An AI-based secondary energy-saving control system and method for old frequency converter equipment

Abstract

The application relates to the technical field of industrial frequency conversion energy-saving control, AI edge computing and motor energy efficiency optimization, and discloses an AI-based secondary energy-saving control system and method for old frequency converter equipment, wherein the AI-based secondary energy-saving control system for old frequency converter equipment comprises an operation data acquisition module, an AI edge computing module, a frequency conversion instruction output module, a safety switching module and a communication module, and the modules are electrically connected; the application has low modification cost and is convenient to deploy; the system is connected in parallel with the original old frequency converter, the whole frequency converter does not need to be replaced, the system does not need to be shut down during the modification process, the wiring is simple, the deployment speed is fast, the equipment procurement cost and the construction cost are greatly reduced, and the loss caused by the interruption of enterprise production is reduced; through AI edge computing, real-time identification of the load and dynamic optimization of energy efficiency are realized, the optimal operation frequency is accurately output, compared with the original operation mode of the old frequency converter, the comprehensive energy-saving rate of the motor system can reach 15%-35%, and the industrial energy consumption is greatly reduced.

Description

Technical Field

[0001] This invention belongs to the fields of industrial frequency conversion energy-saving control, AI edge computing, and motor energy efficiency optimization technology, specifically a secondary energy-saving control system and method for old frequency converter equipment based on AI. Background Technology

[0002] In industrial production sites, frequency converters are widely used in equipment such as fans, water pumps, air compressors, and central air conditioning systems, and many of these frequency converters are old and have been in use for a long time. These old frequency converters generally use a fixed frequency and manually set parameters for control, which cannot perform intelligent energy efficiency optimization based on real-time load changes. Under light load and variable load conditions, the motor operating efficiency is low, resulting in serious energy waste.

[0003] Traditional frequency converter energy-saving retrofit solutions require the complete replacement of old frequency converters. This not only results in high procurement costs for the entire frequency converter unit but also necessitates downtime for construction, leading to complex and lengthy processes and production interruptions, thus reducing companies' willingness to implement energy-saving retrofits. Furthermore, current frequency converter control technologies on the market lack AI self-learning and load-adaptive capabilities. Even with simple parameter adjustments, they cannot dynamically optimize based on long-term changes in load characteristics, making it difficult to maintain optimal long-term operating efficiency of the equipment.

[0004] To address the aforementioned technical issues, there is an urgent need to develop a secondary energy-saving control technology that requires no replacement of the original frequency converter, has low retrofit costs, and possesses intelligent adaptive capabilities. This technology would enable uninterrupted energy-saving retrofits of aging frequency converter systems and improve the overall energy efficiency of motor systems.

[0005] Based on this, an AI-based secondary energy-saving control system and method for old frequency converter equipment is designed. Summary of the Invention

[0006] In view of the above situation and to overcome the shortcomings of the prior art, the present invention provides a secondary energy-saving control system and method for old frequency converter equipment based on AI, which effectively solves the problems mentioned in the background.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a secondary energy-saving control system for old frequency converter equipment based on AI, comprising an operation data acquisition module, an AI edge computing module, a frequency converter command output module, a safety switching module, and a communication module, wherein each module is electrically connected; The operation data acquisition module is used to collect the operating parameters of the motor and load in real time. The operating parameters include motor voltage, current, power, power factor, load rate, and pipeline pressure and flow rate. The AI ​​edge computing module is equipped with an AI load identification model and an energy efficiency optimization algorithm. It is used to perform local real-time inference and parameter optimization on the collected operating parameters, identify load type, load rate and energy efficiency inflection point and predict the optimal operating frequency, and iteratively update the AI ​​load identification model. The frequency conversion command output module is connected to the old frequency converter via an analog interface or an RS485 interface to send the optimal frequency command output by the AI ​​edge computing module to the old frequency converter. The safety switching module is used to monitor the operating status of the motor and the frequency converter, and automatically switch back to the control logic of the old frequency converter under abnormal conditions, including overcurrent, overload, and stall. The communication module supports the Modbus industrial protocol and is used to upload operating parameters and energy-saving data to the remote monitoring platform, while also enabling remote debugging of the system.

[0008] Preferably, the AI ​​edge computing module uses the RK3568G edge computing unit as the hardware core, the AI ​​load identification model is a lightweight load classification neural network, and the energy efficiency optimization algorithm is an energy efficiency curve optimization algorithm.

[0009] Preferably, the analog interface of the frequency converter command output module includes a 0-10V analog interface and a 4-20mA analog interface, and the frequency converter command output module is connected in parallel with the control terminals of the existing old frequency converter.

[0010] Preferably, the safety switching module has a safety switching protection loop, which is electrically connected to the AI ​​edge computing module, the frequency conversion command output module, and the existing old frequency converter to achieve dual-loop control.

[0011] Preferably, the operating data acquisition module is equipped with an isolation sampling circuit for initial electrical isolation and signal acquisition of the acquired operating parameters.

[0012] A secondary energy-saving control method for aging frequency converter equipment based on AI, applied to the aforementioned secondary energy-saving control system, includes the following steps: Step 1: Real-time acquisition of operating parameters: The operating parameters of the motor and load are acquired through the operating data acquisition module. The operating parameters include motor voltage, current, power, power factor, load rate, and pipeline pressure and flow rate. Step 2, Data Filtering and Preprocessing: The operating parameters collected in Step 1 are filtered and noise-reduced to remove invalid data and interference signals, and standardized effective operating parameters are obtained. Step 3: AI Load Identification and Energy Efficiency Optimization: The effective operating parameters obtained in Step 2 are input into the AI ​​edge computing module. A lightweight load classification neural network identifies the load type, load rate, and energy efficiency inflection point. Combined with an energy efficiency curve optimization algorithm, local real-time inference is performed to predict the optimal operating frequency for the motor's next operating cycle. Step 4: Optimal Frequency Command Output: The AI ​​edge computing module transmits the predicted optimal operating frequency to the frequency converter command output module. The frequency converter command output module sends the optimal frequency command to the old frequency converter through the analog interface or RS485 interface, replacing or superimposing the given signal of the old frequency converter, and realizing closed-loop secondary control of the old frequency converter. Step 5: Variable frequency drive motor operation: After receiving the optimal frequency command, the old variable frequency drive adjusts its output frequency according to the command, and drives the motor and load to run at the optimal frequency. Step 6: Operation Status and Energy Efficiency Monitoring: The operation data acquisition module continuously collects energy efficiency data and equipment operation status data after the motor runs, and transmits the data to the AI ​​edge computing module and the safety switching module in real time; Step 7, Anomaly Judgment and Handling: The safety switching module monitors the operating status data collected in Step 6 in real time. If abnormal conditions such as overcurrent, overload, or stall are detected, the safety switching protection circuit is immediately triggered, and the original control logic of the old frequency converter is automatically switched back. If the detected condition is normal operating condition, Step 8 is executed. Step 8: AI Model Self-Update and Iterative Optimization: The AI ​​edge computing module compares and analyzes the energy efficiency data collected in Step 6 with the energy efficiency results of the current optimization strategy. Based on the analysis results, iteratively updates the parameters of the AI ​​load identification model and the energy efficiency optimization algorithm to improve the optimization accuracy. Then, it returns to Step 1 to carry out the next round of operating parameter collection and closed-loop control.

[0013] Preferably, in step four, the frequency conversion command output module is connected in parallel with the original old frequency converter, and the main control circuit and control logic of the original old frequency converter are not modified during the connection process.

[0014] Preferably, in step six, the energy efficiency data includes the motor's real-time energy consumption, power factor, load rate, and energy efficiency value, and the equipment operating status data includes the inverter output frequency, motor current and voltage, and equipment operating fault codes.

[0015] Preferably, in step eight, the model self-update of the AI ​​edge computing module is a local offline update, which does not interrupt the normal operation and energy-saving control of the device during the update process.

[0016] Preferably, the above-mentioned AI-based secondary energy-saving control system or secondary energy-saving control method for old frequency converter equipment is applied to the energy-saving retrofit of in-use old frequency converter systems such as fans, water pumps, air compressors, central air conditioning, and industrial water supply.

[0017] Compared with the prior art, the beneficial effects of the present invention are: 1. The present invention has low transformation cost and convenient deployment. The system of the present invention is connected in parallel with the original old frequency converter, without the need to replace the entire frequency converter. There is no need to stop the machine during the transformation process. The wiring is simple and the deployment speed is fast, which greatly reduces the equipment procurement cost and construction cost, and reduces the loss of production interruption for enterprises. 2. This invention has significant energy-saving effects. It achieves real-time load identification and dynamic optimization of energy efficiency through AI edge computing, and accurately outputs the optimal operating frequency. Compared with the original operation mode of old frequency converters, the comprehensive energy saving rate of the motor system can reach 15%-35%, which greatly reduces industrial energy consumption. 3. Intelligent self-adaptation, no manual adjustment required. The lightweight load classification neural network and energy efficiency curve optimization algorithm have self-learning and self-optimization capabilities. They can iteratively update model parameters according to long-term changes in load characteristics, eliminating the need for frequent manual adjustment of inverter parameters and reducing manual maintenance costs. 4. High safety and reliability: It has a safety switching protection circuit to form a dual-circuit control structure. Under abnormal operating conditions, it can automatically switch back to the original frequency converter control logic without affecting the main control function of the original equipment. It provides dual protection for the safe and stable operation of the equipment from both hardware and software levels. 5. It has strong versatility and wide adaptability. The system is equipped with multiple types of interfaces, supports 0-10V, 4-20mA analog signals and RS485 / Modbus communication, and is compatible with old frequency converters of various brands and eras. It can be widely used for energy-saving retrofitting of various industrial motor loads such as fans, water pumps, air compressors, central air conditioning, and industrial water supply.

[0018] 6. Local real-time computing and remote visualization: The RK3568G edge computing unit enables local real-time inference and optimization without relying on cloud servers, avoiding network latency from affecting control accuracy. It also supports data upload and remote debugging via the Modbus protocol, enabling remote visual monitoring of device operating status and improving operation and maintenance efficiency. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used together with the embodiments of the invention to explain the invention and do not constitute a limitation thereof.

[0020] In the attached diagram: Figure 1 This is a schematic diagram of the AI ​​secondary energy-saving control method of the present invention. Detailed Implementation

[0021] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0022] Depend on Figure 1 The present invention relates to an AI-based secondary energy-saving control system for old frequency converter equipment, including an operation data acquisition module, an AI edge computing module, a frequency converter command output module, a safety switching module, and a communication module. The modules are electrically connected, and the system as a whole is connected in parallel with the original old frequency converter without changing the main control logic and circuit of the original frequency converter.

[0023] Operational data acquisition module: Equipped with an isolated sampling circuit, it achieves electrical isolation and accurate signal acquisition, and is used to collect all-dimensional operating parameters of the motor and load in real time, including motor voltage, current, power, power factor, load rate, as well as pipeline pressure and flow rate, providing a complete data foundation for subsequent AI optimization.

[0024] AI Edge Computing Module: Based on the RK3568G edge computing unit as the hardware core, this module is equipped with a lightweight load classification neural network and energy efficiency curve optimization algorithm. It has local real-time inference and computing capabilities without relying on cloud servers. This module is used to perform local real-time inference and parameter optimization on the collected operating parameters, accurately identify the load type, load rate and energy efficiency inflection point, and predict the optimal operating frequency of the motor in the next operating cycle. At the same time, it can iteratively update the AI ​​load identification model based on real-time energy efficiency data to improve the optimization accuracy.

[0025] Variable frequency command output module: Equipped with a 0-10V analog interface, a 4-20mA analog interface and an RS485 interface, it can be directly connected in parallel with the control terminals of the old variable frequency drive to stably send the optimal frequency command output by the AI ​​edge computing module to the old variable frequency drive. It can replace or superimpose the given signal of the old variable frequency drive to complete the closed-loop secondary control.

[0026] Safety switching module: It is equipped with a safety switching protection loop, which is electrically connected to the AI ​​edge computing module, the frequency converter command output module, and the original old frequency converter to form a dual-loop control structure. This module is used to monitor the operating status of the motor and frequency converter in real time. When abnormal conditions such as overcurrent, overload, or stall are detected, the safety switching protection loop is immediately triggered, and the original control logic of the old frequency converter is automatically switched back to ensure the safe and stable operation of the equipment.

[0027] Communication module: Supports the Modbus industrial general protocol and has the ability to upload data and interact remotely. On the one hand, it can upload motor operating parameters, energy-saving data, and equipment status data to the remote monitoring platform in real time to realize remote data visualization; on the other hand, it can receive instructions from the remote monitoring platform to complete remote system debugging and reduce on-site maintenance costs.

[0028] A secondary energy-saving control method for aging frequency converter equipment based on AI, applied to the aforementioned secondary energy-saving control system, specifically includes the following steps: Step 1: Real-time acquisition of operating parameters: Through the isolated sampling circuit of the data acquisition module, the full-dimensional operating parameters of the motor and load are acquired in real time and accurately. The operating parameters include motor voltage, current, power, power factor, load rate, as well as pipeline pressure and flow. The acquired raw data is transmitted to the AI ​​edge computing module in real time.

[0029] Step 2, Data Filtering and Preprocessing: The AI ​​edge computing module filters and reduces noise from the received raw operating parameters. It uses digital filtering algorithms to remove invalid data, interference signals, and outliers, and standardizes and normalizes the data to obtain effective operating parameters that meet the input requirements of the AI ​​model, ensuring the accuracy of subsequent inference and optimization.

[0030] Step 3, AI Load Identification and Energy Efficiency Optimization: The standardized effective operating parameters are input into the lightweight load classification neural network of the AI ​​edge computing module. The neural network accurately identifies the current load type, real-time load rate, and motor energy efficiency inflection point. Combined with the energy efficiency curve optimization algorithm, the identification results are used for local real-time inference calculation. Based on the motor energy efficiency characteristics and real-time load changes, the optimal operating frequency of the motor in the next operating cycle is predicted, providing accurate instructions for frequency conversion control.

[0031] Step 4: Optimal Frequency Command Output: The AI ​​edge computing module transmits the predicted optimal operating frequency data to the frequency converter command output module. The frequency converter command output module selects 0-10V analog, 4-20mA analog, or RS485 communication mode according to the interface type of the old frequency converter and sends the optimal frequency command to the control terminal of the old frequency converter. This command replaces or superimposes the manual input signal of the old frequency converter, realizing closed-loop secondary control of the old frequency converter, and does not change the main control circuit and control logic of the original frequency converter during the connection process.

[0032] Step 5: Variable frequency drive drives the motor: After receiving the optimal frequency command sent by the variable frequency command output module, the old variable frequency drive automatically adjusts its own output frequency according to the command, driving the motor and load to run at the optimal frequency, thereby optimizing the motor's energy efficiency.

[0033] Step Six: Operation Status and Energy Efficiency Monitoring: The operation data acquisition module continuously collects energy efficiency data and equipment operation status data after the motor operates at the optimal frequency. The energy efficiency data includes the motor's real-time energy consumption, power factor, load rate, and energy efficiency value. The equipment operation status data includes the inverter output frequency, motor current and voltage, and equipment operation fault codes. The collected data is transmitted synchronously to the AI ​​edge computing module and the safety switching module in real time to provide data support for model updates and safety monitoring.

[0034] Step 7, Anomaly Judgment and Handling: The safety switching module monitors and analyzes the received equipment operating status data in real time. If abnormal conditions such as overcurrent, overload, or stall are detected, the safety switching protection circuit is immediately triggered to cut off the frequency command output of the system and automatically switch back to the original control logic of the old frequency converter until the equipment returns to normal. If the monitoring result indicates that the equipment is operating normally, the energy efficiency data and status data are transmitted to the AI ​​edge computing module, and step 8 is executed.

[0035] Step 8: AI Model Self-Update and Iterative Optimization: The AI ​​edge computing module compares and analyzes the real-time collected energy efficiency data with the energy efficiency results under the current optimization strategy, calculates the energy efficiency improvement rate and optimization deviation, and iteratively updates the weight parameters of the AI ​​load identification model and the calculation parameters of the energy efficiency optimization algorithm based on the analysis results, continuously improving the model's load identification accuracy and energy efficiency optimization accuracy. The model update is a local offline update, which does not interrupt the normal operation and energy-saving control of the equipment. After the update is completed, it returns to Step 1 to carry out the next round of operating parameter collection and closed-loop control, realizing long-term dynamic self-optimization.

[0036] The present invention relates to an AI-based secondary energy-saving control system for old frequency converter equipment. The core hardware adopts the RK3568G edge computing unit, which is equipped with an isolation sampling circuit, a command output circuit, a safety switching circuit, and a communication circuit. Each circuit corresponds to one of the five major modules of the system. The AI ​​algorithm adopts a lightweight load classification neural network + energy efficiency curve optimization algorithm. The algorithm is embedded in the RK3568G edge computing unit to achieve local real-time operation.

[0037] The system deployment process is as follows: wiring and power-on → equipment self-learning → automatic commissioning → continuous AI optimization. Specifically: the system's frequency converter command output module is connected in parallel with the control terminals of the existing old frequency converter, and the operation data acquisition module is connected to the sensor detection terminals of the motor and load. After wiring is completed, the system is powered on; the system automatically collects 30 minutes of raw operating data from the equipment to complete the initial self-learning of the AI ​​model and identify the basic load characteristics of the equipment; after self-learning is completed, the system is automatically commissioned and enters closed-loop energy regulation; after commissioning, the system continuously updates the AI ​​model based on real-time energy efficiency data to achieve long-term dynamic optimization.

Claims

1. A secondary energy-saving control system for old frequency converter equipment based on AI, characterized in that, It includes a data acquisition module, an AI edge computing module, a frequency conversion command output module, a safety switching module, and a communication module, with each module electrically connected. The operation data acquisition module is used to collect the operating parameters of the motor and load in real time. The operating parameters include motor voltage, current, power, power factor, load rate, and pipeline pressure and flow rate. The AI ​​edge computing module is equipped with an AI load identification model and an energy efficiency optimization algorithm. It is used to perform local real-time inference and parameter optimization on the collected operating parameters, identify load type, load rate and energy efficiency inflection point and predict the optimal operating frequency, and iteratively update the AI ​​load identification model. The frequency conversion command output module is connected to the old frequency converter via an analog interface or an RS485 interface to send the optimal frequency command output by the AI ​​edge computing module to the old frequency converter. The safety switching module is used to monitor the operating status of the motor and the frequency converter, and automatically switch back to the control logic of the old frequency converter under abnormal conditions, including overcurrent, overload, and stall. The communication module supports the Modbus industrial protocol and is used to upload operating parameters and energy-saving data to the remote monitoring platform, while also enabling remote debugging of the system.

2. The secondary energy-saving control system for old frequency converter equipment based on AI according to claim 1, characterized in that, The AI ​​edge computing module uses the RK3568G edge computing unit as its hardware core, the AI ​​load identification model is a lightweight load classification neural network, and the energy efficiency optimization algorithm is an energy efficiency curve optimization algorithm.

3. The secondary energy-saving control system for old frequency converter equipment based on AI according to claim 1, characterized in that, The analog interface of the frequency converter command output module includes a 0-10V analog interface and a 4-20mA analog interface. The frequency converter command output module is connected in parallel with the control terminals of the existing old frequency converter.

4. The secondary energy-saving control system for old frequency converter equipment based on AI according to claim 1, characterized in that, The safety switching module is equipped with a safety switching protection loop, which is electrically connected to the AI ​​edge computing module, the frequency conversion command output module, and the existing old frequency converter to achieve dual-loop control.

5. The secondary energy-saving control system for old frequency converter equipment based on AI according to claim 1, characterized in that, The operational data acquisition module is equipped with an isolation sampling circuit, which is used to perform initial electrical isolation and signal acquisition of the acquired operational parameters.

6. A secondary energy-saving control method for aging frequency converter equipment based on AI, applied to the secondary energy-saving control system for aging frequency converter equipment based on AI as described in any one of claims 1-5, characterized in that, Includes the following steps: Step 1: Real-time acquisition of operating parameters: The operating parameters of the motor and load are acquired through the operating data acquisition module. The operating parameters include motor voltage, current, power, power factor, load rate, and pipeline pressure and flow rate. Step 2, Data Filtering and Preprocessing: The operating parameters collected in Step 1 are filtered and noise-reduced to remove invalid data and interference signals, and standardized effective operating parameters are obtained. Step 3: AI Load Identification and Energy Efficiency Optimization: The effective operating parameters obtained in Step 2 are input into the AI ​​edge computing module. A lightweight load classification neural network identifies the load type, load rate, and energy efficiency inflection point. Combined with an energy efficiency curve optimization algorithm, local real-time inference is performed to predict the optimal operating frequency for the motor's next operating cycle. Step 4: Optimal Frequency Command Output: The AI ​​edge computing module transmits the predicted optimal operating frequency to the frequency converter command output module. The frequency converter command output module sends the optimal frequency command to the old frequency converter through the analog interface or RS485 interface, replacing or superimposing the given signal of the old frequency converter, and realizing closed-loop secondary control of the old frequency converter. Step 5: Variable frequency drive motor operation: After receiving the optimal frequency command, the old variable frequency drive adjusts its output frequency according to the command, and drives the motor and load to run at the optimal frequency. Step 6: Operation Status and Energy Efficiency Monitoring: The operation data acquisition module continuously collects energy efficiency data and equipment operation status data after the motor runs, and transmits the data to the AI ​​edge computing module and the safety switching module in real time; Step 7, Anomaly Judgment and Handling: The safety switching module monitors the operating status data collected in Step 6 in real time. If abnormal conditions such as overcurrent, overload, or stall are detected, the safety switching protection circuit is immediately triggered, and the original control logic of the old frequency converter is automatically switched back. If the detected condition is normal operating condition, Step 8 is executed. Step 8: AI Model Self-Update and Iterative Optimization: The AI ​​edge computing module compares and analyzes the energy efficiency data collected in Step 6 with the energy efficiency results of the current optimization strategy. Based on the analysis results, iteratively updates the parameters of the AI ​​load identification model and the energy efficiency optimization algorithm to improve the optimization accuracy. Then, it returns to Step 1 to carry out the next round of operating parameter collection and closed-loop control.

7. The secondary energy-saving control method for old frequency converter equipment based on AI according to claim 6, characterized in that, In step four, the frequency conversion command output module is connected in parallel with the original old frequency converter, and the main control circuit and control logic of the original old frequency converter are not modified during the connection process.

8. The secondary energy-saving control method for old frequency converter equipment based on AI according to claim 6, characterized in that, In step six, the energy efficiency data includes the motor's real-time energy consumption, power factor, load rate, and energy efficiency value, while the equipment operating status data includes the inverter output frequency, motor current and voltage, and equipment operating fault codes.

9. A secondary energy-saving control method for old frequency converter equipment based on AI according to claim 6, characterized in that, In step eight, the model self-update of the AI ​​edge computing module is a local offline update, which does not interrupt the normal operation and energy-saving control of the device during the update process.

10. The AI-based secondary energy-saving control system for aging frequency converters according to any one of claims 1-5, or the AI-based secondary energy-saving control method for aging frequency converters according to any one of claims 6-9, characterized in that, Energy-saving retrofitting of old variable frequency systems in use, such as fans, water pumps, air compressors, central air conditioning, and industrial water supply.

Images