Variable frequency fan control system

Abstract

The invention discloses a frequency conversion fan control system which comprises a sensor network module used for collecting various data of environment and fan operation, including temperature, humidity, pressure, vibration and air quality; the data processing module is used for carrying out preprocessing, denoising and data fusion on the data acquired by the sensor network module and providing standardized input for the machine learning module; the machine learning model integrates a self-adaptive control algorithm based on machine learning and is used for learning an optimal fan control strategy from historical data and sensor real-time data; and the controller module is responsible for receiving and executing the control instruction of the machine learning module. According to the method, dynamic adjustment can be carried out in combination with real-time data of the sensors, the method adapts to complex and variable environmental conditions, the method has very high flexibility and expandability, and the system can add more types of sensors or adjust a machine learning model according to different scenes and requirements, so that the system adapts to more complex application scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of variable frequency fan control, and in particular to a variable frequency fan control system. Background Art

[0002] In recent years, variable frequency fans are devices that adjust the fan speed by changing the power supply frequency of the fan motor. Compared with traditional fixed-speed fans, variable frequency fans can dynamically adjust the speed according to actual needs, thereby improving energy efficiency and saving energy. Variable frequency fan control system is a system that controls the fan speed by adjusting the power supply frequency of the fan motor. The system usually includes components such as sensors, controllers and inverters.

[0003] In the current market, wind turbine control systems in existing technologies usually use single sensor data to execute simple control strategies. Systems that lack data fusion find it difficult to comprehensively analyze complex environmental conditions, making it difficult to provide intelligent control strategies. Lacking adaptive control methods based on historical data and machine learning, the system finds it difficult to predict and dynamically adjust future environmental changes. Summary of the invention

[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.

[0005] In view of the above problems existing in the existing variable frequency fan control system, the present invention is proposed.

[0006] Therefore, the purpose of the present invention is to provide a variable frequency fan control system, which is suitable for solving the problem that there is a lack of adaptive control methods based on historical data and machine learning, and the system is difficult to predict and dynamically adjust future environmental changes.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a variable frequency fan control system, comprising:

[0008] Sensor network module: used to collect various data about the environment and fan operation, including temperature, humidity, pressure, vibration, and air quality;

[0009] Data processing module: pre-processes, denoises and fuses the data collected by the sensor network module to provide standardized input for the machine learning module;

[0010] Machine learning model: This model integrates a machine learning-based adaptive control algorithm to learn the best wind turbine control strategy from historical data and real-time sensor data;

[0011] Controller module: responsible for receiving and executing control instructions from the machine learning module.

[0012] As a preferred solution of the variable frequency fan control system described in the present invention, it is assumed that the input data set where x i represents the input feature, y i represents the target value, N is the number of samples, and a prediction error calculation formula is constructed according to the adaptive control algorithm of machine learning. The prediction error calculation formula is as follows:

[0013]

[0014] E i Represents the prediction error of the i-th sample, which is used to measure the difference between the model's prediction of the sample and the true value; y i represents the true label of the i-th sample and represents the target value in the data set; is the predicted value of the i-th sample, which is calculated by the model based on the input feature x i Make predictions; It is part of the cross entropy loss function and is used to calculate the prediction error of the sample.

[0015] As a preferred solution of the variable frequency fan control system described in the present invention, wherein: in the output result of the prediction error calculation formula, the prediction error threshold is set to τ1;

[0016] If E i >τ1, indicating that the model has a large prediction error for this sample and further parameter optimization is needed;

[0017] If E i ≤τ1, indicating that the model has a small prediction error for this sample and no further parameter optimization is required.

[0018] As a preferred solution of the variable frequency fan control system described in the present invention, an average error calculation formula is constructed according to the output result of the prediction error calculation formula, and the average error calculation formula is as follows:

[0019]

[0020] is the average prediction error, which represents the average value of the prediction error on the entire training set; N is the number of samples, which represents the number of samples in the training set.

[0021] As a preferred solution of the variable frequency fan control system described in the present invention, wherein: in the output result of the average error calculation formula, the average prediction error threshold is set to τ2;

[0022] like It means that the model has a large error on the entire training set and needs to be adjusted;

[0023] like Indicates that the model meets expectations and no adjustment is needed.

[0024] As a preferred solution of the variable frequency fan control system described in the present invention, a regularized loss weighted total error formula is constructed according to the input result of the average error calculation formula, and the regularized loss weighted total error formula is as follows:

[0025]

[0026] in, is the total loss function, which combines the average error and the regularization term to measure the overall error of the model, taking into account both prediction accuracy and model complexity; λ is the regularization coefficient, which is used to control model complexity and prevent overfitting; ||θ|| 2 Represents the sum of squares of model parameters and is used to limit the complexity of the model.

[0027] As a preferred solution of the variable frequency fan control system described in the present invention, wherein: in the output result of the regularized loss weighted total error formula, the total loss threshold is set to τ3;

[0028] like It means that the total loss of the model is large and the model parameters need to be adjusted;

[0029] like It means that the total loss of the model is small and the performance meets expectations.

[0030] As a preferred solution of the variable frequency fan control system described in the present invention, when the total loss of the model is large, it is also adjusted through an optimization algorithm.

[0031] As a preferred solution of a variable frequency fan control system described in the present invention, the control system also includes a remote monitoring and cloud platform, which is used to upload the data of the fan system to the cloud and provide remote control and maintenance functions. The remote monitoring and cloud platform is connected to the controller module through a network.

[0032] As a preferred solution of the variable frequency fan control system described in the present invention, the controller module adjusts the input frequency of the fan through a variable frequency driver to directly control the rotation speed and output power of the fan.

[0033] Beneficial effects of the present invention:

[0034] By introducing the sensor network module, the present invention can monitor the environment and the operating status of the fan in real time, and collect data in multiple dimensions including temperature, humidity, pressure, vibration, air quality, etc. This multi-dimensional data collection not only provides comprehensive environmental information, but also enhances the system's ability to adapt to different environmental changes, so that the fan can be adaptively controlled according to real-time data;

[0035] The optimal control strategy for wind turbines is learned through machine learning models. The model can not only extract rules from historical data, but also dynamically adjust based on real-time sensor data to adapt to complex and changing environmental conditions. This adaptive control strategy can effectively improve the operating efficiency of wind turbines and reduce energy consumption.

[0036] This system relies on modular designs such as sensor networks and machine learning models, and has strong flexibility and scalability. The system can add more types of sensors or adjust machine learning models according to different scenarios and needs to adapt to more complex application scenarios and further enhance its market application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0038] Figure 1 This is a schematic diagram of the overall framework of a variable frequency fan control system proposed by the present invention;

[0039] Figure 2 A schematic diagram showing the energy consumption and time comparison structure of a variable frequency fan control system proposed in the present invention compared with an existing system. DETAILED DESCRIPTION

[0040] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0041] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0042] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0043] Secondly, the present invention is described in detail with reference to the schematic diagram. When describing the embodiments of the present invention in detail, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0044] Embodiment 1

[0045] Reference Figure 1 , is an embodiment of the present invention, and provides a variable frequency fan control system, comprising:

[0046] Sensor network module: used to collect various data on the environment and fan operation, including temperature, humidity, pressure, vibration, and air quality. This module collects multi-dimensional data on the environment and fan operation status, such as temperature, humidity, pressure, vibration, and air quality. The collaborative work of multiple sensors can not only fully grasp the environmental conditions, but also provide comprehensive data support for the precise control of the system. By collecting real-time data of different dimensions through multiple sensors, the incomplete data caused by a single sensor is avoided.

[0047] Data processing module: pre-processes, denoises and fuses the data collected by the sensor network module to provide standardized input for the machine learning module;

[0048] Machine learning model: This model integrates an adaptive control algorithm based on machine learning to learn the best fan control strategy from historical data and real-time sensor data; it learns the optimal control strategy from historical data and real-time sensor data and dynamically adjusts the fan speed and motor frequency. This innovative control algorithm goes beyond traditional feedback control and achieves intelligent self-learning and optimization through machine learning technology. It can respond quickly to real-time changes in the external environment. The machine learning model is adaptive and can optimize the control strategy from the continuously accumulated historical data, so as to achieve the best performance and lowest energy consumption in future fan operation.

[0049] Controller module: responsible for receiving and executing control instructions from the machine learning module; compared with traditional controllers, the controller module of the present invention combines intelligent algorithms to make control instructions more precise and dynamic, and the speed and output power of the fan can be adjusted in real time as the environment changes.

[0050] The control system also includes a remote monitoring and cloud platform, which is used to upload the data of the fan system to the cloud and provide remote control and maintenance functions. The remote monitoring and cloud platform is connected to the controller module through the network; users can monitor and adjust the system anytime and anywhere through the network. The remote monitoring function not only enhances the maintainability and flexibility of the system, but also enables big data analysis of the fan operation status to further optimize the control strategy.

[0051] The controller module adjusts the input frequency of the fan through the variable frequency drive and directly controls the speed and output power of the fan.

[0052] Assume the input dataset is where x i represents the input feature, y i Represents the target value, N is the number of samples, and the prediction error calculation formula is constructed according to the adaptive control algorithm of machine learning. The prediction error calculation formula is as follows:

[0053]

[0054] E i Represents the prediction error of the i-th sample, which is used to measure the difference between the model's prediction of the sample and the true value; y i represents the true label of the i-th sample and represents the target value in the data set; is the predicted value of the i-th sample, which is calculated by the model based on the input feature x i Make predictions; It is part of the cross entropy loss function and is used to calculate the prediction error of the sample.

[0055] In the output result of the prediction error calculation formula, the prediction error threshold is set to τ1, and the prediction error threshold is set to 0.1 in the present invention;

[0056] If E i >τ1, indicating that the model has a large prediction error for this sample and further parameter optimization is needed;

[0057] If E i ≤τ1, indicating that the model has a small prediction error for this sample and no further parameter optimization is required.

[0058] The average error calculation formula is constructed according to the output result of the prediction error calculation formula, and the average error calculation formula is as follows:

[0059]

[0060] is the average prediction error, which represents the average value of the prediction error on the entire training set; N is the number of samples, which represents the number of samples in the training set.

[0061] In the output result of the average error calculation formula, the average prediction error threshold is set to τ2. In the present invention, the average prediction error threshold is 0.05;

[0062] like It means that the model has a large error on the entire training set and needs to be adjusted;

[0063] like Indicates that the model meets expectations and no adjustment is needed.

[0064] The regularized loss weighted total error formula is constructed according to the input result of the average error calculation formula, and the regularized loss weighted total error formula is as follows:

[0065]

[0066] in, is the total loss function, which combines the average error and the regularization term to measure the overall error of the model, taking into account both prediction accuracy and model complexity; λ is the regularization coefficient, which is used to control model complexity and prevent overfitting; ||θ|| 2 Represents the sum of squares of model parameters and is used to limit the complexity of the model.

[0067] In the output result of the regularized loss weighted total error formula, the total loss threshold is set to τ3, and in the present invention, the total loss threshold is 0.1;

[0068] like It means that the total loss of the model is large and the model parameters need to be adjusted. When the total loss of the model is large, it can also be adjusted through the optimization algorithm;

[0069] like It means that the total loss of the model is small and the performance meets expectations.

[0070] Embodiment 2

[0071] Reference Figure 2 , which is the second embodiment of the present invention. This embodiment is different from the first embodiment in that, in order to verify its beneficial effects, experimental comparison data between the present invention and the prior art are provided.

[0072] In order to verify the advantages of our variable frequency fan control system, especially the effectiveness of its adaptive control algorithm based on machine learning model, this embodiment selects two systems for comparison: the traditional variable frequency fan control system (system A) and the variable frequency fan control system with adaptive control based on machine learning model designed by us (system B). The test environment is set to simulate the industrial environment, including scenarios with large changes in temperature, humidity, pressure and air quality. The experiment selected five typical working conditions for testing, namely: large fluctuations in ambient temperature, sudden changes in humidity, increased pressure, deterioration of air quality, and simultaneous fluctuations of multiple factors.

[0073] During the experiment, the sensor network module will collect the fan operation data and environmental parameters in real time, and transmit the data to the data processing module for denoising and standardization. Then, the machine learning model will combine these data with historical data to dynamically adjust the fan speed and motor frequency. The controller module will control the operation of the fan in real time according to the output of the machine learning model to ensure that the system can maintain efficient and stable operation under complex working conditions.

[0074] In order to fully demonstrate the advantages of our system in energy saving, stability and response speed, the following key indicators were monitored during the experiment: energy consumption, response time (the time required for the fan to adjust to a stable state from environmental changes), system stability (vibration frequency and amplitude) and air quality improvement effect (the degree of optimization of air quality through fan operation).

[0075] Table 1: Performance comparison test data of variable frequency fan control system

[0076]

[0077] It can be clearly seen from the data in the above table that system B is superior to traditional system A in various indicators such as energy consumption, response time, system stability and air quality improvement effect. The specific analysis is from the following four aspects:

[0078] Energy consumption: The energy consumption of system B under all working conditions is significantly lower than that of system A, saving an average of about 15% to 20% of electricity. This shows that by optimizing the fan speed and frequency through machine learning algorithms, system B can dynamically adjust the fan operation according to actual needs, avoiding unnecessary energy waste;

[0079] In terms of response time: System B responds much faster than System A, and can adjust the fan's operating state more quickly when the environment changes greatly. In particular, when humidity changes suddenly and air quality deteriorates, the response time is shortened by half. This reflects the efficiency and real-time performance of our adaptive control algorithm, which enables the system to quickly adapt to environmental changes and ensure that the fan operates in the best condition.

[0080] System stability: Vibration frequency and amplitude are key indicators for measuring the stability of fan operation. The vibration frequency and amplitude of system B are significantly lower than those of system A, indicating that it operates more smoothly, causes less wear and tear on the equipment during long-term operation, and greatly improves system reliability.

[0081] In terms of air quality improvement effect: under conditions of deteriorating air quality and fluctuations in multiple factors, the air quality improvement effect of system B is significantly better than that of system A, with an improvement of 15% to 20%. This shows that the machine learning model can not only optimize the working state of the fan, but also reasonably adjust the fan output according to environmental data, thereby more effectively improving air quality.

[0082] Through the above data comparison: the variable frequency fan control system invented by us, through integrated sensor network, data fusion processing and machine learning adaptive control algorithm, has significantly improved the system's energy utilization efficiency, response speed and operation stability, and has obvious advantages in dealing with complex environmental conditions.

[0083] The present invention significantly improves the efficiency and accuracy of variable frequency fan control by organically combining technologies such as multi-sensor network, data preprocessing and fusion, and machine learning adaptive control. It has a wide range of uses and is suitable for fields with high requirements for environmental control. It has good energy-saving effects, flexibility, and scalability, and can effectively overcome the limitations of existing technologies.

[0084] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A variable frequency fan control system, characterized in that: include: Sensor network module: used to collect various data about the environment and fan operation, including temperature, humidity, pressure, vibration, and air quality; Data processing module: pre-processes, denoises and fuses the data collected by the sensor network module to provide standardized input for the machine learning module; Machine learning model: This model integrates a machine learning-based adaptive control algorithm to learn the best wind turbine control strategy from historical data and real-time sensor data; Controller module: responsible for receiving and executing control instructions from the machine learning module.

2. A variable frequency fan control system according to claim 1, characterized in that: Assume the input dataset is where x i represents the input feature, y i represents the target value, N is the number of samples, and a prediction error calculation formula is constructed according to the adaptive control algorithm of machine learning. The prediction error calculation formula is as follows: E i Represents the prediction error of the i-th sample, which is used to measure the difference between the model's prediction of the sample and the true value; y i represents the true label of the i-th sample and represents the target value in the data set; is the predicted value of the i-th sample, which is calculated by the model based on the input feature x i Make predictions; It is part of the cross entropy loss function and is used to calculate the prediction error of the sample.

3. A variable frequency fan control system according to claim 2, characterized in that: In the output result of the prediction error calculation formula, the prediction error threshold is set to τ1; If E i >τ1, indicating that the model has a large prediction error for this sample and further parameter optimization is needed; If E i ≤τ1, indicating that the model has a small prediction error for this sample and no further parameter optimization is required.

4. A variable frequency fan control system according to claim 2, characterized in that: The average error calculation formula is constructed according to the output result of the prediction error calculation formula, and the average error calculation formula is as follows: is the average prediction error, which represents the average value of the prediction error on the entire training set; N is the number of samples, which represents the number of samples in the training set.

5. A variable frequency fan control system according to claim 4, characterized in that: In the output result of the average error calculation formula, the average prediction error threshold is set to τ2; like It means that the model has a large error on the entire training set and needs to be adjusted; like Indicates that the model meets expectations and no adjustment is needed.

6. A variable frequency fan control system according to claim 4, characterized in that: The regularized loss weighted total error formula is constructed according to the input result of the average error calculation formula, and the regularized loss weighted total error formula is as follows: in, is the total loss function, which combines the average error and the regularization term to measure the overall error of the model, taking into account both prediction accuracy and model complexity; λ is the regularization coefficient, which is used to control model complexity and prevent overfitting; ||θ|| 2 Represents the sum of squares of model parameters and is used to limit the complexity of the model.

7. A variable frequency fan control system according to claim 6, characterized in that: In the output of the regularized loss weighted total error formula, the total loss threshold is set to τ3; like It means that the total loss of the model is large and the model parameters need to be adjusted; like It means that the total loss of the model is small and the performance meets expectations.

8. A variable frequency fan control system according to claim 7, characterized in that: When the total loss of the model is large, it is also adjusted through the optimization algorithm.

9. A variable frequency fan control system according to claim 1, characterized in that: The control system also includes a remote monitoring and cloud platform, which is used to upload the data of the fan system to the cloud and provide remote control and maintenance functions. The remote monitoring and cloud platform is connected to the controller module through a network.

10. A variable frequency fan control system according to claim 1, characterized in that: The controller module adjusts the input frequency of the fan through a variable frequency driver to directly control the speed and output power of the fan.

Images