factory production line includes the following steps: S1: Multi-source data perception and interaction module, which drives vision and sensing devices to capture the spatial pose coordinates of components, thermal parameters of independent kiln temperature zones and driving current of transmission equipment, and generates a full-element production operation status dataset based on synchronous clock signals. The process by which the multi-source data perception and interaction module generates a dataset of all-factor production operation status specifically includes: A high-precision master clock synchronization signal is broadcast to the vision acquisition devices, thermocouple sensors and frequency converters distributed at various nodes of the production line to trigger the data synchronization sampling logic of multiple devices. Real-time acquisition of visual image streams containing the spatial pose coordinates of components, time-series data of temperature and pressure containing the thermal parameters of independent kiln temperature zones, and waveforms of electrical parameters containing the drive current of transmission equipment. The time sequence of multi-source heterogeneous data is aligned based on the timestamp of the synchronization clock signal, and the visual coordinates are spatially registered with the physical sensor positions using spatial calibration parameters. The aligned data is then denoised and interpolated, and fused to generate a full-element production and operation status dataset with unified spatiotemporal labels. The data denoising and interpolation process specifically includes: An adaptive median filtering algorithm is used to smooth out abrupt noise points in the thermodynamic parameters of the independent kiln temperature zone, while preserving the true temperature fluctuation trend. Detect missing frames or anomalous jump segments in the spatial pose coordinate sequence of components, and use cubic spline interpolation function to predict and fill in the missing pose data based on the motion vectors of the preceding and following frames; Wavelet transform threshold denoising is performed on the high-frequency noise components of the drive current of the transmission equipment to reconstruct the fundamental current signal that can truly reflect load changes. Verify the completeness and logical consistency of each data dimension after processing, and remove invalid samples that do not meet the confidence threshold.

[0020] A master clock signal conforming to the IEEE 1588PTP protocol is sent to the Basleraceac A2500 industrial cameras, K-type armored thermocouples, and Yaskawa servo drives distributed at key nodes of the production line to calibrate the time deviation of all acquisition terminals to within 1 millisecond. The acquisition devices deployed at each monitoring point are activated to capture component transmission image streams with a resolution of 2048×1536 pixels at a frequency of 30 frames per second, read the temperature and pressure values ​​of each temperature zone of the kiln at a sampling rate of 10 Hz, and record the instantaneous waveform of the three-phase current of the transmission motor at a high-frequency sampling rate of 20 kHz. Based on the 64-bit nanosecond-level timestamp carried by the synchronization clock signal, the above heterogeneous data streams are aligned on the time axis. For spatial registration of visual coordinates and physical sensor positions, the pre-calibrated camera intrinsic matrix (focal length f=12 mm, principal point coordinates cx=1024, cy=768) and extrinsic rotation and translation matrix are read to transform the centroid position of the component in the image pixel coordinate system to the world coordinate system with the production line entrance as the origin, with the transformation error controlled within 5 mm. Differentiated processing is performed for the noise characteristics of different types of data. A data buffer with a window size of 5×1 is called to perform adaptive median filtering on thermal parameters. Replacement is only performed when the difference between the center pixel value and the median in the window exceeds 2 times the standard deviation, effectively filtering out spike pulses caused by electromagnetic interference. The component spatial pose coordinate sequence is traversed to identify missing segments with fewer than 10 consecutive frames. A cubic spline interpolation function is constructed using the known pose data of the previous and next 5 frames to calculate and fill in the coordinate values ​​of the missing time. Wavelet transform algorithm is applied to the motor drive current data. The Daubechies4 (db4) wavelet basis is selected for 4-level decomposition. A soft threshold rule is set to filter out high-frequency noise components and retain the low-frequency fundamental signal that reflects the load characteristics. Finally, the integrity of the cleaned data is verified. If the percentage of valid data is less than 95% within a certain time period, the sample is marked as invalid and removed. The final output is a full-element production operation status dataset containing timestamps, three-dimensional coordinates, temperature and pressure, and clean current waveforms.

[0021] The aforementioned IEEE 1588PTP protocol refers to a precision clock synchronization protocol for network measurement and control systems. It achieves sub-microsecond clock synchronization accuracy in an Ethernet environment by exchanging messages with precise timestamps between master and slave clocks, calculating and compensating for network transmission delays.

[0022] The cubic spline interpolation function mentioned above refers to an interpolation method that uses piecewise cubic polynomials to fit data points. Its characteristic is that not only are the function values ​​continuous at the interpolation nodes, but the first and second derivatives are also continuous, thereby ensuring that the generated motion trajectory is smooth and conforms to the laws of physical motion.

[0023] Table 1 lists the configuration parameters and preprocessing settings of the multi-source data acquisition equipment, providing a basic guarantee for subsequent analysis.

[0024] Table 1 Multi-source data acquisition configuration table

[0025] Please see Figure 1 and Figure 3 S2: Thermodynamic Energy Efficiency Dynamic Analysis Module. Based on the spatial orientation coordinates of the components, it identifies the heating, constant temperature and cooling process stages. It inputs the full-element production operation status dataset into the three-dimensional thermodynamic distribution model for calculation, calculates the difference between the theoretical heat demand and the actual energy supply heat flux of each independent temperature zone, and generates a regional thermal efficiency deviation matrix. The process by which the dynamic analysis module for thermal efficiency generates the regional thermal efficiency deviation matrix specifically includes: The spatial pose coordinates of the component are mapped to the preset digital twin topology map of the kiln to accurately determine the specific process temperature zone where the component is currently located and the corresponding heating, constant temperature or cooling state. The material thermal property parameters that match the current state are called, and the environmental boundary conditions in the full-factor production operation status dataset are combined to drive the three-dimensional thermodynamic distribution model to perform unsteady heat transfer simulation and calculate the theoretical heat demand of each temperature zone. The energy supply heat flux of the actual input temperature zone is calculated based on the thermal parameters of the independent kiln temperature zone, and then corrected by combining the combustion efficiency factor. Calculate the relative deviation between the theoretical heat demand and the actual heat flux of each temperature zone, and generate a regional thermal efficiency deviation matrix according to the physical arrangement of the temperature zones. The formulas for calculating each element in the regional thermal efficiency deviation matrix are as follows: ; in, Representing the The independent temperature zone in the first The deviation of thermal efficiency at any given time. This represents the theoretical heat demand obtained through model calculations. This represents the actual heat flux supplied based on monitoring parameters. This represents the preset energy conversion efficiency coefficient of the combustion system. This represents the standard deviation of the temperature range within the current time window. This represents the average process set temperature for this temperature range. The weighted penalty factor represents the impact of temperature stability on energy efficiency.

[0026] The real-time pose coordinates of the component in the world coordinate system Mapped to a pre-built digital twin model of the kiln, this model divides the 150-meter-long kiln into 30 independent physical control units, based on the current longitudinal coordinates of the components. The value determines the unit number it belongs to, and then identifies the current process stage as either the heating stage (0 to 50 meters), the isothermal stage (50 to 100 meters), or the cooling stage (100 to 150 meters). The specific heat capacity at the current temperature is obtained from the material thermophysical property database corresponding to the component. Thermal conductivity and density By combining the ambient temperature boundary conditions, the finite difference method is used to drive the iterative solution of the three-dimensional unsteady heat conduction equation. The spatial step size is set to 0.1 meters and the time step size is 1 second. The theoretical heat demand required for the component to reach the process set temperature curve within the target temperature range at the current moment is calculated. Simultaneously, the parameters of the natural gas flow meter and combustion fan in this temperature zone are read, the actual input chemical energy heat flux is calculated, and multiplied by the combustion efficiency factor of 0.92 to obtain the actual effective energy supply heat flux. .

[0027] The aforementioned finite difference method refers to discretizing the continuous solution domain into grid nodes, using Taylor series expansion to approximate the derivative terms in the partial differential equation as the difference quotients of the node function values, thereby transforming the differential equation into a system of algebraic equations for numerical solution.

[0028] The theoretical heat demand of the fifth independent temperature zone at time 100 seconds. (Value is 45,000 watts), actual energy supply heat flux (Value is 52,000 watts), Combustion system energy conversion efficiency coefficient (Value is 0.88), Temperature Standard Deviation of Temperature Zone (Value is 3.5 degrees Celsius), average process set temperature (Value is 1150 degrees Celsius) and temperature stability weighting penalty factor (The value is 2.5) Substitute it into the formula for calculating the deviation matrix of regional thermal efficiency. First, calculate the ratio of the difference between the actual effective heat supply and the theoretical heat demand, i.e. and product minus Take the absolute value and divide by The baseline deviation rate is obtained; then the penalty term due to temperature fluctuations is calculated, i.e. Multiply Divide by The exponential function value is then used; finally, the basic deviation rate is multiplied by the penalty term to calculate the thermal efficiency deviation value at that spatiotemporal node. The value is 0.017. This result indicates that under the current operating conditions, although there is a slight redundancy in the heat supply, the overall thermal efficiency deviation remains at a low level (less than 0.05) due to the small temperature fluctuations, falling within the high-efficiency operating range. Experimental data shows that compared to the traditional control method based solely on temperature deviation, introducing the deviation matrix reduces the overall energy consumption of the kiln by 8.5%.

[0029] Please see Figure 1 and Figure 4 S3: Logistics transmission power consumption assessment module. Based on the driving current of the transmission equipment and the spatial pose coordinates of the components, a mechanical transmission damping loss model is constructed. The mechanical transmission damping loss model is used to accurately distinguish between normal transmission work and abnormal slippage idling energy consumption, and to generate the mechanical energy efficiency ratio of logistics transmission. The process by which the logistics transmission power consumption assessment module generates the mechanical energy efficiency ratio for logistics transmission specifically includes: Spectral analysis of the drive current of the transmission equipment is performed to extract the load torque component, and the gravity component and running resistance of the transmission chain under the current path are calculated by combining the spatial orientation coordinates of the components. By using a mechanical transmission damping loss model to quantify the frictional loss work between the chain and the guide rail, as well as the energy consumption of abnormal slippage and idling caused by mechanical clearance, the effective transmission work for component displacement is separated from the total input electrical energy. Obtain the total power consumption of the transmission system within the statistical period, and define the ratio of effective transmission work to total power consumption as the instantaneous energy efficiency index; The instantaneous energy efficiency indicators of multiple sampling periods are weighted and averaged to generate the mechanical energy efficiency ratio of logistics transmission that can reflect the health of the transmission system; The formula for calculating the energy efficiency ratio of logistics transportation machinery is: ; in, This represents the calculated energy efficiency ratio. The value representing the traction force of the transmission chain. The real-time transmission linear velocity value representing the component. This represents the velocity-related frictional damping power dissipation value estimated based on the model. and These represent the real-time values ​​of the input line voltage and line current of the drive motor, respectively. Represents the power factor. This represents the time integration interval for the evaluation calculation.

[0030] A Fast Fourier Transform (FFT) is performed on the acquired stator current signal of the transmission motor to extract the amplitude of the fundamental component at a frequency of 50 Hz. High-order harmonics generated by the inverter switching frequency are filtered out, and the amplitude is then combined with the motor torque constant. Calculate the electromagnetic torque. Based on the real-time spatial coordinates of the components, calculate the slope angle of the transmission chain on the undulating road section. Project the total mass of the components and chain (valued at 5000 kg) onto the tangential direction to calculate the gravity component. Call the mechanical transmission damping loss model, which defines frictional resistance as a quadratic polynomial of velocity. , where the coefficient The current transmission speed was determined through no-load testing and calibration. The normal mechanical friction loss power is calculated. The total mechanical power generated by the electromagnetic torque is subtracted from the work done by gravity and normal friction loss; the remaining portion is determined to be the energy consumption due to abnormal slippage and idling caused by chain slack or guide rail wear. Energy consumption data for a complete production cycle (e.g., 1 hour) is statistically analyzed, and the effective work used for component displacement is compared with the total electrical energy input.

[0031] The aforementioned Fast Fourier Transform refers to an efficient algorithm for calculating the Discrete Fourier Transform (DFT) and its inverse transform. It can decompose a time-domain signal into sinusoidal components of different frequencies, thereby accurately extracting frequency domain features such as the fundamental amplitude.

[0032] Evaluation interval Set for 60 seconds, during which time the chain traction force is transmitted. The average value is 7050 Newtons, and the linear velocity of the component transmission is... Maintaining a speed of 0.15 m / s, the friction damping power dissipation value is estimated based on the model. 300 watts, drive motor input line voltage It is 380 volts, line current. It is 8.5 amperes, with a power factor of 1.5. The value is 0.82. Substituting the above parameters into the formula for calculating the energy efficiency ratio of logistics transportation machinery, we first calculate the integral of the numerator, that is, the time accumulation of effective mechanical power (traction force multiplied by speed minus friction loss). The calculation process is as follows: Next, calculate the integral of the denominator, which is the time-cumulative amount of the three-phase power (√3 multiplied by voltage, current, and power factor). The calculation process is as follows: Finally, the ratio of the two is calculated to obtain the energy efficiency ratio of logistics transportation machinery. The value is 0.165. This result indicates that the current transmission system has low mechanical efficiency (below the baseline value of 0.25), with significant mechanical stagnation or abnormal losses, suggesting the need for lubrication maintenance or tension adjustment of the transmission chain.

[0033] Please see Figure 1 and Figure 5 S4: Global Energy Efficiency Optimization Decision Module, which maps the regional thermal efficiency deviation matrix and the energy efficiency ratio of logistics transmission machinery to a multi-dimensional energy efficiency evaluation space, executes an adaptive weighted fusion algorithm to solve the optimal energy efficiency operating point of the production line, and generates a closed-loop feedback energy-saving control strategy in real time based on the dynamic energy efficiency benchmark range; The process by which the global energy efficiency optimization decision module generates a closed-loop feedback energy-saving control strategy specifically includes: The thermal deviation values ​​and the mechanical energy efficiency ratio of logistics transportation in the regional thermal efficiency deviation matrix are normalized to eliminate dimensional differences in order to construct a system state feature vector. The state feature vector is projected onto the multi-dimensional energy efficiency evaluation space, and the Euclidean distance between the current operating point and the theoretical optimal region is calculated based on the preset energy efficiency benchmark surface. The adaptive weighted fusion algorithm is run to iteratively search for the combination of key control parameters that minimizes the Euclidean distance, while satisfying the product process quality constraints. The optimal temperature setpoint and transmission speed command obtained from the solution are encoded into a closed-loop feedback energy-saving control strategy and sent to the underlying controller for execution. The execution process of the adaptive weighted fusion algorithm specifically includes: Construct a multi-objective optimization function that includes the objectives of minimizing total system energy consumption and minimizing product quality fluctuations, and define the upper and lower limits of temperature and the transmission speed limit for each independent temperature zone as constraints. Based on the coupling relationship between energy efficiency and quality in historical operating data, the weighting coefficients of the regional thermal efficiency deviation matrix and the energy efficiency ratio of logistics transportation machinery in the optimization function are dynamically adjusted. An improved particle swarm optimization algorithm is adopted, using the current production line operating parameters as the initial population, and performing parallel search and iterative updates within the solution space limited by constraints. When the rate of change of the population fitness value is lower than the preset convergence threshold, the parameter set corresponding to the global optimal particle position is output as the optimal energy efficiency operating point of the production line.

[0034] The generated regional thermal efficiency deviation matrix (30×1) and the energy efficiency ratio of logistics transportation machinery (scalar) are subjected to max-min normalization to map all indicators to the [0,1] interval, constructing a system state feature vector composed of 31 dimensions. In the multi-dimensional energy efficiency evaluation space, the theoretical optimal operating point is defined as the origin (i.e., the ideal state with thermal deviation of 0 and transmission energy efficiency ratio of 1). The Euclidean distance between the current state vector and the optimal operating point is calculated to quantify the overall energy efficiency level of the entire line. An adaptive weighted fusion algorithm is launched to construct a multi-objective optimization function. ,in This represents the total energy consumption of the system. This is the product quality loss function. The Particle Swarm Optimization (PSO) algorithm is initialized with a population size of 50 particles and a maximum iteration count of 100. Each particle represents a set of control parameters including the set temperature and transmission speed for each temperature zone. During the iteration process, the weights are dynamically adjusted based on the Pearson correlation coefficient between energy efficiency and quality in the historical database. When a quality index fluctuation greater than 2% is detected, the quality weights are automatically adjusted. The value was increased from 0.4 to 0.7. The algorithm performs optimization under the constraints of the upper and lower temperature limits (±10 degrees Celsius) and the transmission speed limit (0.1 to 0.3 m / s) of each temperature zone. It is considered to have converged when the rate of change of the fitness value of the population is less than 0.1% for 5 consecutive generations, and the corresponding optimal parameter set is output.

[0035] The aforementioned particle swarm optimization algorithm is an evolutionary computational technique that simulates the foraging behavior of bird flocks. It iteratively searches for the optimal solution of the objective function in a multidimensional search space by updating individual positions and sharing optimal information of the group.

[0036] The optimal parameter combination obtained from the solution (e.g., lowering the constant temperature zone setting from 1150 degrees Celsius to 1142 degrees Celsius, and increasing the transmission speed from 0.15 m / s to 0.16 m / s) is encoded into Modbus TCP instructions and directly sent to the PLC controller for closed-loop regulation. Experimental data shows that after applying this strategy, the overall energy efficiency ratio of the production line increased by 12.4%, while the product yield remained above 98.5%.

[0037] An intelligent factory production line energy efficiency data acquisition and analysis system is provided. This system is used to execute the aforementioned intelligent factory production line energy efficiency data acquisition and analysis method. The system includes: The multi-source data sensing and interaction module is configured to broadcast IEEE 1588PTP synchronization messages via a high-throughput industrial Ethernet interface, controlling 50 data acquisition nodes distributed on the production line to achieve sub-millisecond clock synchronization. Internally, this module integrates an image preprocessing engine based on a Xilinx Artix-7 FPGA, capable of parallel processing 2K resolution video streams from 10 Basler industrial cameras, and real-time calculation of the three-dimensional spatial pose of components. Simultaneously, its analog input channel quantizes the analog signals from K-type thermocouples and pressure sensors at 16-bit resolution, and digitizes the motor current signal at a 20kHz frequency via a high-speed A / D converter, ultimately constructing a full-element production operation status dataset with a unified UTC timestamp in memory.

[0038] The dynamic thermal efficiency analysis module is deployed on a computing server equipped with an Intel Xeon Gold processor and incorporates a kiln thermal field simulation engine based on computational fluid dynamics (CFD). This module is configured to read component pose data in real time, automatically match the corresponding temperature zone thermal model, and complete a single heat demand calculation within 0.5 seconds using finite element analysis (FEA). The module acquires real-time gas flow data from the on-site combustion controller via the OPCUA protocol, dynamically generates and updates a regional thermal efficiency deviation matrix in its internal memory. This matrix visually maps the energy efficiency mismatch of 30 independent temperature zones within the kiln, providing accurate spatial distribution characteristics for subsequent optimization.

[0039] The aforementioned OPCUA protocol refers to a cross-platform industrial automation communication standard designed to provide a secure, reliable, and vendor-independent data transmission mechanism, enabling seamless information integration between different devices and systems.

[0040] The logistics transmission power consumption assessment module integrates a TITMS320C6678 digital signal processing (DSP) chip, specifically designed for motor current characteristic analysis. This module is configured to perform a 1024-point Fast Fourier Transform on the current signal, separating the fundamental and higher harmonic components, and, combined with a pre-built library of mechanical dynamics models, calculates the friction coefficient and running resistance of the transmission chain in real time. Internally, the module runs an energy efficiency assessment algorithm that refreshes the logistics transmission machinery energy efficiency ratio index every 1 second. It also features an anomaly pattern recognition function, automatically triggering a mechanical fault warning signal when the energy efficiency ratio falls below a preset threshold of 0.15.

[0041] The global energy efficiency optimization decision-making module is equipped with an NVIDIA Tesla T4 AI inference accelerator card and runs a customized multi-objective particle swarm optimization algorithm. This module is configured to receive thermal and transport energy efficiency indicators in real time and perform a global search within a 31-dimensional state space. It incorporates an expert knowledge base and a self-learning algorithm, enabling it to adaptively adjust the weight coefficients of the optimization objective function based on the current production schedule and raw material properties. Through rapid iterative calculations, the module generates a closed-loop feedback energy-saving control strategy every 5 minutes, containing the optimal temperature setpoints and optimal transport speeds for each temperature zone, and sends the instructions to the underlying actuators via an industrial bus interface.

[0042] The above embodiments illustrate preferred embodiments of the present invention. Any equivalent adjustments to the technical solution based on software engineering methods are within the scope of protection, including but not limited to: implementing algorithm logic using different programming languages, refactoring functional modules into services, adjusting data interaction protocols, and optimizing resource scheduling strategies. Any implementation scheme derived from reasonable modifications to the data processing flow, service call chain, or system architecture layer without departing from the core technology of the present invention should be considered within the protection scope defined by the technical solution of the present invention.

Claims

1. A method for collecting and analyzing energy efficiency data in an intelligent factory production line, characterized in that, Includes the following steps: The multi-source data perception and interaction module drives vision and sensing devices to capture the spatial pose coordinates of components, the thermal parameters of independent kiln temperature zones and the driving current of transmission equipment, and generates a full-element production operation status dataset based on a synchronous clock signal. The thermal efficiency dynamic analysis module identifies the heating, constant temperature and cooling process stages based on the spatial orientation coordinates of the components, inputs the full-element production and operation status dataset into the three-dimensional thermodynamic distribution model for calculation, calculates the difference between the theoretical heat demand and the actual energy supply heat flux of each independent temperature zone, and generates a regional thermal efficiency deviation matrix. The logistics transmission power consumption assessment module constructs a mechanical transmission damping loss model based on the driving current of the transmission equipment and the spatial pose coordinates of the component. The mechanical transmission damping loss model is used to accurately distinguish between normal transmission work and abnormal slippage idling energy consumption, and to generate the logistics transmission mechanical energy efficiency ratio. The global energy efficiency optimization decision module maps the regional thermal efficiency deviation matrix and the energy efficiency ratio of logistics transmission machinery to a multi-dimensional energy efficiency evaluation space, executes an adaptive weighted fusion algorithm to solve for the optimal energy efficiency operating point of the production line, and generates a closed-loop feedback energy-saving control strategy in real time based on the dynamic energy efficiency benchmark range.

2. The method for collecting and analyzing energy efficiency data in an intelligent factory production line according to claim 1, characterized in that, The process by which the multi-source data perception and interaction module generates the all-factor production operation status dataset specifically includes: A high-precision master clock synchronization signal is broadcast to the vision acquisition devices, thermocouple sensors and frequency converters distributed at various nodes of the production line to trigger the data synchronization sampling logic of multiple devices. Real-time acquisition of visual image streams containing the spatial pose coordinates of the component, temperature and pressure time-series data containing the thermal parameters of the independent kiln temperature zone, and electrical parameter waveforms containing the drive current of the transmission device. The multi-source heterogeneous data is time-aligned based on the timestamp of the synchronization clock signal, and the visual coordinates and physical sensor positions are spatially registered using spatial calibration parameters. The aligned data is then denoised and interpolated to generate the full-element production operation status dataset with a unified spatiotemporal label.

3. The method for collecting and analyzing energy efficiency data in an intelligent factory production line according to claim 1, characterized in that, The process by which the dynamic analysis module for thermal efficiency generates the regional thermal efficiency deviation matrix specifically includes: The spatial pose coordinates of the component are mapped to a preset digital twin topology map of the kiln to accurately determine the specific process temperature zone where the component is currently located and the corresponding heating, constant temperature or cooling state. The material thermal property parameters that match the current state are called, and combined with the environmental boundary conditions in the full-factor production operation status dataset, the three-dimensional thermodynamic distribution model is driven to perform unsteady-state heat transfer simulation to calculate the theoretical heat demand of each temperature zone. The heat flux supplied to the actual input temperature zone is calculated based on the thermal parameters of the independent kiln temperature zone, and then corrected by combining the combustion efficiency factor. The relative deviation between the theoretical heat demand and the actual heat flux supplied in each temperature zone is calculated one by one, and the regional thermal efficiency deviation matrix is ​​generated according to the physical arrangement order of the temperature zones.

4. The method for collecting and analyzing energy efficiency data in an intelligent factory production line according to claim 1, characterized in that, The process by which the logistics transmission power consumption assessment module generates the logistics transmission mechanical energy efficiency ratio specifically includes: Spectral analysis is performed on the drive current of the transmission equipment to extract the load torque component, and the gravity component and running resistance of the transmission chain under the current path are calculated in combination with the spatial pose coordinates of the component. The mechanical transmission damping loss model is used to quantify the frictional loss work between the chain and the guide rail, as well as the energy consumption of abnormal slippage and idling caused by mechanical clearance, and to separate the effective transmission work for component displacement from the total input electrical energy. Obtain the total power consumption of the transmission system within the statistical period, and define the ratio of the effective transmission work to the total power consumption as the instantaneous energy efficiency index; The instantaneous energy efficiency indexes of multiple sampling periods are weighted and averaged to generate the energy efficiency ratio of the logistics transmission machinery, which can reflect the health of the transmission system.

5. The method for collecting and analyzing energy efficiency data in an intelligent factory production line according to claim 1, characterized in that, The process by which the global energy efficiency optimization decision module generates the closed-loop feedback energy-saving control strategy specifically includes: The thermal deviation values ​​in the regional thermal efficiency deviation matrix and the mechanical energy efficiency ratio of the logistics transmission are normalized to eliminate dimensional differences and construct a system state feature vector. The state feature vector is projected onto the multidimensional energy efficiency evaluation space, and the Euclidean distance between the current operating point and the theoretical optimal region is calculated based on the preset energy efficiency reference surface. The adaptive weighted fusion algorithm is run to iteratively search for the combination of key control parameters that minimizes the Euclidean distance, while satisfying product process quality constraints. The optimal temperature setpoint and transmission speed command obtained from the solution are encoded into the closed-loop feedback energy-saving control strategy and sent to the underlying controller for execution.

6. The method for collecting and analyzing energy efficiency data in an intelligent factory production line according to claim 2, characterized in that, The denoising and interpolation process for the data specifically includes: An adaptive median filtering algorithm is used to smooth out abrupt noise points in the thermal parameters of the independent kiln temperature zone, while preserving the true temperature fluctuation trend. Detect missing frames or abnormal jump segments in the spatial pose coordinate sequence of the component, and use cubic spline interpolation function to predict and fill in the missing pose data based on the motion vectors of the preceding and following frames; Wavelet transform threshold denoising is performed on the high-frequency noise component of the drive current of the transmission device to reconstruct the fundamental current signal that can truly reflect the load change. Verify the completeness and logical consistency of each data dimension after processing, and remove invalid samples that do not meet the confidence threshold.

7. The method for collecting and analyzing energy efficiency data in an intelligent factory production line according to claim 3, characterized in that, The formula for calculating each element in the regional thermal efficiency deviation matrix is ​​as follows: ； in, Representing the The independent temperature zone in the first The deviation of thermal efficiency at any given time. This represents the theoretical heat demand obtained through model calculation. This represents the actual heat flux supplied, calculated based on monitoring parameters. This represents the preset energy conversion efficiency coefficient of the combustion system. This represents the standard deviation of the temperature range within the current time window. This represents the average process set temperature for this temperature range. The weighted penalty factor represents the impact of temperature stability on energy efficiency.

8. The method for collecting and analyzing energy efficiency data in an intelligent factory production line according to claim 4, characterized in that, The formula for calculating the energy efficiency ratio of logistics transportation machinery is as follows: ； in, This represents the calculated energy efficiency ratio. The value representing the traction force of the transmission chain. The real-time transmission linear velocity value representing the component. This represents the velocity-related frictional damping power dissipation value estimated based on the model. and These represent the real-time values ​​of the input line voltage and line current of the drive motor, respectively. Represents the power factor. This represents the time integration interval for the evaluation calculation.

9. The method for collecting and analyzing energy efficiency data in an intelligent factory production line according to claim 5, characterized in that, The execution process of the adaptive weighted fusion algorithm specifically includes: Construct a multi-objective optimization function that includes the objectives of minimizing total system energy consumption and minimizing product quality fluctuations, and define the upper and lower limits of temperature and the transmission speed limit for each independent temperature zone as constraints. Based on the coupling relationship between energy efficiency and quality in historical operating data, the weighting coefficients of the regional thermal efficiency deviation matrix and the energy efficiency ratio of logistics transportation machinery in the optimization function are dynamically adjusted. An improved particle swarm optimization algorithm is adopted, using the current production line operating parameters as the initial population, and performing parallel search and iterative updates within the solution space defined by the constraints. When the rate of change of the population fitness value is lower than the preset convergence threshold, the parameter set corresponding to the global optimal particle position is output as the optimal energy efficiency operating point of the production line.

10. An intelligent factory production line energy efficiency data acquisition and analysis system, characterized in that, The system is used to implement the intelligent factory production line energy efficiency data acquisition and analysis method according to any one of claims 1-9, and the system includes: The multi-source data perception and interaction module is configured to drive vision and sensing devices to synchronously capture the spatial pose coordinates of components, the thermal parameters of independent kiln temperature zones, and the driving current of transmission equipment, and generate a full-element production operation status dataset. The thermal efficiency dynamic analysis module is configured to identify the process stage based on the spatial orientation coordinates of the component, calculate the theoretical heat demand using a three-dimensional thermodynamic distribution model and compare it with the actual energy supply heat flux to generate a regional thermal efficiency deviation matrix. The logistics transmission power consumption assessment module is configured to analyze the driving current of the transmission equipment and the spatial pose coordinates of the component using a mechanical transmission damping loss model, distinguish between effective work and abnormal loss, and generate the logistics transmission mechanical energy efficiency ratio. The global energy efficiency optimization decision module is configured to map energy efficiency indicators to a multi-dimensional energy efficiency evaluation space, use an adaptive weighted fusion algorithm to solve for the optimal operating point, and generate a closed-loop feedback energy-saving control strategy.

Images