An energy consumption monitoring and analyzing system for a smart factory

Abstract

The present application belongs to the technical field of energy consumption monitoring and analysis, and specifically relates to an energy consumption monitoring and analysis system for a smart factory, comprising an energy data acquisition module for acquiring real-time energy consumption data and energy medium fluctuation data; an energy consumption influence analysis module for obtaining production load energy consumption data, extracting energy consumption parameter change data and energy quality data; an energy consumption prediction control module for formulating a basic energy consumption control strategy, predicting energy consumption changes to obtain energy consumption prediction data; and an energy consumption strategy optimization module for formulating an energy consumption dynamic adjustment strategy according to the production load energy consumption data and the energy consumption prediction data, and adjusting the basic energy consumption control strategy in combination with factory production scheduling requirements and the energy quality data to obtain an optimized energy consumption control scheme; the present application can improve energy management efficiency, reduce energy consumption costs, and help enterprises achieve energy saving and carbon reduction and efficient production.

Description

Technical Field

[0001] This invention belongs to the field of energy consumption monitoring and analysis technology, specifically an energy consumption monitoring and analysis system for smart factories. Background Technology

[0002] In the process of industrial digital transformation, factory energy consumption accounts for a very high proportion of total industrial energy consumption. The level of refinement and intelligence in energy management directly affects enterprises' production costs and the achievement of dual-carbon goals. Currently, traditional factory energy management systems generally suffer from many technical bottlenecks and are no longer suitable for the efficient operation requirements of smart factories.

[0003] First, multi-source heterogeneous data is difficult to process uniformly. The equipment used for collecting data from various energy sources within the factory, such as electricity, water, gas, steam, and compressed air, varies in model and communication protocols are incompatible, leading to data fragmentation and silos. This makes unified integration and standardized management of all energy types impossible, resulting in extremely low data reliability and utilization. Second, there is a lack of global coordination and collaborative optimization across multiple energy sources. Existing systems primarily monitor individual energy sources independently, lacking a multi-energy flow collaborative calculation mechanism. This makes it impossible to quantify the impact of various energy sources on production equipment and hinders the overall allocation of energy across the plant. Third, energy consumption management lacks foresight. Traditional management models rely heavily on post-event statistics and passive alarms, failing to employ scientific time-series forecasting methods. This makes it impossible to predict energy consumption trends in advance and avoid energy waste and supply-demand imbalances. Furthermore, the dynamic mismatch between energy supply and demand is prominent. Control strategies do not consider production scheduling needs and differences in energy quality, resulting in insufficient energy supply for high-priority processes, unreasonable energy resource allocation, and low energy utilization efficiency. Simultaneously, energy waste is widespread in industrial production, and existing systems struggle to accurately identify the locations and root causes of energy losses.

[0004] In summary, there is an urgent need for an energy consumption monitoring and analysis system for smart factories to address the pain points of existing technologies, improve energy management efficiency, reduce energy costs, and help enterprises achieve energy conservation, carbon reduction, and efficient production. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, this invention proposes an energy consumption monitoring and analysis system for smart factories. This invention primarily addresses the problems in traditional factory energy management systems, such as the difficulty in unifying multi-source heterogeneous data, the lack of global coordination and collaborative optimization across multiple energy media, the lack of foresight in energy consumption control, and the dynamic mismatch between energy supply and demand.

[0006] The technical solution adopted by this invention to solve its technical problem is: This invention provides an energy consumption monitoring and analysis system for smart factories, comprising:

[0007] The energy data acquisition module is used to collect data on energy media within the factory and monitor the energy consumption of each energy-consuming department in real time to obtain real-time energy consumption data and energy media fluctuation data.

[0008] The energy consumption impact analysis module is used to analyze the impact of energy medium fluctuation data on the overall production equipment of the factory based on the multi-energy collaborative calculation method to obtain production load energy consumption data, and extract energy consumption parameter change data and energy quality data from real-time energy consumption data.

[0009] The energy consumption prediction and control module is used to formulate basic energy consumption control strategies based on changes in energy consumption parameters. It obtains energy consumption prediction data by predicting energy consumption changes based on changes in energy consumption parameters using the energy consumption time series prediction method.

[0010] The energy consumption strategy optimization module is used to formulate dynamic energy consumption adjustment strategies based on production load energy consumption data and energy consumption forecast data, and to adjust the basic energy consumption control strategy in conjunction with factory production scheduling requirements and energy quality data to obtain an optimized energy consumption control scheme.

[0011] The present invention provides an energy consumption monitoring and analysis system for smart factories. The steps for the energy data acquisition module to obtain energy medium fluctuation data include:

[0012] Based on their energy properties, factory energy media are classified into five categories: electrical energy, water energy, coal gas energy, steam energy, and compressed air energy.

[0013] Data on power distribution circuit load, voltage, current, harmonics, and time-of-use electricity consumption are collected for electrical energy. Data on water inflow and outflow rates and water pressure loss in each workshop are collected for water energy.

[0014] Data on gas supply flow rate, pressure, concentration, and consumption rate in the gas pipeline network are collected for coal gas energy; data on transmission temperature, pressure, flow rate, and heat loss are collected for steam energy.

[0015] Data on compressed air energy collection pipeline pressure, flow rate, leakage, and equipment gas load are collected, and multi-source energy consumption data are normalized and anomaly screening is performed to obtain energy medium fluctuation data.

[0016] The present invention provides an energy consumption monitoring and analysis system for smart factories, wherein the steps for the energy data acquisition module to obtain real-time energy consumption data include:

[0017] By using an edge computing gateway compatible with multiple industrial communication protocols, energy medium fluctuation data can be preprocessed locally, converted according to protocols, and resumed from where it was interrupted, to obtain preliminary real-time data.

[0018] Outlier detection and cleaning are performed on the initial real-time data to remove noise data caused by equipment failure and network fluctuations, thus obtaining real-time energy consumption data.

[0019] This invention provides an energy consumption monitoring and analysis system for smart factories, the energy consumption impact analysis module of which includes:

[0020] The energy consumption characteristic analysis unit is used to determine the energy consumption characteristics of the corresponding energy type based on the physical transmission and consumption characteristics of various energy media, collect the operating parameters of factory production equipment, and analyze the energy consumption and energy consumption fluctuation of different production processes to obtain the energy consumption characteristics of the equipment.

[0021] The energy consumption impact assessment unit is used to establish an energy consumption factor analysis model based on energy consumption characteristics and equipment energy consumption characteristics, analyze the quantitative relationship between energy parameters and equipment operating performance, determine the impact weight of each parameter in the energy parameters on equipment energy consumption, and calculate the degree of impact of different types of energy media on the actual energy consumption of factory equipment.

[0022] The load coordination calculation unit is used to establish a plant-wide energy balance model to calculate the total comprehensive energy flow value, and to calculate the production load energy consumption data under the coordinated management of multiple energy media by combining the production line design production load and the actual energy consumption impact.

[0023] The present invention provides an energy consumption monitoring and analysis system for smart factories, wherein the steps for the load coordination calculation unit to calculate the total comprehensive energy flow include:

[0024] The energy input items are determined based on the electricity input required for factory production, fuel consumption, purchased steam, and the energy carried by the medium itself.

[0025] Energy output items are determined based on energy consumption in the production of finished products, energy consumption in process by-products, and energy loss from end-of-pipe emissions.

[0026] The total comprehensive energy flow value is calculated based on energy input and energy output items, combined with pipeline transmission losses, equipment no-load losses, and production condition fluctuation losses.

[0027] The present invention provides an energy consumption monitoring and analysis system for smart factories. The steps of the energy consumption impact analysis module to obtain energy consumption parameter change data and energy quality data include:

[0028] By analyzing the production load energy consumption data of each energy-consuming node, the difference in electricity load between peak and off-peak periods is calculated to obtain load change data. In addition, the inlet and outlet flow data of the energy pipeline network are statistically analyzed to calculate the flow fluctuation of various media and obtain flow parameter change data.

[0029] Pressure parameter change data is obtained by calculating the pressure loss and dynamic change range of the pipeline network and energy-consuming equipment through pressure monitoring data of energy transmission pipeline network and energy-consuming equipment. This data is then combined with load change data and flow parameter change data to form energy consumption parameter change data.

[0030] Based on load change data, energy medium parameters, and specific heat capacity, the total energy consumption and unit product consumption are calculated to obtain the consumption quality, and the temperature quality for stable energy supply is determined.

[0031] By analyzing the changing trends of energy consumption parameters with production time, the stability of energy supply and demand is assessed, and energy quality data is formed together with consumption quality and temperature quality.

[0032] This invention provides an energy consumption monitoring and analysis system for smart factories. The steps of the energy consumption prediction and control module in formulating basic energy consumption control strategies include:

[0033] Based on the factory's energy conservation and emission reduction targets and dual carbon control indicators, the control scale of energy recovery and intensive utilization is determined to obtain the energy control target. The production energy demand is obtained by analyzing the factory's production scheduling plan and the energy demand of each process for water, electricity, gas, steam, and compressed air.

[0034] By identifying the type and magnitude of energy consumption parameter fluctuations through load change data and flow change data, and extracting the time-series characteristics of energy consumption parameter change data.

[0035] Energy supply and demand matching strategies are formulated based on energy management objectives and production energy demand. Energy consumption parameter fluctuation scenarios are generated according to the type, magnitude and time series characteristics of parameter fluctuations, and key management factors for different scenarios are identified.

[0036] Based on the fluctuation scenarios of energy consumption parameters, select the corresponding energy consumption control and energy-saving adjustment methods, and adjust them according to the degree of influence of key control factors on the energy supply and demand matching strategy to obtain the basic energy consumption control strategy.

[0037] The present invention provides an energy consumption monitoring and analysis system for smart factories. The steps for the energy consumption prediction and control module to generate energy consumption prediction data include:

[0038] A time-series energy consumption prediction model based on random forest is established. The changing patterns and characteristics of energy consumption parameters in each production cycle are analyzed to obtain predicted load, predicted consumption, and predicted pipeline pressure. Production condition characteristics, environmental climate characteristics, and time-series cycle characteristics are extracted from the energy consumption parameter change data.

[0039] Using production operating conditions, environmental climate characteristics, and time-series cycle characteristics as inputs, and predicted load, predicted consumption, and predicted pipeline pressure as outputs, the energy consumption time-series prediction model is trained, and model parameters that meet the test accuracy are retained.

[0040] Based on the current production conditions, current environmental climate, and current time series, extract the corresponding real-time load, real-time consumption, and real-time pipeline pressure, and input them into the trained energy consumption time series prediction model to obtain predicted values ​​of energy consumption parameters.

[0041] The energy transmission loss rate is calculated based on pipeline fluid dynamics, and the natural loss rate of the medium is obtained by combining the loss rate of compressed air and steam with the ambient temperature and humidity.

[0042] Energy consumption prediction data is calculated by combining predicted energy consumption parameters, energy transmission loss rate, and natural loss rate of the medium.

[0043] This invention provides an energy consumption monitoring and analysis system for smart factories. The steps of the energy consumption strategy optimization module in formulating a dynamic energy consumption adjustment strategy include:

[0044] Energy consumption prediction data and production load energy consumption data are converted into a unified time series granularity through interpolation algorithms, and abnormal energy consumption values ​​are removed and smoothed.

[0045] A two-dimensional supply-demand matching matrix is ​​established by describing the matching relationship between different production periods and energy supply grades.

[0046] Calculate the energy supply and demand gap for each production period, classify the gaps according to their severity, and establish a multi-objective energy efficiency optimization model with energy utilization efficiency, production and operation economic costs, and plant carbon emissions as optimization objectives.

[0047] The optimal energy consumption value is obtained by iteratively transitioning the state of the multi-objective energy efficiency optimization model using dynamic programming algorithm, and a graded load response mechanism is formulated in combination with different levels of supply and demand gap to obtain a dynamic energy consumption adjustment strategy.

[0048] This invention provides an energy consumption monitoring and analysis system for smart factories, comprising:

[0049] The steps by which the energy consumption strategy optimization module obtains an optimized energy consumption management scheme include:

[0050] The energy needs of each production process and energy-consuming department in the factory are classified according to their urgency and importance. The energy needs are then prioritized by combining the annual and monthly production schedules to estimate the scale and duration of various energy needs at different times.

[0051] By combining energy quality data and energy consumption forecast data, the total amount, spatiotemporal distribution and supply and demand fluctuation characteristics of various energy resources in the whole plant are assessed, and the supply and demand matching assessment results are obtained by combining the degree of matching with the energy demand of each department.

[0052] Based on the priority ranking of energy demand and the assessment results of supply and demand matching, the basic energy consumption management strategy is adjusted, energy resources are dynamically reallocated, and corresponding energy-saving control and scheduling methods are determined for different energy quality levels and different energy demand to obtain an optimized energy consumption management plan.

[0053] The beneficial effects of this invention are as follows:

[0054] 1. This invention leverages edge gateway multi-protocol adaptation, breakpoint resumption, and anomaly cleaning and smoothing to effectively resist industrial electromagnetic interference and network fluctuations, avoiding data loss, jumps, and distortions, thus improving energy consumption data quality and acquisition stability. It accurately quantifies the weight and degree of influence of various energy parameters on equipment energy consumption, clarifying the root causes of energy consumption fluctuations. Through unified calculation of energy input, output, and multiple types of losses, it accurately calculates the total comprehensive energy flow, clearly locating waste points such as pipeline losses, equipment idle losses, and operating condition fluctuation losses, providing direction for energy-saving renovations. It breaks down parameter changes from load, flow, and pressure, and assesses energy quality from consumption, temperature, and supply-demand stability, achieving refined energy consumption analysis. Based on random forest multi-feature modeling and dual correction of pipeline losses and environmental losses, energy consumption time-series prediction is more closely aligned with actual site conditions, enabling early prediction of peak and valley loads and supply-demand changes, supporting proactive scheduling. Through multi-objective energy efficiency optimization models and dynamic programming algorithms, it improves energy utilization efficiency while reducing production energy operating costs and factory carbon emissions, meeting the factory's energy conservation, cost reduction, and dual-carbon management needs. Differentiated control methods are adopted to match different energy qualities, so as to comprehensively improve the overall energy utilization rate of the factory and reduce the overall energy consumption and energy cost. Attached Figure Description

[0055] The invention will now be further described with reference to the accompanying drawings.

[0056] Figure 1 This is a schematic diagram of a smart factory energy consumption monitoring and analysis system provided in an embodiment of the present invention;

[0057] Figure 2 This is a schematic diagram of the process for obtaining energy quality data in an energy consumption monitoring and analysis system for a smart factory, provided by an embodiment of the present invention.

[0058] Figure 3 This is a flowchart illustrating the process of obtaining dynamic energy consumption adjustment strategies in an energy consumption monitoring and analysis system for a smart factory, as provided in an embodiment of the present invention. Detailed Implementation

[0059] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0060] like Figures 1 to 3 As shown in the figure, an energy consumption monitoring and analysis system for a smart factory is provided in an embodiment of the present invention. The system includes:

[0061] The energy data acquisition module is used to collect data on energy media within the factory and monitor the energy consumption of each energy-consuming department in real time to obtain real-time energy consumption data and energy media fluctuation data.

[0062] The steps by which the energy data acquisition module obtains energy medium fluctuation data include:

[0063] Based on their energy properties, factory energy media are classified into five categories: electrical energy, water energy, coal gas energy, steam energy, and compressed air energy.

[0064] Data on load, voltage, current, harmonics, and time-of-use electricity consumption in power distribution circuits are collected. Data on inflow and outflow water and water pressure loss in each workshop are collected for water energy. Intelligent multi-functional power meters, power quality analyzers, and edge acquisition terminals are deployed in the plant's main power distribution room, workshop distribution cabinets, and independent circuits of key production equipment. Modbus-RTU and OPCUA industrial communication protocols are used to access the power distribution system PLC in real time. Raw electrical parameters such as instantaneous load, three-phase voltage, three-phase current, total harmonic distortion rate, and harmonic components are obtained through high-frequency sampling at the second level. Combined with peak-valley-flat time-of-use pricing rules, daily / shift / monthly time-of-use electricity consumption is automatically calculated, and equipment start / stop status and production condition tags are synchronously associated to complete the power source data association and tagging.

[0065] Electromagnetic flow meters, pressure transmitters, and remote water meters are installed in the main water supply network of the plant area, the water inlet branch pipelines of each workshop, and the production circulating return water pipeline. Real-time instantaneous flow rate of incoming water, cumulative water consumption, and return water flow rate are collected using pulse signals and 485 bus transmission. By comparing the difference between the water supply pressure at the front end of the pipeline and the water pressure at the end, the system automatically calculates water pressure loss data such as pipeline friction loss and local throttling loss.

[0066] Data on gas supply flow rate, pressure, concentration, and consumption rate in the gas pipeline network are collected for coal gas energy; data on transmission temperature, pressure, flow rate, and heat loss are collected for steam energy.

[0067] Gas turbine flow meters, intelligent pressure sensors, and gas concentration detectors are installed at the main gas inlet, workshop distribution network, and front-end gas-consuming equipment such as kilns. Industrial LoRa and industrial Ethernet are used to achieve stable long-distance data transmission. Real-time data collection of instantaneous gas supply flow, static and dynamic pressure, and combustible gas concentration thresholds in the pipeline network is performed. The real-time gas consumption rate is calculated by converting the cumulative consumption within a unit time period. At the same time, abnormal triggering conditions such as concentration exceeding limits and sudden pressure drop are set to complete the real-time marking of the original gas collection data.

[0068] Vortex flow meters, high-temperature pressure transmitters, and platinum resistance temperature sensors are installed at the main steam transmission pipeline, the inlet of the steam distribution cylinder, and the heat exchange front end of the heat-using equipment. Real-time data on real-time steam transmission temperature, pipeline operating pressure, instantaneous and cumulative steam flow are collected. Combined with the steam enthalpy formula, pipeline insulation parameters and transmission distance, the heat loss data caused by pipeline heat dissipation and medium phase change are calculated in real time. The system can also distinguish between saturated steam and superheated steam types and adapt to the steam energy consumption data calibration rules under different operating conditions.

[0069] Data on compressed air energy collection pipeline pressure, flow rate, leakage, and equipment gas load are collected, and multi-source energy consumption data are normalized and anomaly screening is performed to obtain energy medium fluctuation data.

[0070] First, the heterogeneous data collected from five energy sources underwent unified dimension conversion, timestamp alignment, and sampling frequency normalization to eliminate data format differences caused by different instruments and communication protocols. Then, the Laida criterion and box plot analysis were used to identify outliers, eliminating outliers caused by data jumps, communication packet loss, and sensor malfunctions, and completing missing data through time-series interpolation. Finally, the parameter fluctuations and supply-demand volatility characteristics of various energy media were aggregated and analyzed across multiple time dimensions, including daily, hourly, and shift-based approaches, generating standardized and structured energy media volatility data.

[0071] The steps by which the energy data acquisition module obtains real-time energy consumption data include:

[0072] By using an edge computing gateway compatible with multiple industrial communication protocols, energy medium fluctuation data can be preprocessed locally, converted according to protocols, and resumed from where it was interrupted, to obtain preliminary real-time data.

[0073] Built-in drivers for mainstream industrial communication protocols and proprietary protocols enable automatic identification of communication interface types for various metering instruments and sensors used in the plant for electricity, water, gas, steam, and compressed air. This allows for seamless access and protocol compatibility for heterogeneous devices from multiple brands and models. Energy medium fluctuation data undergoes local preprocessing, protocol conversion, and breakpoint resumption. The edge gateway performs on-site parsing of the raw data format, byte order regularization, data point mapping, and variable classification, unifying fragmented energy medium fluctuation data under different protocols into standard industrial data packets. Simultaneously, a built-in protocol conversion engine automatically converts non-standard proprietary protocols into the platform-recognizable standard MQTT / OPCUA universal protocol format. A local flash storage module caches collected data in real time and automatically activates when the industrial network is interrupted or experiences link jitter.

[0074] Outlier detection and cleaning are performed on the initial real-time data to remove noise data caused by equipment failure and network fluctuations, thus obtaining real-time energy consumption data.

[0075] A two-layer anomaly detection rule is constructed using a time-series sliding window algorithm combined with the Laida 3σ criterion and box plot quartile method. This rule performs a full-domain screening of preliminary real-time data from three dimensions: amplitude abrupt changes, time-series jumps, and long-term constant invariance. Noisy data caused by equipment failure and network fluctuations is removed. Precise identification and removal are performed for numerical drift anomalies caused by sensor aging and drift, metering instrument failure, strong electromagnetic interference in the workshop, and pulse noise and abrupt outliers caused by network packet loss and instantaneous signal interruptions. Short-term missing energy consumption time-series data is repaired using linear interpolation and nearest-neighbor mean completion methods. Continuous long-term abnormal periods are tagged with abnormal operating conditions and the original logs are retained. Simultaneously, data timestamp alignment, unit unification, sampling frequency regularization, and redundant and duplicate data are eliminated.

[0076] The energy consumption impact analysis module is used to analyze the impact of energy medium fluctuation data on the overall production equipment of the factory based on the multi-energy collaborative calculation method to obtain production load energy consumption data, and extract energy consumption parameter change data and energy quality data from real-time energy consumption data.

[0077] The energy consumption impact analysis module includes:

[0078] The energy consumption characteristic analysis unit is used to determine the energy consumption characteristics of the corresponding energy type based on the physical transmission and consumption characteristics of various energy media, collect the operating parameters of factory production equipment, and analyze the energy consumption and energy consumption fluctuation of different production processes to obtain the energy consumption characteristics of the equipment.

[0079] By retrieving real-time energy consumption data and energy medium fluctuation data that have been collected and preprocessed in the early stage through the energy consumption data platform, and combining the inherent physical characteristics of various energy media, such as voltage level and frequency characteristics of electricity, temperature, pressure and enthalpy parameters of steam, calorific value and density of coal gas, specific heat capacity and water quality parameters of water, pressure level and dew point characteristics of compressed air, and other attributes of the transportation process, such as pipeline material, transportation distance, insulation / corrosion protection method, leakage loss benchmark, and production consumption and usage attributes, such as energy use scenario, consumption rate, linkage with production process, and start-stop response characteristics, a classified energy consumption characteristic database is established to clarify the energy consumption benchmark threshold, fluctuation range, loss law and suitable energy use scenario for each type of energy medium.

[0080] By connecting to the factory's MES system, PLC control system, and equipment operation and maintenance management system via industrial Ethernet, the system collects real-time operating parameters of various production equipment, including core data such as rated power, actual operating load, start-up and shutdown status, process parameters, runtime, and maintenance records. Simultaneously, it categorizes data by production process, linking energy consumption data points corresponding to each process, statistically analyzing the energy consumption and consumption rate of different processes under various production conditions, analyzing the fluctuation patterns of energy consumption with process progress, equipment operating status, and production cycle time, identifying key characteristics such as energy consumption differences between processes, equipment no-load energy consumption, and energy consumption fluctuations caused by process adjustments, and combining comparative analysis of equipment rated energy consumption standards and actual energy consumption data to extract and classify equipment energy consumption characteristics.

[0081] The energy consumption impact assessment unit is used to establish an energy consumption factor analysis model based on energy consumption characteristics and equipment energy consumption characteristics, analyze the quantitative relationship between energy parameters and equipment operating performance, determine the impact weight of each parameter in the energy parameters on equipment energy consumption, and calculate the degree of impact of different types of energy media on the actual energy consumption of factory equipment.

[0082] An energy consumption factor analysis model is constructed based on the fusion of multiple linear regression and random forest algorithms. Key parameters of energy consumption characteristics, such as the pressure, flow rate, temperature, and loss rate of the energy medium, are used as input variables. Core indicators of equipment energy consumption characteristics, such as actual energy consumption, operating efficiency, and energy consumption fluctuation coefficient, are used as output variables. The model is trained, iterated, and validated using sample data including historical energy consumption data from the factory, equipment operation logs, and production process parameters. Model parameters are optimized to improve analytical accuracy and ensure that the model can accurately fit the correlation between energy parameters and equipment operating performance. The quantitative relationship between energy parameters and equipment operating performance is analyzed. Through the variable correlation coefficients output by the model, the influence of various energy parameters on equipment operating performance is quantitatively analyzed, clarifying the positive and negative correlations between each energy parameter and equipment energy efficiency and energy cost, and generating a quantitative correlation report of energy parameters and equipment performance. The impact weights of each energy parameter on equipment energy consumption were determined. The Analytic Hierarchy Process (AHP) combined with the entropy weight method was used, comprehensively considering factors such as production priority, energy cost, and equipment operation and maintenance requirements. The impact of each energy parameter was weighted and calculated, eliminating redundant parameters with weak influence, and identifying the weight proportions of core energy parameters, such as electricity load, steam pressure, and gas concentration, thus forming an energy parameter impact weight matrix. The actual energy consumption of different types of energy media on factory equipment was calculated. Combining the energy consumption characteristics of various energy media, equipment energy consumption characteristics, and energy parameter impact weights, an energy consumption impact assessment index system was established. Assessment indicators such as energy consumption impact factor, equipment efficiency decay coefficient, and energy cost proportion were introduced to quantify the actual energy consumption of five types of energy media—electricity, water, gas, steam, and compressed air—on different production equipment and different production processes. Core influencing energy and auxiliary influencing energy were distinguished, and the type of energy media with the greatest impact on equipment operating efficiency and total energy consumption was identified.

[0083] The load coordination calculation unit is used to establish a plant-wide energy balance model to calculate the total comprehensive energy flow value, and to calculate the production load energy consumption data under the coordinated management of multiple energy media by combining the production line design production load and the actual energy consumption impact.

[0084] The steps for the load coordination calculation unit to calculate the total integrated energy flow include:

[0085] The energy input items are determined based on the electricity input required for factory production, fuel consumption, purchased steam, and the energy carried by the medium itself.

[0086] Energy output items are determined based on energy consumption in the production of finished products, energy consumption in process by-products, and energy loss from end-of-pipe emissions.

[0087] The total comprehensive energy flow value is calculated based on energy input and energy output items, combined with pipeline transmission losses, equipment no-load losses, and production condition fluctuation losses.

[0088] A plant-wide energy balance model is established, clarifying the accounting logic and calculation formulas: Total comprehensive energy flow = Sum of energy inputs - Sum of energy outputs - Sum of all types of losses. Pipeline transmission losses are calculated by retrieving energy medium fluctuation data and pipeline operating parameters, such as pipeline material, transmission distance, and insulation coefficient, combined with the transmission characteristics of various energy media. Power line losses, steam pipeline heat dissipation losses, gas / compressed air pipeline leakage losses, and water pipeline losses are calculated separately, and the total pipeline transmission loss is obtained. Equipment no-load losses are calculated by collecting no-load operating parameters of various production equipment and combining them with equipment rated parameters. The energy consumption losses of equipment in standby and no-load states are calculated, distinguishing the differences in no-load losses among different equipment types, and the total equipment no-load loss is obtained. Production condition fluctuation losses are calculated by analyzing production load fluctuation data and process adjustment records, identifying operating conditions such as production changeovers, frequent start-ups and shutdowns, and sudden load changes. Combined with historical energy consumption data, the additional energy consumption losses caused by operating condition fluctuations are calculated, including energy waste during load adjustments and energy increases due to process parameter fluctuations, and the total production condition fluctuation losses are obtained. Finally, the total energy input is subtracted from the total energy output, and then the sum of pipeline transmission losses, equipment no-load losses, and production condition fluctuation losses is subtracted to calculate the total comprehensive energy flow value.

[0089] The steps for the energy consumption impact analysis module to obtain energy consumption parameter change data and energy quality data include:

[0090] By analyzing the production load energy consumption data of each energy-consuming node, the difference in electricity load between peak and off-peak periods is calculated to obtain load change data. In addition, the inlet and outlet flow data of the energy pipeline network are statistically analyzed to calculate the flow fluctuation of various media and obtain flow parameter change data.

[0091] Pressure parameter change data is obtained by calculating the pressure loss and dynamic change range of the pipeline network and energy-consuming equipment through pressure monitoring data of energy transmission pipeline network and energy-consuming equipment. This data is then combined with load change data and flow parameter change data to form energy consumption parameter change data.

[0092] Based on load change data, energy medium parameters, and specific heat capacity, the total energy consumption and unit product consumption are calculated to obtain the consumption quality, and the temperature quality for stable energy supply is determined.

[0093] By analyzing the changing trends of energy consumption parameters with production time, the stability of energy supply and demand is assessed, and energy quality data is formed together with consumption quality and temperature quality.

[0094] The energy consumption prediction and control module is used to formulate basic energy consumption control strategies based on changes in energy consumption parameters. It obtains energy consumption prediction data by predicting energy consumption changes based on changes in energy consumption parameters using the energy consumption time series prediction method.

[0095] The steps for the energy consumption prediction and control module to formulate basic energy consumption control strategies include:

[0096] Based on the factory's energy conservation and emission reduction targets and dual carbon control indicators, the control scale of energy recovery and intensive utilization is determined to obtain the energy control target. The production energy demand is obtained by analyzing the factory's production scheduling plan and the energy demand of each process for water, electricity, gas, steam, and compressed air.

[0097] By identifying the type and magnitude of energy consumption parameter fluctuations through load change data and flow change data, and extracting the time-series characteristics of energy consumption parameter change data.

[0098] Energy supply and demand matching strategies are formulated based on energy management objectives and production energy demand. Energy consumption parameter fluctuation scenarios are generated according to the type, magnitude and time series characteristics of parameter fluctuations, and key management factors for different scenarios are identified.

[0099] Based on the fluctuation scenarios of energy consumption parameters, select the corresponding energy consumption control and energy-saving adjustment methods, and adjust them according to the degree of influence of key control factors on the energy supply and demand matching strategy to obtain the basic energy consumption control strategy.

[0100] The steps for the energy consumption prediction and control module to generate energy consumption prediction data include:

[0101] A time-series energy consumption prediction model based on random forest is established. The changing patterns and characteristics of energy consumption parameters in each production cycle are analyzed to obtain predicted load, predicted consumption, and predicted pipeline pressure. Production condition characteristics, environmental climate characteristics, and time-series cycle characteristics are extracted from the energy consumption parameter change data.

[0102] Using production operating conditions, environmental climate characteristics, and time-series cycle characteristics as inputs, and predicted load, predicted consumption, and predicted pipeline pressure as outputs, the energy consumption time-series prediction model is trained, and model parameters that meet the test accuracy are retained.

[0103] Based on the current production conditions, current environmental climate, and current time series, extract the corresponding real-time load, real-time consumption, and real-time pipeline pressure, and input them into the trained energy consumption time series prediction model to obtain predicted values ​​of energy consumption parameters.

[0104] The energy transmission loss rate is calculated based on pipeline fluid dynamics, and the natural loss rate of the medium is obtained by combining the loss rate of compressed air and steam with the ambient temperature and humidity. The formula is expressed as:

[0105]

[0106]

[0107] In the formula, It is the energy transmission loss rate. It is the thermal conductivity of the medium in the pipeline network. It is the conveying cross-sectional area. It is the pipeline operating pressure. It is the environmental benchmark pressure. It refers to the pipeline transmission length. It is the total energy supply. It is the production time. This is the initial media reserve. It is the loss coefficient. It is the temperature of the steam medium. It is the ambient air temperature. It is the natural loss rate of the medium.

[0108] Energy consumption prediction data is calculated by combining predicted energy consumption parameters, energy transmission loss rate, and natural loss rate of the medium.

[0109] The energy consumption strategy optimization module is used to formulate dynamic energy consumption adjustment strategies based on production load energy consumption data and energy consumption forecast data, and to adjust the basic energy consumption control strategy in conjunction with factory production scheduling requirements and energy quality data to obtain an optimized energy consumption control scheme.

[0110] The steps for the energy consumption strategy optimization module to formulate a dynamic energy consumption adjustment strategy include:

[0111] Energy consumption prediction data and production load energy consumption data are converted into a unified time series granularity through interpolation algorithms, and abnormal energy consumption values ​​are removed and smoothed.

[0112] The differences in the original time-series granularity between the two types of data were identified. Energy consumption prediction data is typically predicted at the hourly level, while production load energy consumption data is collected in real-time at the minute level. A linear interpolation method combined with a cubic spline interpolation algorithm was used to convert both types of data into a standard time-series granularity of 15 minutes / 30 minutes. Simultaneously, pre-processed real-time energy consumption data and energy medium fluctuation data were retrieved as references. A two-layer anomaly detection mechanism combining the Laida 3σ criterion and box plot analysis was employed to accurately remove abnormal energy consumption values ​​caused by prediction deviations, equipment failures, and sudden changes in operating conditions. The time-series data after outlier removal was smoothed using a moving average method to eliminate data fluctuation noise and fill in short-term data gaps.

[0113] A two-dimensional supply-demand matching matrix is ​​established by describing the matching relationship between different production periods and energy supply grades.

[0114] Production periods are categorized into peak, off-peak, low-peak, and transitional periods based on factory production schedules and process progress. The production load demand, energy priority, and energy medium demand type for each period are clearly defined. Simultaneously, based on energy quality data, energy supply grade levels are defined, such as: High grade: stable energy parameters, low losses, suitable for high-load production; Medium grade: parameter fluctuations within acceptable range, moderate losses, suitable for routine production; General grade: larger parameter fluctuations, higher losses, suitable for low-load production or auxiliary energy use. A two-dimensional supply-demand matching matrix is ​​established, with production periods as rows and energy supply grades as columns. Matrix elements quantitatively represent the matching degree between different production periods and different energy supply grades. The matching degree calculation considers factors such as energy demand, energy quality requirements, and energy efficiency for each period, and the weights are determined using the analytic hierarchy process (AHP) to ultimately generate a standardized two-dimensional supply-demand matching matrix.

[0115] Calculate the energy supply and demand gap for each production period, classify the gaps according to their severity, and establish a multi-objective energy efficiency optimization model with energy utilization efficiency, production and operation economic costs, and plant carbon emissions as optimization objectives.

[0116] Based on energy consumption datasets, production load energy consumption data, and energy consumption forecast data, and combined with a two-dimensional supply-demand matching matrix, the total supply and demand of various energy media are calculated for each production period. Using the supply-demand difference formula (supply-demand gap = total demand - total supply), the supply-demand gap for each production period and for each type of energy media is accurately calculated. A positive gap indicates insufficient supply, while a negative gap indicates oversupply. Considering factory production priorities and energy cost affordability, the degree of supply-demand gap is divided into three levels: mild gap (gap of 5%-15% of total demand), moderate gap (gap of 15%-30% of total demand), and severe gap (gap of ≥30% of total demand), with the impact range corresponding to each level marked. Subsequently, a multi-objective energy efficiency optimization model is built, clarifying three major optimization objectives: first, to improve energy utilization efficiency; second, to reduce production and operating economic costs; and third, to reduce carbon emissions from the plant area. The model constraints include total energy supply constraints, production load constraints, equipment operating parameter constraints, energy quality constraints, and carbon emission limit constraints. At the same time, weighting coefficients are introduced and determined in conjunction with the factory's energy-saving goals, dual-carbon requirements, and economic benefit priorities. This transforms the multi-objective optimization into a solvable single-objective optimization problem, ensuring that the model fits the actual production needs of the factory.

[0117] The optimal energy consumption value is obtained by iteratively transitioning the state of the multi-objective energy efficiency optimization model using dynamic programming algorithm, and a graded load response mechanism is formulated in combination with different levels of supply and demand gap to obtain a dynamic energy consumption adjustment strategy.

[0118] The core parameters of the dynamic programming algorithm are determined, and the production period is divided into several discrete states. Energy supply and demand gap, energy efficiency, economic cost, and carbon emissions for each period are used as state variables, and energy consumption adjustment is used as the decision variable. A state transition equation is constructed, clarifying the decision rules for different states. Based on historical energy consumption data and production condition data, the model is iteratively trained. Through gradual optimization, the optimal energy consumption value for each production period (i.e., the optimal energy consumption allocation scheme that satisfies the three optimization objectives and meets the constraints) is obtained. Subsequently, a differentiated graded load response mechanism is formulated based on three levels of supply and demand gaps: mild gaps adopt conventional adjustment strategies, such as optimizing auxiliary equipment energy use, fine-tuning energy allocation ratios, and prioritizing energy supply for core processes. Moderate gaps adopt key adjustment strategies, such as adjusting production schedules, optimizing equipment operating parameters, and activating backup energy to balance supply and demand. Severe gaps adopt emergency adjustment strategies, such as suspending non-core processes, activating emergency energy supply, and linking energy storage systems to replenish energy, to avoid affecting core production. At the same time, the activation conditions, execution steps, responsible departments and adjustment time limits of each level of response mechanism are clearly defined, the optimal energy consumption utilization value is deeply integrated with the hierarchical response mechanism, and the effect prediction and risk prevention and control measures after energy consumption adjustment are supplemented, so as to form a complete dynamic energy consumption adjustment strategy.

[0119] The steps by which the energy consumption strategy optimization module obtains an optimized energy consumption management scheme include:

[0120] The energy needs of each production process and energy-consuming department in the factory are classified according to their urgency and importance. The energy needs are then prioritized by combining the annual and monthly production schedules to estimate the scale and duration of various energy needs at different times.

[0121] The entire production process and energy-consuming departments of the factory were analyzed. By connecting with the MES production management system and ERP management system, production process documents, capacity planning, order delivery cycles, and energy consumption ledgers and historical energy consumption data of each department were retrieved to clarify the types of energy needs for each process and department. Using the Analytic Hierarchy Process (AHP), a classification and evaluation system was constructed from three dimensions: production importance, energy urgency, and order delivery priority, to classify each process and department into levels. Subsequently, the annual production plan and monthly production schedule were retrieved to clarify the production tasks and process progress for each time period. Combined with the unit-time energy consumption standards of each process and department, the specific scale and duration of various energy needs in different time periods were estimated. Simultaneously, the energy demand estimates were dynamically revised based on order urgency and production progress. Finally, a priority ranking algorithm was used to rank various energy needs based on the classification results, energy urgency, and production task priority, determining that the energy priority of core processes and key departments is the highest, and auxiliary energy needs are the lowest, forming a standardized energy demand priority ranking table.

[0122] By combining energy quality data and energy consumption forecast data, the total amount, spatiotemporal distribution and supply and demand fluctuation characteristics of various energy resources in the whole plant are assessed, and the supply and demand matching assessment results are obtained by combining the degree of matching with the energy demand of each department.

[0123] Energy consumption forecast data and energy quality data were retrieved to build an energy resource assessment model, defining the assessment dimensions as follows: First, a total energy resource assessment was conducted, summarizing the predicted total supply of five energy media—electricity, water, gas, steam, and compressed air—and comparing it with historical supply data and total production energy demand to assess the sufficiency of energy resources. Second, a spatiotemporal distribution assessment was performed, analyzing the supply distribution of various energy media according to production periods and energy consumption areas, identifying peak and off-peak periods and key supply areas, and recognizing spatiotemporal imbalances. Third, a supply-demand fluctuation assessment was conducted, combining energy consumption parameter change data to analyze the synchronicity between energy supply fluctuations and energy demand fluctuations, predicting the risk of supply-demand imbalances at different times. Subsequently, a supply-demand matching assessment system was built, based on an energy demand priority ranking table, and considering the energy demand type, scale, and quality requirements of each process and department, to quantitatively calculate the matching degree between various energy resources and different energy demands. The matching degree calculation covered three core indicators: energy quality suitability, supply timing suitability, and usage scale suitability. Finally, a supply-demand matching assessment report was generated.

[0124] Based on the priority ranking of energy demand and the assessment results of supply and demand matching, the basic energy consumption management strategy is adjusted, energy resources are dynamically reallocated, and corresponding energy-saving control and scheduling methods are determined for different energy quality levels and different energy demand to obtain an optimized energy consumption management plan.

[0125] Based on the retrieved basic energy consumption control strategy and combined with the priority ranking of energy demand and the assessment results of supply and demand matching, a strategy adjustment plan was formulated: For high-matching links, the original control model was maintained, and detailed parameters were optimized, such as fine-tuning the energy supply timing to improve energy efficiency. For medium-matching and low-matching links, the focus was on adjusting the energy allocation ratio, prioritizing high-priority energy demand, and reasonably compressing low-priority energy demand to achieve dynamic redistribution of energy resources. During the energy allocation process, the allocation plan was adjusted in real time based on energy consumption forecast data and energy medium fluctuation data to avoid supply and demand imbalance. At the same time, key parameters in the allocation process, such as the energy allocation amount and allocation time period for each department, were recorded to ensure traceability of the allocation process. Subsequently, personalized energy-saving control and scheduling methods were formulated for different energy quality levels and differentiated energy demand: For high-quality energy, priority was given to core processes and key equipment, and a precise control mode was adopted, such as real-time monitoring of energy consumption status and dynamic adjustment of supply parameters to maximize the utilization efficiency of high-quality energy. For medium-quality energy, it was allocated to regular production processes and conventional energy-saving control was adopted. For general energy, it was allocated to auxiliary departments and low-load processes and energy-saving constraint control was adopted. Simultaneously, supplementary energy-saving measures are added, such as waste heat recovery and utilization, equipment frequency conversion control, and pipeline leakage control. The details of these measures are optimized for different energy consumption scenarios; for example, core processes utilize high-efficiency waste heat recovery, while auxiliary processes employ simple energy-saving management. Finally, the adjusted energy allocation scheme, personalized energy-saving control methods, and supporting energy-saving measures are integrated to form a complete optimized energy consumption management scheme, clearly defining the implementation steps, responsible departments, and control indicators.

[0126] In summary, the smart factory energy consumption monitoring and analysis system provided in this embodiment breaks down data barriers for water, electricity, gas, and steam by normalizing and screening multi-source data for initial anomalies. It can clearly show the flow of every drop of water and the consumption point of every kilowatt-hour of electricity in real time, achieving data transparency. By leveraging a random forest time-series prediction model, it anticipates production peaks and troughs, intervening before energy consumption anomalies occur and nipping energy waste in the bud, significantly reducing the risk of sudden energy losses. Combining production scheduling and energy quality, it dynamically allocates resources using a multi-objective optimization model, ensuring energy supply for high-priority production processes and cleverly utilizing a tiered response mechanism for energy supply and demand gaps, minimizing energy costs and related carbon emissions while ensuring production.

[0127] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0128] 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 the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An energy consumption monitoring and analysis system for a smart factory, characterized in that, include: The energy data acquisition module is used to collect data on energy media within the factory and monitor the energy consumption of each energy-consuming department in real time to obtain real-time energy consumption data and energy media fluctuation data. The energy consumption impact analysis module is used to analyze the impact of the energy medium fluctuation data on the overall production equipment of the factory according to the multi-energy collaborative calculation method to obtain production load energy consumption data, and to extract energy consumption parameter change data and energy quality data from the real-time energy consumption data. The energy consumption prediction and control module is used to formulate basic energy consumption control strategies based on the energy consumption parameter change data, and to obtain energy consumption prediction data by predicting energy consumption changes based on the energy consumption parameter change data through the energy consumption time series prediction method. The energy consumption strategy optimization module is used to formulate a dynamic energy consumption adjustment strategy based on the production load energy consumption data and the energy consumption prediction data, and to adjust the basic energy consumption control strategy in combination with the factory production scheduling requirements and the energy quality data to obtain an optimized energy consumption control scheme.

2. The energy consumption monitoring and analysis system for a smart factory according to claim 1, characterized in that: The steps by which the energy data acquisition module obtains the energy medium fluctuation data include: Based on their energy properties, factory energy media are classified into five categories: electrical energy, water energy, coal gas energy, steam energy, and compressed air energy. Data on power distribution circuit load, voltage, current, harmonics, and time-of-use electricity consumption are collected for the electrical energy source; data on water inflow and outflow rates and water pressure loss are collected for the water energy source in each workshop. Data on the gas supply flow rate, pressure, concentration, and consumption rate of the gas energy pipeline were collected; data on the steam energy transmission temperature, pressure, flow rate, and heat loss were collected. The compressed air energy collection network includes data on air pressure, flow rate, leakage, and equipment air load. Multi-source energy consumption data is then normalized and initial anomaly screening is performed to obtain the energy medium fluctuation data.

3. The energy consumption monitoring and analysis system for a smart factory according to claim 2, characterized in that: The steps by which the energy data acquisition module obtains the real-time energy consumption data include: By using an edge computing gateway compatible with multiple industrial communication protocols, the energy medium fluctuation data is preprocessed locally, converted according to protocols, and resumed from breakpoints to obtain preliminary real-time data. The preliminary real-time data is subjected to outlier detection and cleaning to remove noise data caused by equipment failure and network fluctuations, thus obtaining the real-time energy consumption data.

4. The energy consumption monitoring and analysis system for a smart factory according to claim 1, characterized in that: The energy consumption impact analysis module includes: The energy consumption characteristic analysis unit is used to determine the energy consumption characteristics of the corresponding energy type based on the physical transmission and consumption characteristics of various energy media, collect the operating parameters of factory production equipment, and analyze the energy consumption and energy consumption fluctuation of different production processes to obtain the energy consumption characteristics of the equipment. The energy consumption impact assessment unit is used to establish an energy consumption factor analysis model based on the energy consumption characteristics and the equipment energy consumption characteristics, analyze the quantitative relationship between energy parameters and equipment operating performance, determine the impact weight of each parameter in the energy parameters on equipment energy consumption, and calculate the degree of impact of different types of energy media on the actual energy consumption of factory equipment. The load coordination calculation unit is used to establish a plant-wide energy balance model to calculate the total comprehensive energy flow value, and to calculate the production load energy consumption data under the coordinated management of multiple energy media by combining the production line design load and the actual energy consumption impact.

5. The energy consumption monitoring and analysis system for a smart factory according to claim 4, characterized in that: The steps for the load coordination calculation unit to calculate the total integrated energy flow include: The energy input items are determined based on the electricity input required for factory production, fuel consumption, purchased steam, and the energy carried by the medium itself. Energy output items are determined based on energy consumption in the production of finished products, energy consumption in process by-products, and energy loss from end-of-pipe emissions. The total comprehensive energy flow value is calculated based on the energy input and energy output items, combined with pipeline transmission losses, equipment no-load losses, and production condition fluctuation losses.

6. The energy consumption monitoring and analysis system for a smart factory according to claim 1, characterized in that: The steps by which the energy consumption impact analysis module obtains the energy consumption parameter change data and the energy quality data include: By analyzing the production load energy consumption data of each energy-consuming node, the difference in electricity load between peak and off-peak periods is calculated to obtain load change data. In addition, the inlet and outlet flow data of the energy pipeline network are statistically analyzed, and the flow fluctuation of various media is calculated to obtain flow parameter change data. Pressure parameter change data is obtained by calculating the pressure loss and dynamic change range of the energy transmission pipeline network and energy-consuming equipment through pressure monitoring data of energy transmission pipeline network and energy-consuming equipment. This data, together with the load change data and the flow parameter change data, constitutes the energy consumption parameter change data. Based on the load change data, energy medium parameters and specific heat capacity, the total energy consumption and unit product consumption are calculated to obtain the consumption quality, and the temperature quality for stable energy supply is determined. By analyzing the changing trends of the energy consumption parameters with the production timeline, the stability of energy supply and demand is assessed, and the energy quality data is formed together with the consumption quality and the temperature quality.

7. The energy consumption monitoring and analysis system for a smart factory according to claim 6, characterized in that: The steps for the energy consumption prediction and control module to formulate the basic energy consumption control strategy include: Based on the factory's energy conservation and consumption reduction targets and dual carbon control indicators, the control scale of energy recovery and intensive utilization is determined to obtain the energy control targets. The production energy demand is obtained by analyzing the factory's production scheduling plan and the energy demand of each process for water, electricity, gas, steam, and compressed air. The type and magnitude of energy consumption parameter fluctuations are identified by the load change data and the flow change data, and the time-series characteristics of the energy consumption parameter change data are extracted. Based on the energy management objectives and the energy demand for production, an energy supply and demand matching strategy is formulated. Based on the type and magnitude of parameter fluctuations and the time series characteristics, energy consumption parameter fluctuation scenarios are generated, and key management factors for different scenarios are determined. Based on the energy consumption parameter fluctuation scenario, select the corresponding energy consumption control and energy-saving adjustment method, and adjust it according to the degree of influence of the key control factors on the energy supply and demand matching strategy to obtain the basic energy consumption control strategy.

8. The energy consumption monitoring and analysis system for a smart factory according to claim 1, characterized in that: The steps for the energy consumption prediction and control module to generate the energy consumption prediction data include: A time-series energy consumption prediction model based on random forest is established. The change patterns and characteristics of the energy consumption parameters in each production cycle are analyzed to obtain the predicted load, predicted consumption, and predicted pipeline pressure. Production condition characteristics, environmental climate characteristics, and time-series cycle characteristics are extracted from the energy consumption parameter change data. The production condition characteristics, environmental climate characteristics, and time-series periodic characteristics are used as inputs, and the predicted load, predicted consumption, and predicted pipeline pressure are used as outputs to train the energy consumption time-series prediction model, retaining model parameters that meet the test accuracy. Based on the current production conditions, current environmental climate, and current time series, extract the corresponding real-time load, real-time consumption, and real-time pipeline pressure, and input them into the trained energy consumption time series prediction model to obtain predicted values ​​of energy consumption parameters. The energy transmission loss rate is calculated based on pipeline fluid dynamics, and the natural loss rate of the medium is obtained by combining the loss rate of compressed air and steam with the ambient temperature and humidity. The predicted energy consumption data is calculated by combining the predicted energy consumption parameters, the energy transmission loss rate, and the natural loss rate of the medium.

9. The energy consumption monitoring and analysis system for a smart factory according to claim 8, characterized in that: The steps for the energy consumption strategy optimization module to formulate the dynamic energy consumption adjustment strategy include: The energy consumption prediction data and the production load energy consumption data are converted into a unified time series granularity through an interpolation algorithm, and abnormal energy consumption values ​​are removed and smoothed. A two-dimensional supply-demand matching matrix is ​​established by describing the matching relationship between different production periods and energy supply grades; Calculate the energy supply and demand gap for each production period, classify the gaps according to their severity, and establish a multi-objective energy efficiency optimization model with energy utilization efficiency, production and operation economic costs, and plant carbon emissions as optimization objectives. The optimal energy consumption utilization value is obtained by iteratively transitioning the state of the multi-objective energy efficiency optimization model using a dynamic programming algorithm, and the energy consumption dynamic adjustment strategy is obtained by formulating a graded load response mechanism in combination with different levels of supply and demand gaps.

10. The energy consumption monitoring and analysis system for a smart factory according to claim 8, characterized in that: The steps by which the energy consumption strategy optimization module obtains the optimized energy consumption management scheme include: The energy needs of each production process and energy-consuming department in the factory are classified according to their urgency and importance. The scale and duration of various energy needs at different times are estimated in conjunction with the annual and monthly production schedules to obtain the priority ranking of energy needs. By combining the energy quality data and the energy consumption forecast data, the total amount, spatiotemporal distribution and supply and demand fluctuation characteristics of all types of energy resources in the plant are assessed, and the supply and demand matching assessment results are obtained by combining the degree of matching with the energy demand of each department. The basic energy consumption control strategy is adjusted based on the energy demand priority ranking and the supply and demand matching assessment results. Energy resources are dynamically reallocated, and corresponding energy-saving control and scheduling methods are determined for different energy quality levels and different energy demand to obtain the optimized energy consumption control scheme.

Images