An intelligent energy-saving system and energy-saving control method for air compressor stations

Through multi-source sensing and edge intelligent decision-making of the air compressor station intelligent energy-saving system, the coordinated scheduling of air compressor clusters, air storage, purification and waste heat recovery is realized, which solves the problems of energy waste and production instability of air compressor stations, improves energy efficiency and operational stability, and reduces maintenance costs.

CN123067378APending Publication Date: 2026-11-10ZHONGTIAN DIGITAL ENERGY MANAGEMENT (HANGZHOU) CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611219902.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-08-12
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing air compressor station control technology cannot achieve optimal overall energy efficiency, resulting in energy waste and production instability. Furthermore, the lack of predictive maintenance leads to frequent unplanned equipment shutdowns and high maintenance costs.

Method used

An intelligent energy-saving system for air compressor stations is adopted. Data is collected through multi-source sensing units, and multi-modal load prediction and global multi-objective optimization are performed using edge intelligent decision-making units. This enables coordinated scheduling of air compressor clusters, graded air storage, drying and purification, and waste heat recovery. Predictive maintenance is also performed in conjunction with closed-loop self-learning.

Benefits of technology

It has achieved a significant improvement in overall station energy efficiency, reduced pipeline pressure fluctuations, increased energy saving rate, enhanced equipment operation stability, and reduced maintenance costs.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123067378A_ABST
    Figure CN123067378A_ABST
Patent Text Reader

Abstract

This invention discloses an intelligent energy-saving system and control method for air compressor stations, belonging to the field of industrial energy conservation and intelligent air compressor control technology. The system includes a multi-source sensing unit, an edge intelligent decision-making unit, and a full-link execution unit encompassing air compressor clusters, graded gas storage, drying and purification, and waste heat recovery. It uses a multi-modal fusion model to predict gas load in advance, and then generates a full-link collaborative scheduling scheme through a global multi-objective optimization algorithm, achieving global optimized control of gas production, storage, purification, and waste heat recovery. It also possesses closed-loop self-learning and predictive maintenance capabilities. This invention breaks through the limitations of traditional fragmented control of air compressor stations, effectively improving the overall energy saving rate and simultaneously controlling pipeline pressure fluctuations. It combines multiple advantages such as improved energy efficiency, stable gas supply, waste heat efficiency enhancement, and intelligent operation and maintenance, and can be widely applied to energy-saving renovation and intelligent upgrading of air compressor stations in various industrial scenarios.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application belongs to the field of industrial energy conservation and intelligent control technology of air compressors. Specifically, it relates to an intelligent energy-saving system and energy-saving control method for an air compressor station. Background Technology

[0002] Compressed air is the second most widely used power source in the industrial sector, supporting production operations in industries such as machinery manufacturing, chemicals, metallurgy, and electronics. As the core supply unit for compressed air, air compressor stations typically account for 10% to 35% of the total electricity consumption of industrial enterprises, making them a key target for industrial energy-saving renovations. Currently, air compressor station control technology has gradually evolved from single-unit power frequency control to variable frequency unit group control, with some solutions incorporating simple load forecasting to assist unit scheduling. However, certain shortcomings still exist overall.

[0003] In existing technologies, the four main units—air compressor cluster, air storage system, drying and purification, and waste heat recovery—operate and are individually controlled, which can only achieve local optimization for a single device and cannot achieve optimal energy efficiency for the entire station. The overall energy saving rate of conventional retrofits may not be high, and the energy saving potential is not fully released. Moreover, traditional control relies on PID feedback regulation based on real-time pipeline pressure, which has a lag response when the gas load fluctuates. The pipeline pressure fluctuation is usually high, which not only affects the stability of production gas supply but may also lead to frequent loading and unloading of air compressors, resulting in a large amount of additional no-load energy consumption.

[0004] The energy consumption of air compressors during unloading and idling accounts for 10% to 30% of the total energy consumption; the dryer operates at constant power, and the drying depth does not match the actual air load and ambient humidity, resulting in ineffective energy consumption; the waste heat recovery system is disconnected from the air production demand, and waste heat recovery squeezes out air production capacity during peak air consumption, while waste heat is not fully utilized during off-peak air consumption, resulting in low energy cascade utilization rate; existing technologies generally rely on manual periodic inspections and post-fault repairs, which may not be able to predict equipment deterioration trends in advance, and in actual operation, unplanned downtime is likely to occur, resulting in high maintenance costs and production losses. Summary of the Invention

[0005] To address the aforementioned problems and technical deficiencies, this application adopts the following technical solution: an intelligent energy-saving system for air compressor stations, wherein the full-link execution unit includes an air compressor cluster unit, a graded air storage unit, a drying and purification unit, and a waste heat recovery unit, covering the entire energy link of air compressor station gas production, storage, purification, and waste heat recovery.

[0006] The multi-source sensing unit is used to collect operating data, pipeline operating data, environmental data and production plan data of each subsystem of the full-link execution unit, and transmit them to the edge intelligent decision-making unit.

[0007] Furthermore, the multi-source sensing unit collects data including: air compressor outlet pressure, temperature, current, and vibration data; air storage tank graded pressure data; dryer dew point and energy consumption data; waste heat recovery system temperature and flow data; pipeline end pressure data; ambient temperature and humidity data; and production shift gas usage plan data, thereby achieving comprehensive and multi-dimensional perception of the entire station's operating status.

[0008] The edge intelligent decision-making unit has a built-in multimodal load prediction module and a global multi-objective optimization module, which are used to generate gas load prediction results based on the collected data, solve for the whole-link collaborative scheduling scheme, and send the scheduling instructions to each subsystem of the whole-link execution unit.

[0009] Furthermore, the multimodal load prediction module adopts a prediction model that integrates LSTM and Transformer. The input features include historical gas load, pipeline pressure, equipment status, environmental weather, and production plan. The output is a gas load prediction curve for the next 1 to 24 hours, which enables advanced prediction of gas demand and solves the lag problem of traditional feedback control from the root.

[0010] Furthermore, the global multi-objective optimization module adopts an improved NSGA-II algorithm. The optimization objectives include minimizing the total energy consumption of the entire station, minimizing pipeline pressure fluctuations, optimizing the average load rate of equipment, and maximizing the waste heat recovery rate. The constraints include the allowable range of gas supply pressure, the rated operating range of equipment, and the threshold for achieving the drying dew point standard. The optimal scheduling scheme for the coordinated operation of each subsystem is obtained through parallel solution of multiple objectives, thereby achieving the optimal global energy efficiency of the entire station.

[0011] The air compressor cluster unit, the staged air storage unit, the drying and purification unit, and the waste heat recovery unit are used to receive scheduling instructions and operate in coordination to achieve full-chain control of air production, air storage, purification, and waste heat recovery.

[0012] Furthermore, the air compressor cluster unit includes variable frequency and fixed frequency air compressors, which, together with the staged air storage unit, achieve zero unloading operation, maintain the equipment load rate in the high-efficiency range of 70% to 85%, and completely eliminate the energy consumption of air compressor unloading and idling.

[0013] Furthermore, the drying and purification unit performs dew point adaptive adjustment, dynamically matching the drying depth according to the gas load and ambient humidity, thereby reducing drying and purification energy consumption while ensuring gas supply quality.

[0014] Furthermore, the waste heat recovery unit implements load linkage control, prioritizing gas supply during peak gas consumption periods and maximizing waste heat recovery during off-peak gas consumption periods, thereby achieving efficient tiered utilization of energy.

[0015] Furthermore, the edge intelligent decision-making unit also has a built-in closed-loop self-learning and predictive maintenance module, which is used to collect the operation effect data of the execution unit in real time, correct the parameters of the load prediction and optimization model, and continuously improve the control accuracy; at the same time, it performs health status assessment and fault warning based on equipment vibration and temperature data, realizing the upgrade from passive operation and maintenance to predictive operation and maintenance.

[0016] A method for intelligent energy-saving control of an air compressor station, implemented using the aforementioned intelligent energy-saving system for the air compressor station, includes the following steps: S1. Multi-source data acquisition: The multi-source sensing unit collects operational data, pipeline condition data, environmental data, and production plan data from the entire air compressor station chain, providing comprehensive data support for subsequent intelligent decision-making.

[0017] S2. Multimodal advance load forecasting: Based on the collected multi-dimensional data, a gas load forecast curve for the next 1 to 24 hours is generated.

[0018] Furthermore, in step S2, a prediction model combining LSTM and Transformer is used for load forecasting. The input features cover historical gas load, pipeline pressure, equipment status, environmental weather, and production plan, taking into account the time-series change patterns and multi-dimensional influencing factors, thus significantly improving the prediction accuracy.

[0019] S3. Global multi-objective optimization solution: With the optimization objectives of minimizing total energy consumption, minimizing pipeline pressure fluctuation, optimizing equipment load rate, and maximizing waste heat recovery rate, the scheduling schemes for air compressor clusters, staged gas storage, drying and purification, and waste heat recovery are obtained.

[0020] Furthermore, in step S3, the improved NSGA-II algorithm is used to solve the multi-objective optimization problem. The allowable range of gas supply pressure, the rated operating range of equipment, and the dry dew point threshold are used as constraints to output the globally optimal collaborative scheduling strategy.

[0021] S4. Full-link coordinated execution: The scheduling scheme is broken down into control commands for each subsystem, and coordinated control of gas production scheduling, gas storage charging and discharging, dew point adjustment, and waste heat linkage is executed synchronously.

[0022] Furthermore, step S4 specifically includes: coordinating the charging and discharging of the graded gas storage with the air compressor scheduling to ensure that the air compressor operates in the high-efficiency load range of 70% to 85% and avoids unloading and idling; dynamically adjusting the drying depth according to the real-time gas load and ambient humidity to reduce drying energy consumption; and establishing a linkage mechanism between the gas load and waste heat recovery power to achieve energy cascade utilization.

[0023] S5. Closed-loop feedback self-learning: Collects actual system operation data, reversely corrects the prediction model and optimization algorithm parameters, and continuously iterates to optimize control accuracy.

[0024] Furthermore, step S5 also includes: assessing the health status of the equipment based on the collected equipment vibration, temperature, and current data, providing early warnings of potential faults, and realizing predictive maintenance.

[0025] Compared to existing technologies, the beneficial effects of this application are as follows: (1) This application incorporates air compressor clusters, staged air storage, drying and purification, and waste heat recovery into a unified multi-objective optimization framework to achieve full-link coordinated scheduling. Compared with the energy saving rate of traditional single-point optimization schemes, the comprehensive energy saving rate of this scheme is effectively improved, and the energy efficiency improvement is far greater than the expected level of conventional technical means in the prior art. (2) This application adopts a multimodal fusion prediction model to achieve advance prediction of gas load from 1 to 24 hours. Combined with the peak shaving and valley filling effect of graded gas storage, it effectively controls pipeline pressure fluctuations, which not only ensures the stability of gas consumption for production, but also avoids the hidden energy consumption waste caused by excessive pressure.

[0026] (3) This application eliminates the no-load energy consumption of the air compressor through zero unloading control, reduces the drying and purification energy consumption through dew point adaptive adjustment, and realizes the resource utilization of waste heat through load linkage waste heat recovery. It simultaneously explores the energy-saving potential from multiple dimensions such as gas production, gas storage, purification and waste heat, thereby realizing the cascade efficient utilization of energy. Attached Figure Description

[0027] In the attached diagram: Figure 1 This is a system diagram from an embodiment of this application; Figure 2 This is a flowchart of an embodiment of this application. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of this application, but not all embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can be arranged and designed in various different configurations.

[0029] Example 1: This embodiment is applied to an air compressor station in a machinery manufacturing plant. The original air compressor station was equipped with 5 air compressors (3 industrial frequency and 2 variable frequency), and was equipped with conventional air storage tanks and refrigerated dryers. Waste heat was not effectively recovered. The traditional PLC group control method was used, resulting in low overall energy efficiency and large pressure fluctuations.

[0030] The architecture of the intelligent energy-saving air compressor station system built in this embodiment is as follows: Figure 1As shown, it includes a multi-source sensing unit, an edge intelligent decision-making unit, and a full-link execution unit. The full-link execution unit covers an air compressor cluster unit, a graded air storage unit, a drying and purification unit, and a waste heat recovery unit.

[0031] Multi-source sensing unit: Pressure, temperature, current, and vibration sensors are installed at the outlets of each air compressor; pressure sensors are installed at the staged air storage tanks; dew point and energy consumption sensors are installed at the dryers; temperature and flow sensors are installed at the waste heat recovery system; pressure sensors are installed at the end of the pipeline network; and ambient temperature and humidity data and factory production scheduling data are integrated to achieve comprehensive collection of all station operation data, with a data sampling period of 1 second.

[0032] Edge Intelligent Decision Unit: Deployed at the edge of the factory's industrial control computer, it has built-in multimodal load prediction module, global multi-objective optimization module, and closed-loop self-learning and predictive maintenance module, and is responsible for data processing, prediction and solution and command issuance.

[0033] Air compressor cluster unit: The original 5 variable frequency and fixed frequency air compressors are connected to the decision unit for control, realizing the coordinated scheduling of unit start-up and shutdown and variable frequency speed regulation.

[0034] Tiered gas storage unit: A high-pressure gas storage tank is added to form a two-stage gas storage system. It receives dispatch instructions to execute gas filling and discharging control, and works with the air compressor to achieve peak shaving and valley filling and zero unloading operation.

[0035] Drying and purification unit: The original dryer is equipped with an intelligent adjustment module that receives scheduling instructions and performs dew point adaptive adjustment.

[0036] Waste heat recovery unit: A new air compressor waste heat recovery system is added and connected to the decision unit to achieve linkage control with the gas load, and the recovered waste heat is used for the supply of domestic hot water in the plant area.

[0037] Reference Figure 2 In actual operation, the steps are as follows: Multi-source data acquisition: The multi-source sensing unit collects operating parameters of various equipment, pipeline conditions, environmental data and production plan data in real time, and uploads them to the edge intelligent decision-making unit to build a data foundation for the entire station's operation.

[0038] Multimodal advance load forecasting: Based on historical gas consumption data, pipeline pressure, equipment status, ambient temperature and humidity, production schedule and other multi-dimensional features, the decision unit generates hourly gas load forecast curves for the next 24 hours through the fusion model of LSTM and Transformer. The forecast accuracy can reach more than 95%, providing a basis for advance scheduling.

[0039] Global multi-objective optimization solution: With the optimization objectives of minimizing total energy consumption, minimizing pipeline pressure fluctuation, optimizing equipment load rate, and maximizing waste heat recovery rate, and with constraints of air supply pressure of 0.6~0.7MPa, equipment rated load range, and dew point ≤-20℃, the improved NSGA-II algorithm is used to generate a coordinated scheduling strategy for future periods, including air compressor start-up and shutdown speed regulation schemes, air tank charging and discharging plans, dryer dew point adjustment schemes, and waste heat recovery power schemes.

[0040] Air compressor cluster: Adjust the number of operating units and frequency converter output in advance according to the scheduling plan, and coordinate with the charging and discharging of the staged air storage tank to ensure that the air compressor always operates in the high-efficiency load range of 70% to 85%, avoiding unloading and idling conditions throughout the process, and reducing unloading energy consumption by more than 90%.

[0041] Tiered gas storage: During periods of low gas demand, the excess gas production of the air compressor is used to fill the high-pressure gas storage tank for energy storage; during periods of high gas demand, the high-pressure stored gas is released to supplement the gas supply, smooth out gas demand fluctuations, and maintain stable pipeline pressure.

[0042] Drying and purification: The dryer's operating power is dynamically adjusted according to the real-time gas load and ambient humidity, reducing drying energy consumption by about 20% while meeting dew point requirements.

[0043] Waste heat recovery: During peak gas consumption periods, the waste heat recovery power is reduced to prioritize the supply of compressed air; during off-peak gas consumption periods, the waste heat recovery power is maximized, and the recovered waste heat is used to supply hot water for production and domestic use in the plant area, realizing the cascade utilization of energy.

[0044] Closed-loop feedback and predictive maintenance: The system collects real-time data on actual operating pressure, energy consumption, and equipment status, and reverse-corrects the load prediction model and optimizes the algorithm parameters to continuously improve control accuracy; at the same time, it monitors the health status of equipment by monitoring vibration and temperature data, and provides early warnings of potential faults such as bearing wear and motor overheating 7 to 15 days in advance, thus achieving predictive maintenance.

[0045] In actual operation, according to testing and statistics, the air compressor station achieved a comprehensive energy saving rate of 30%, the pipeline pressure fluctuation was controlled within ±0.08 bar, the waste heat recovery could meet 60% of the domestic hot water demand of the plant area, and the equipment failure downtime was reduced by 40%, achieving significant energy-saving benefits and improved operational stability.

[0046] The above embodiments only illustrate preferred embodiments of this application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of this application's patent. It should be noted that those skilled in the art can make various modifications, improvements, and substitutions without departing from the concept of this application, and these all fall within the protection scope of this application.

Claims

1. An intelligent energy-saving system for an air compressor station, characterized in that, It includes multi-source sensing units, edge intelligent decision-making units, and end-to-end execution units; The full-link execution unit includes an air compressor cluster unit, a staged air storage unit, a drying and purification unit, and a waste heat recovery unit; The multi-source sensing unit is used to collect the operating data, pipeline condition data, environmental data and production plan data of each subsystem of the full-link execution unit, and transmit them to the edge intelligent decision-making unit; The edge intelligent decision-making unit has a built-in multimodal load prediction module and a global multi-objective optimization module, which are used to generate gas load prediction results based on the collected data, solve for the whole-link collaborative scheduling scheme, and send the scheduling instructions to each subsystem of the whole-link execution unit. The air compressor cluster unit, the staged air storage unit, the drying and purification unit, and the waste heat recovery unit are used to receive scheduling instructions and operate in coordination to achieve full-chain control of air production, air storage, purification, and waste heat recovery.

2. The intelligent energy-saving system for an air compressor station according to claim 1, characterized in that, The data collection dimensions of the multi-source sensing unit include: air compressor outlet pressure, temperature, current, and vibration data; air tank graded pressure data; dryer dew point and energy consumption data; waste heat recovery system temperature and flow data; pipeline end pressure data; ambient temperature and humidity data; and production shift gas usage plan data.

3. The intelligent energy-saving system for an air compressor station according to claim 1, characterized in that, The multimodal load prediction module adopts a prediction model that combines LSTM and Transformer. The input features include historical gas load, pipeline pressure, equipment status, environmental weather, and production plan. The output is a gas load prediction curve for the next 1 to 24 hours. The global multi-objective optimization module adopts the improved NSGA-II algorithm. The optimization objectives include minimizing the total energy consumption of the entire station, minimizing pipeline pressure fluctuations, optimizing the average load rate of equipment, and maximizing the waste heat recovery rate. The constraints include the allowable range of gas supply pressure, the rated operating range of equipment, and the threshold for meeting the drying dew point standard.

4. The intelligent energy-saving system for an air compressor station according to claim 1, characterized in that, The air compressor cluster unit includes variable frequency and fixed frequency air compressors, which, together with the staged air storage unit, achieve zero unloading operation and maintain the equipment load rate in the high-efficiency range of 70% to 85%. The drying and purification unit performs dew point adaptive adjustment, dynamically matching the drying depth according to the air load and ambient humidity. The waste heat recovery unit performs load linkage control, prioritizing air supply during peak air consumption and maximizing waste heat recovery during off-peak air consumption.

5. The intelligent energy-saving system for an air compressor station according to claim 1, characterized in that, The edge intelligent decision-making unit also has a built-in closed-loop self-learning and predictive maintenance module, which is used to collect the operation effect data of the execution unit in real time, correct the parameters of the load prediction and optimization model, and at the same time perform health status assessment and fault warning based on equipment vibration and temperature data.

6. A method for intelligent energy-saving control of an air compressor station, characterized in that, The intelligent energy-saving system for air compressor stations according to any one of claims 1-5 includes the following steps: S1. Multi-source data acquisition: Collects operational data, pipeline condition data, environmental data, and production plan data of the entire air compressor station chain through multi-source sensing units; S2. Multimodal advance load forecasting: Based on the collected multi-dimensional data, a gas load forecast curve for the next 1 to 24 hours is generated; S3. Global multi-objective optimization solution: Taking the minimum total energy consumption of the whole station, the minimum pipeline pressure fluctuation, the optimal equipment load rate, and the maximum waste heat recovery rate as the optimization objectives, the scheduling scheme of air compressor cluster, staged air storage, drying and purification, and waste heat recovery is obtained. S4. Full-link coordinated execution: The scheduling scheme is broken down into control commands for each subsystem, and coordinated control of gas production scheduling, gas storage charging and discharging, dew point adjustment, and waste heat linkage is executed synchronously. S5. Closed-loop feedback self-learning: Collects actual system operation data, reversely corrects the prediction model and optimization algorithm parameters, and continuously iterates to optimize control accuracy.

7. The intelligent energy-saving control method for an air compressor station according to claim 6, characterized in that, In step S2, a prediction model combining LSTM and Transformer is used for load forecasting. The input features include historical gas load, pipeline pressure, equipment status, environmental weather, and production plan.

8. The intelligent energy-saving control method for an air compressor station according to claim 6, characterized in that, In step S3, the improved NSGA-II algorithm is used to solve the multi-objective optimization problem, with the allowable range of gas supply pressure, the rated operating range of the equipment, and the threshold for meeting the drying dew point standard as constraints.

9. The intelligent energy-saving control method for an air compressor station according to claim 6, characterized in that, Step S4 specifically includes: coordinating the charging and discharging of the graded gas storage with the air compressor scheduling to ensure that the air compressor operates in the high-efficiency load range of 70% to 85% and avoids unloading and idling; dynamically adjusting the drying depth according to the real-time gas load and ambient humidity; and establishing a linkage mechanism between the gas load and the waste heat recovery power to achieve energy cascade utilization.

10. The intelligent energy-saving control method for an air compressor station according to claim 6, characterized in that, Step S5 also includes: assessing the health status of the equipment based on the collected equipment vibration, temperature, and current data, providing early warnings of potential faults, and realizing predictive maintenance.