Monitoring and feedback pump and fan operating environment control method and system

Abstract

The invention discloses a monitoring and feedback pump and fan operation environment control method and system, and belongs to the technical field of power equipment monitoring, and the method comprises the steps: obtaining equipment operation parameters, and constructing an equipment state monitoring model; collecting vibration, noise and temperature data in real time through the equipment state monitoring model, identifying an equipment operation state, and determining an abnormal point; dynamically generating a vibration, noise and temperature control strategy model through the abnormal point data in combination with the environmental parameters, wherein the model is used for first control of the equipment operation state; dynamic characteristics of vibration, noise, and temperature are analyzed in conjunction with the first control. Through the monitoring model, abnormity is recognized in real time, the control strategy is dynamically generated, first control is rapidly implemented, and the equipment state is stabilized. And a multi-dimensional control threshold and a self-adaptive control algorithm are generated by quantizing vibration, noise and temperature coupling effects, so that the control precision and the intelligent level are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power equipment monitoring, in particular to a pump and fan operating environment control method and system with monitoring and feedback. BACKGROUND

[0002] In modern industrial and civil fields, pumps and fans, as key fluid mechanical equipment, are widely used in water supply, heating, air conditioning, chemical industry, power and environmental protection, etc. These devices not only provide power and transport fluids, but also often produce vibrations and noises during operation. Vibration not only may cause early fatigue and damage to the structure of the equipment, but also may cause resonance of the pipeline system, thereby causing safety hazards. In addition, continuous noise pollution can have a negative impact on the hearing and mental health of workers, reduce work efficiency, and may cause dissatisfaction and complaints from surrounding residents, constituting environmental pollution problems.

[0003] Traditional vibration and noise control methods usually rely on passive vibration isolation and sound absorption measures. These methods often require rich experience and repeated tests during design and implementation, and once installed, their parameters are difficult to dynamically adjust according to actual working conditions. With the progress of industrial automation and intelligent technology, real-time monitoring and intelligent control of equipment operating state have become possible. However, current intelligent control systems mainly focus on performance monitoring and fault diagnosis of equipment, and there are still problems in comprehensive control of vibration and noise. SUMMARY

[0004] To solve the above technical problems, a pump and fan operating environment control method with monitoring and feedback is proposed, which includes obtaining device operating parameters and constructing a device state monitoring model;

[0005] Real-time collection of vibration, noise and temperature data is performed through the device state monitoring model, the device operating state is identified, and abnormal points are determined;

[0006] A vibration, noise and temperature control strategy model is dynamically generated by combining abnormal point data with environmental parameters, which is used for first control of the device operating state;

[0007] In combination with the first control, the dynamic characteristics of vibration, noise and temperature are analyzed to obtain vibration-noise-temperature multi-dimensional control thresholds for second control of the device operating state;

[0008] Based on a closed-loop feedback mechanism, a device optimization control model is generated in combination with the coupling characteristics of vibration-noise-temperature;

[0009] Through the device optimization control model, a time-varying performance optimization model is generated in combination with long-term operating data for continuous optimization control of device vibration, noise and temperature.

[0010] As a preferred scheme of the pump and fan operation environment control method of monitoring and feedback, wherein: the vibration, noise and temperature data are collected in real time through the equipment state monitoring model, the equipment operation state is identified, and the abnormal points are determined, including:

[0011] The vibration, noise and temperature data of the equipment are collected in real time through the equipment state monitoring model, and the feature distribution diagram of the equipment operation state is generated.

[0012] The maximum value region of vibration, noise or temperature is determined, and the normal operation mode, slight abnormality or serious fault state of the equipment is determined.

[0013] As a preferred scheme of the pump and fan operation environment control method of monitoring and feedback, wherein: the vibration, noise and temperature control strategy model is dynamically generated by combining the abnormal point data with the environmental parameters, including:

[0014] Based on spectrum analysis and time-frequency analysis, the characteristic frequency and transient characteristics of the abnormal points are extracted;

[0015] The damper damping coefficient, sound absorption material configuration and fan blade angle are adjusted through the characteristic frequency and transient characteristics.

[0016] As a preferred scheme of the pump and fan operation environment control method of monitoring and feedback, wherein: in combination with the first control, the dynamic characteristics of vibration, noise and temperature are analyzed to obtain the vibration-noise-temperature multi-dimensional control threshold, and the second control of the equipment operation state is performed, including:

[0017] The vibration, noise and temperature feature matrix of the abnormal points is generated through the pattern recognition algorithm;

[0018] Based on historical data and real-time data, the coupling effect of vibration, noise and temperature is quantified, and a vibration correction term is generated;

[0019] Through the environmental parameters and the equipment operation state, the influence of noise on vibration control is quantified, and a noise correction term is generated;

[0020] Based on the temperature data and the equipment operation state, the influence of temperature on vibration and noise control is quantified, and a temperature correction term is generated;

[0021] Based on the vibration correction term, the noise correction term and the temperature correction term, the vibration-noise-temperature multi-dimensional control threshold is generated in combination with the reference control parameter;

[0022] Based on the comparison between the multi-dimensional control threshold and the real-time monitoring data, the second control of the equipment operation state is performed.

[0023] As a preferred scheme of the pump and fan operation environment control method of monitoring and feedback, wherein: based on the closed-loop feedback mechanism, the coupling characteristics of vibration, noise and temperature are combined to generate a device optimization control model, including:

[0024] Based on the actual effect of the first control and the second control, a feedback matrix of vibration, noise and temperature is generated;

[0025] Combined with the coupling characteristics of vibration, noise and temperature, the deviation characteristics of the control effect are extracted;

[0026] Based on the deviation characteristics and historical control data, an adaptive control algorithm is generated;

[0027] The adaptive control algorithm includes a combination of genetic algorithm, neural network and fuzzy logic control, wherein:

[0028] Through the genetic algorithm, the control parameters are optimized based on the dynamic fitness function;

[0029] Through the neural network, the local features and time series characteristics of the vibration, noise and temperature signals are extracted;

[0030] Through the fuzzy logic control, multi-dimensional reasoning is performed based on the automatically generated fuzzy rules to generate control instructions.

[0031] As a preferred scheme of the pump and fan operation environment control method of monitoring and feedback, wherein: through the device optimization control model, a time-varying performance optimization model is generated based on long-term operation data, including:

[0032] Through the device optimization control model, the long-term performance characteristics of the device operation state are generated;

[0033] Based on the long-term operation data, the performance decay correction term is generated by combining the environmental changes and the device aging characteristics;

[0034] Through the performance decay correction term and real-time control data, a time-varying performance optimization model is generated.

[0035] Another object of the present application is to provide a pump and fan operation environment control system of monitoring and feedback, which solves the problem that the device operation state monitoring and control in the prior art usually relies on manual inspection, static threshold judgment or single physical field analysis, and lacks real-time comprehensive analysis of multi-dimensional data such as vibration, noise and temperature. This leads to the fact that in complex industrial environments, the system is difficult to accurately capture the dynamic operation state of the device, and the timeliness and accuracy of abnormal detection are insufficient. In addition, the traditional control strategy usually uses fixed parameter setting, ignores the influence of environmental parameters (such as load, running speed, environmental temperature) and device aging on the running state, and the control effect is unstable under dynamic working conditions, which is difficult to adapt to long-term running requirements.

[0036] As a preferred scheme of the monitoring and feedback pump and fan operating environment control system, it is characterized in that, comprising:

[0037] The acquisition module is used for acquiring device operating parameters and constructing a device state monitoring model;

[0038] The first analysis module is used for collecting vibration, noise and temperature data in real time through the device state monitoring model, identifying the device operating state, and determining abnormal points;

[0039] The first control model construction module is used for dynamically generating vibration, noise and temperature control strategy models through abnormal point data combined with environmental parameters, for the first control of the device operating state;

[0040] The second control model construction module is used for combining the first control, analyzing the dynamic characteristics of vibration, noise and temperature, obtaining vibration-noise-temperature multi-dimensional control thresholds, and performing the second control of the device operating state;

[0041] The optimization model construction module is used for generating a device optimization control model based on a closed-loop feedback mechanism combined with the coupling characteristics of vibration-noise-temperature;

[0042] The performance optimization model construction module is used for generating a time-varying performance optimization model based on long-term operating data through the device optimization control model, for continuous optimization control of device vibration, noise and temperature.

[0043] A computer device comprising a memory and a processor, the memory storing a computer program, and the processor implementing the steps of the monitoring and feedback pump and fan operating environment control method when executing the computer program.

[0044] A computer readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps of the monitoring and feedback pump and fan operating environment control method.

[0045] The beneficial effects of the present application are: through the monitoring model, the abnormality is identified in real time and the control strategy is dynamically generated, the first control is quickly implemented, and the device state is stabilized. Then, through the quantification of the coupling effect of vibration, noise and temperature, a multi-dimensional control threshold and an adaptive control algorithm are generated, the control precision and the intelligent level are improved. Based on long-term data, a time-varying performance optimization model is generated, and the parameters are dynamically adjusted to cope with aging and working condition changes. The overall effect significantly improves the device operating efficiency, reliability and environmental adaptability, prolongs the service life, and reduces the maintenance cost. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0047] Figure 1 A general flowchart of a monitoring and feedback pump and fan operating environment control method is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to make the above-mentioned purposes, features and advantages of the present application more apparent and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings in the specification. Obviously, the described embodiments are only some embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present application.

[0049] Embodiment 1, refer to Figure 1 For the first embodiment of the present application, the embodiment provides a monitoring and feedback pump and fan operating environment control method, comprising:

[0050] S101: Obtain device operating parameters and construct a device state monitoring model;

[0051] In S101, the vibration, noise and temperature data of the device are collected in real time by the device state monitoring model, and the feature distribution graph of the device operating state is generated. The maximum value region of vibration, noise or temperature is determined, and the normal operation mode, slight abnormality or serious fault state of the device is determined.

[0052] The method completes the collection of device operating parameters and the construction of the state monitoring model through the collaborative work of the intelligent monitoring module and the central processing unit.

[0053] In a preferred embodiment of the present application, the intelligent monitoring module is composed of high-precision vibration sensors, sound level meters and temperature sensors, which are respectively installed at the key positions (such as motor base, pump body, fan blades, etc.) of the pump and fan, to collect vibration, noise and temperature data in real time.

[0054] The measurement accuracy of the vibration sensor is ±5%, and the frequency response range is 0.1 Hz to 1000 Hz; the measurement range of the sound level meter is 30 dB to 130 dB; the accuracy of the temperature sensor is ±1℃. These sensors collect data at a sampling frequency of at least 10 times per second and transmit them to the central processing unit through the wireless ZigBee protocol, ensuring the real-time and reliability of the data. The central processing unit uses fast Fourier transform (FFT) and wavelet transform algorithms to perform spectral analysis and time-frequency analysis on the collected data, extracting feature vectors of vibration, noise and temperature, including time domain features (such as mean, variance, peak value), frequency domain features (such as main frequency components of FFT analysis) and time-frequency domain features (such as mutation features of wavelet transform).

[0055] Based on these feature vectors, a device state monitoring model is constructed by machine learning algorithms such as support vector machine (SVM), and a feature distribution map of the device operating state is generated. The feature distribution map displays the distribution of vibration, noise and temperature in three dimensions, and the device state is determined by analyzing the maximum value region (such as vibration amplitude exceeding 3.0 mm / s, noise exceeding 80 dB or temperature exceeding 40℃). For example, if the vibration amplitude is between 3.0 mm / s and 3.5 mm / s, it is marked as "mild abnormality"; if it exceeds 3.5 mm / s, it is marked as "serious failure". In this way, the normal operation mode, slight abnormality or serious failure state of the device can be accurately identified.

[0056] In an optional embodiment of the present application, the feature distribution map generated in step S101 can be realized by a convolutional neural network (CNN). The vibration, noise and temperature data are converted into time series or frequency domain signals and input into the CNN model to automatically extract features and generate a distribution map. The maximum value region and the device state are output by the classification layer.

[0057] It should be noted that in the prior art, vibration and noise monitoring usually relies on discrete manual inspection or simple threshold judgment, which lacks real-time and comprehensiveness, and it is difficult to accurately capture the dynamic operating state and potential abnormalities of the device. In addition, traditional methods often cannot effectively extract features and perform comprehensive analysis when dealing with multi-dimensional data (such as vibration, noise, temperature) in complex industrial environments, resulting in insufficient accuracy and timeliness of fault diagnosis. The preferred solution of the present application is to deploy an intelligent monitoring module composed of sensors, combined with fast Fourier transform (FFT), wavelet transform and machine learning algorithms in the central processing unit, to construct a device state monitoring model, realize real-time collection of device operating parameters, feature extraction and state evaluation.

[0058] The method of the embodiment ensures comprehensive monitoring of the running state of the equipment through real-time collection and multi-dimensional analysis, and significantly improves the accuracy and timeliness of abnormality detection. The combination of FFT and wavelet transform can accurately extract periodic and transient characteristics, making up for the shortcomings of traditional methods in non-stationary signal processing; the introduction of a machine learning model realizes intelligent classification of the state, reducing the need for manual intervention. In addition, the generation of feature distribution graphs facilitates users to intuitively understand the state of the equipment, providing a reliable basis for the development of subsequent control strategies.

[0059] S102: The vibration, noise and temperature control strategy model is dynamically generated by combining abnormal point data with environmental parameters, including:

[0060] In S102, the characteristic frequency and transient characteristics of the abnormal point are extracted based on spectral analysis and time-frequency analysis; the vibration, noise and temperature control strategy model is generated by adjusting the damper coefficient of the vibration isolator, the configuration of the sound-absorbing material and the fan blade angle based on the characteristic frequency and transient characteristics.

[0061] In a preferred embodiment of the present application, in the time-frequency analysis stage, wavelet transform is used to process non-stationary signals, Morlet wavelet is selected as the basis function, and 5-layer wavelet decomposition is performed. Wavelet transform can capture transient characteristics in vibration and noise signals, such as signal mutations when the equipment starts or the load suddenly changes. By analyzing the wavelet coefficients, the transient characteristics of the abnormal point are extracted, such as a sudden increase in the amplitude of a certain layer of wavelet coefficients, indicating transient vibration or noise abnormalities. These transient characteristics are compared with the pre-set abnormal feature library to determine the nature of the abnormal point.

[0062] The control strategy model is dynamically generated in combination with environmental parameters, and the specific adjustments are as follows:

[0063] Vibration isolator damper coefficient adjustment: The damper coefficient of the electromagnetic vibration isolator is dynamically adjusted according to the characteristic frequency range and environmental parameters (such as load and running speed). For low-frequency vibration (<50Hz), which is usually related to low-speed operation (<1000rpm) or low load (load <50% of rated value), the damper coefficient is adjusted to 0.6-0.8 to enhance low-frequency vibration absorption; for medium-frequency vibration (50Hz-200Hz), which is common in medium-speed operation (1000rpm-3000rpm) or medium load (50%-80% of rated value), the damper coefficient is adjusted to 0.8-1.0 to balance the rigidity and damping effect; for high-frequency vibration (>200Hz), which is usually found in high-speed operation (>3000rpm) or high load (>80% of rated value), the damper coefficient is adjusted to 1.0-1.2 to reduce high-frequency vibration transmission. For example, when the load increases by 20% and high-frequency vibration abnormalities are detected, the damper coefficient is adjusted from 0.8 to 1.2 by increasing the current, while considering the running speed to avoid excessive damping leading to increased energy consumption.

[0064] Absorbing material configuration optimization: Adjust the thickness, density, and porosity of the absorbing material according to the characteristic frequency of the noise signal and environmental parameters (such as ambient temperature and equipment operating speed). For low-frequency noise (<500 Hz), which is usually caused by low-speed operation or aerodynamic effects, increase the thickness of the absorbing material to 15 cm when the ambient temperature is high (>30℃) to enhance low-frequency absorption; for medium-frequency noise (500 Hz-2000 Hz), which is common in medium-speed operation, adjust the porosity to 60% to optimize the sound absorption effect; for high-frequency noise (>2000 Hz), which is usually caused by high-speed operation, increase the surface micro-pore density to 200 per square centimeter. For example, when the ambient temperature is 35℃ and high-frequency noise anomalies are detected, the micro-pore structure is optimized by electric adjustment devices to improve the sound absorption effect by at least 5 dB.

[0065] Fan blade angle adjustment: Adjust the fan blade angle according to the characteristic frequency of the aerodynamic noise and environmental parameters (such as operating speed and load). For low-speed operation (<1000 rpm) and low load (<50% of rated value), adjust the blade angle to 10°-15° to reduce vortex noise; for medium-speed operation (1000 rpm-3000 rpm) and medium load (50%-80% of rated value), adjust the blade angle to 15°-20° to balance noise and ventilation efficiency; for high-speed operation (>3000 rpm) and high load (>80% of rated value), adjust the blade angle to 20°-25° to reduce airflow impact noise. For example, when the operating speed exceeds 3000 rpm and aerodynamic noise anomalies are detected, adjust the blade angle from 15° to 25° to reduce noise, while adjusting the angle according to the load to maintain ventilation efficiency.

[0066] Temperature control strategy: Combine temperature sensor data and environmental parameters (such as ambient temperature and load) to control temperature by increasing cooling power or adjusting operating parameters (such as reducing load) when the temperature exceeds 40℃ and is accompanied by vibration or noise anomalies. For example, if the ambient temperature is 38℃, the load increases by 20%, and the temperature anomaly occurs simultaneously with high-frequency vibration, prioritize adjusting the damper damping coefficient to 1.2 while increasing the cooling power by 20% to ensure that the temperature remains below 35℃.

[0067] In an optional embodiment of the present application, the vibration, noise, and temperature control strategy model generated in step S102 can also be realized by a machine learning method, inputting the characteristic frequency and transient characteristics of the anomaly point into a pre-trained neural network model, and outputting the optimized damper damping coefficient, absorbing material configuration, and fan blade angle through the model.

[0068] It should be noted that in the prior art, vibration and noise control usually adopts static parameter setting, ignoring the influence of environmental parameters (such as load, running speed, environmental temperature) on the running state of the equipment, resulting in unstable control effect under dynamic working conditions. The preferred solution of the present application is to extract the characteristic frequency and transient characteristics of the abnormal point through frequency spectrum analysis and time-frequency analysis, and dynamically adjust the control parameters combined with environmental parameters. This method can respond to changes in equipment running state and environmental conditions in real time, ensuring the accuracy and stability of vibration, noise and temperature control.

[0069] The method of this embodiment extracts the characteristic frequency and transient characteristics of the abnormal point through FFT and wavelet transform, and dynamically adjusts the damper coefficient of the vibration isolator, the configuration of the sound-absorbing material and the angle of the fan blade combined with environmental parameters (such as load, running speed, environmental temperature) to generate an adaptive control strategy model. In actual operation, it can reduce vibration to below 2.5mm / s, reduce noise to below 75dB, maintain temperature below 35℃, control effect deviation within ±5%, significantly improve the running efficiency and reliability of the pump and fan equipment, and provide a quiet and comfortable working environment for the operators.

[0070] S103: Dynamically generate vibration, noise and temperature control strategy model through abnormal point data combined with environmental parameters, for the first control of the equipment running state;

[0071] In a preferred embodiment of the present application, the first control refers to the initial control measure that is preferentially executed after detecting the abnormality of the equipment running state, aiming to quickly correct the abnormality, prevent it from further deterioration, and restore the equipment running state to the normal or acceptable range. The specific meaning is as follows:

[0072] The first control is the primary response to the abnormal state, which is usually triggered immediately after the abnormal point is identified. For example, when the vibration amplitude exceeds 3.5mm / s, the damper coefficient of the vibration isolator will be adjusted to 1.0-1.2 immediately to reduce vibration transmission. The purpose of this rapid response is to prevent the abnormality from causing equipment damage or failure expansion.

[0073] The goal of the first control is to stabilize the equipment running state and ensure that vibration, noise and temperature are maintained within a safe range (for example, vibration <2.5mm / s, noise <75dB, temperature <35℃). It focuses on solving the current abnormal problem and laying the foundation for subsequent optimization control (such as long-term adaptive adjustment).

[0074] In an optional embodiment of the present application, the dynamically generated vibration, noise and temperature control strategy model can be realized through multi-physical field coupling simulation. Specifically, the abnormal point data and environmental parameters are input into a multi-physical field simulation software to construct a coupling model containing structural mechanics, acoustics and thermodynamics, and the dynamic responses of vibration, noise and temperature are solved by the finite element method or the finite volume method. Based on the simulation results, a closed-loop control algorithm is used to generate real-time control strategies. Compared with the original scheme which may be based on a single physical field or an empirical formula, this method improves the model accuracy through multi-physical field coupling to achieve the same control effect.

[0075] It should be noted that in the prior art, vibration, noise and temperature control usually relies on static parameter setting or simple threshold judgment, lacking comprehensive analysis of abnormal point data and environmental parameters, resulting in unstable control effect under dynamic working conditions. For example, the traditional method may only trigger a unified response when the vibration exceeds a certain fixed threshold, ignoring the changes in load or running speed. The preferred solution of the present application is to extract the characteristic frequency and transient characteristics of the abnormal point through FFT and wavelet transform, and dynamically generate the first control for the device running state in combination with environmental parameters. This method can respond in real time to changes in device running state and environmental conditions, ensuring the accuracy and stability of the control.

[0076] The method of the present embodiment generates an adaptive vibration, noise and temperature control strategy model for the first control of the device running state by analyzing abnormal point data in real time and combining environmental parameters. It can quickly respond when an abnormality occurs, for example, reducing vibration to below 2.5 mm / s, reducing noise to below 75 dB, maintaining temperature below 35°C, and controlling the control effect deviation within ±5%. This first control not only significantly improves the efficiency and reliability of the device, but also provides a quiet and comfortable working environment for the operator, while providing a reliable foundation for subsequent optimization control.

[0077] S104: In combination with the first control, analyze the dynamic characteristics of vibration, noise and temperature to obtain a vibration-noise-temperature multi-dimensional control threshold for the second control of the device running state;

[0078] In S104, a vibration, noise and temperature feature matrix of the abnormal point is generated through a pattern recognition algorithm; the coupling effect of vibration, noise and temperature is quantified based on historical data and real-time data to generate a vibration correction term; the influence of noise on vibration control is quantified through environmental parameters and device running state to generate a noise correction term; the influence of temperature on vibration and noise control is quantified based on temperature data and device running state to generate a temperature correction term; based on the vibration correction term, noise correction term and temperature correction term, a vibration-noise-temperature multi-dimensional control threshold is generated in combination with the reference control parameter; and the second control of the device running state is performed based on the comparison between the multi-dimensional control threshold and the real-time monitoring data.

[0079] In a preferred embodiment of the present application, a pattern recognition algorithm is used to generate a vibration, noise, and temperature feature matrix for abnormal points. The central processing unit receives real-time data from the intelligent monitoring module, including data collected by vibration sensors (measurement accuracy ±5%, frequency response range 0.1 Hz to 1000 Hz), sound level meters (measurement range 30 dB to 130 dB), and temperature sensors (accuracy ±1°C). Through Fast Fourier Transform (FFT, number of sampling points 1024, frequency resolution 1 Hz) and wavelet transform (Morlet wavelet selected, 5 layers of decomposition), spectral analysis and time-frequency analysis are performed on the data, and time domain features (such as mean, variance, peak value), frequency domain features (such as main frequency components), and time-frequency domain features (such as transient mutation characteristics) are extracted. These features are integrated into a feature vector, which is input into a support vector machine (SVM) or a deep convolutional neural network (CNN) for pattern recognition to generate a feature matrix for abnormal points. For example, when the vibration amplitude exceeds 3.5 mm / s, the noise exceeds 85 dB, or the temperature exceeds 40°C, the abnormal point is marked and the corresponding feature matrix is generated, which includes dimensions such as vibration amplitude, frequency, noise intensity, frequency distribution, and temperature value.

[0080] Based on historical data and real-time data, the coupling effect of vibration, noise, and temperature is quantified, and a vibration correction term is generated. Historical data is stored in a local database, including at least 6 months of operation data (1000 data points collected every 24 hours, covering vibration, noise, temperature), through time series analysis and correlation analysis, the mutual influence between vibration, noise and temperature is identified. For example, analysis shows that when the temperature exceeds 40°C, the vibration amplitude may increase by 10%, and the noise level may increase by 3dB. A machine learning algorithm (such as long short-term memory network, LSTM) is used to model the coupling effect and generate a vibration correction term, which is a weighted coefficient (such as 0.1 to 0.3) used to adjust the vibration control parameters. For example, when the temperature rises and causes the vibration amplitude to increase, the vibration correction term will increase the damper coefficient of the vibration isolator by 0.2 to offset the impact of the coupling effect.

[0081] Through environmental parameters (such as load, running speed, environmental temperature) and equipment operating state, the influence of noise on vibration control is quantified, and a noise correction term is generated. The noise correction term is generated based on noise spectrum analysis (main frequency components extracted by FFT) and regression analysis of environmental parameters. For example, when the noise exceeds 85 dB and the running speed exceeds 3000 rpm, the interference of noise on vibration control is detected, causing the vibration amplitude to increase by 5%. The noise correction term is represented as an adjustment factor (such as 0.05 to 0.15) used to optimize the damper coefficient of the vibration isolator or the configuration of the sound-absorbing material. For example, the damper coefficient may be adjusted from 1.0 to 1.15, and the thickness of the sound-absorbing material may be increased to 15 cm to reduce the vibration amplification effect caused by noise.

[0082] Based on temperature data and equipment operating status, quantify the impact of temperature on vibration and noise control, generate temperature correction term. Temperature correction term is generated by analyzing the correlation between temperature and vibration, noise. For example, when the temperature exceeds 40℃ and the load increases by 20%, the vibration amplitude may increase by 8%, and the noise level may rise by 2dB. Using fuzzy logic control algorithm, combined with historical data and real-time data, generate temperature correction term, expressed as a control parameter adjustment value (such as 0.1 to 0.2). For example, it may increase the cooling power by 20%, and at the same time, increase the damper coefficient of the vibration isolator by 0.1, to alleviate the negative impact of high temperature on vibration and noise control.

[0083] Based on vibration correction term, noise correction term and temperature correction term, combined with the reference control parameters (such as vibration threshold 3.0mm / s, noise threshold 80dB, temperature threshold 40℃), generate vibration-noise-temperature multi-dimensional control threshold. Specifically, through the weighted fusion algorithm (the weight is determined according to historical data analysis, for example, vibration: 0.4, noise: 0.3, temperature: 0.3), integrate the correction term to generate the multi-dimensional control threshold.

[0084] Based on the comparison of multi-dimensional control threshold and real-time monitoring data, the second control of equipment operating status is carried out. If the real-time data exceeds the multi-dimensional control threshold, the second control is triggered, including adjusting the damper coefficient of the vibration isolator (such as from 1.0 to 1.2), optimizing the configuration of sound-absorbing material (such as increasing the micro-pore density to 200 per square centimeter), adjusting the fan blade angle (such as from 15° to 20°), and increasing the cooling power (such as increasing by 20%).

[0085] In an optional embodiment of the present application, a deep learning model (such as a convolutional neural network CNN) can also be used to extract features from time series data of vibration, noise and temperature, generating a high-dimensional feature matrix to replace traditional pattern recognition algorithms. Through the long short-term memory network (LSTM) analysis of the time series dependence of historical and real-time data, the dynamic coupling effect of vibration, noise and temperature is quantified, and the vibration correction term is generated. Using adaptive filtering technology (such as Kalman filtering) combined with environmental parameters and equipment operating status, the impact of noise and temperature on vibration control is quantified respectively, and the noise correction term and the temperature correction term are generated. Based on the support vector machine (SVM) regression model, combined with the vibration correction term, the noise correction term, the temperature correction term and the reference control parameters, the vibration-noise-temperature multi-dimensional control threshold is generated. Through the dynamic comparison of real-time monitoring data and multi-dimensional control threshold, the reinforcement learning algorithm (such as Q-learning) is used to optimize the second control of equipment operating status.

[0086] It should be noted that in the prior art, vibration, noise and temperature control is mainly dependent on a single physical field or static threshold, which is difficult to cope with multi-dimensional data coupling and dynamic working conditions, resulting in unstable control effect. For example, the traditional method may only adjust the damping coefficient when the vibration exceeds the standard, ignoring the influence of noise and temperature. The preferred solution of the present application is to generate a feature matrix through pattern recognition, quantify the coupling effect of vibration, noise and temperature, generate dynamic multi-dimensional control threshold, and implement the second control. This method can respond to the device running state and environmental changes in real time, ensuring the accuracy and stability of the control.

[0087] The method of the present embodiment generates multi-dimensional control threshold and implements the second control by analyzing the dynamic characteristics of vibration, noise and temperature, significantly improving the control accuracy and adaptability compared with the first control. This not only improves the efficiency and reliability of the device, but also lays the foundation for subsequent optimization control.

[0088] S105: Based on the closed-loop feedback mechanism, a device optimization control model is generated in combination with the coupling characteristics of vibration-noise-temperature;

[0089] In S105, a feedback matrix of vibration, noise and temperature is generated based on the actual effect of the first control and the second control; the deviation characteristics of the control effect are extracted in combination with the coupling characteristics of vibration, noise and temperature; an adaptive control algorithm is generated based on the deviation characteristics and historical control data; the adaptive control algorithm includes a combination of genetic algorithm, neural network and fuzzy logic control, wherein: the control parameters are optimized based on a dynamic fitness function through the genetic algorithm; the local features and time series characteristics of the vibration, noise and temperature signals are extracted through the neural network; the control instructions are generated through multi-dimensional reasoning based on automatically generated fuzzy rules through the fuzzy logic control.

[0090] In a preferred embodiment of the present application, a feedback matrix of vibration, noise and temperature is generated based on the actual effect of the first control and the second control. The central processing unit receives real-time data from the intelligent monitoring module, including data collected by the vibration sensor, sound level meter and temperature sensor. The first control (refer to S103) quickly responds to abnormalities based on a single threshold, and the second control (refer to S104) makes fine adjustments through multi-dimensional control threshold.

[0091] Through the closed-loop feedback mechanism, the actual control effect (such as the real-time values of vibration, noise and temperature) is compared with the expected target (vibration < 2.5 mm / s, noise < 75 dB, temperature < 35℃), and a feedback matrix is generated. The feedback matrix includes the vibration amplitude, noise level, temperature value and deviation value after control. For example, if the actual vibration is 2.7 mm / s and the deviation is 0.2 mm / s, the matrix records the deviation and the corresponding running state (such as load, running speed).

[0092] Secondly, the deviation features of control effect are extracted by combining the coupling characteristics of vibration, noise and temperature. Based on historical data (stored in local database, containing 6 months of operation data, 1000 data points collected every 24 hours) and real-time data, the coupling effect is quantified through time series analysis and correlation analysis. For example, the analysis shows that when the temperature exceeds 40°C, the vibration amplitude increases by 10% and the noise increases by 3dB. The transient features of non-stationary signals are extracted using wavelet transform (Morlet wavelet is selected, 5 layers of decomposition), combined with fast Fourier transform (FFT, 1024 sampling points, frequency resolution 1Hz) to extract frequency domain features, and a deviation feature vector is generated. The deviation features include time domain deviation (such as vibration mean deviation 0.2mm / s), frequency domain deviation (such as main frequency component deviation 5Hz) and coupling deviation (such as vibration amplitude increment caused by temperature rise). For example, when the temperature rises from 38°C to 42°C, the vibration deviation increases from 0.1mm / s to 0.3mm / s, and the coupling deviation feature is recorded.

[0093] Subsequently, an adaptive control algorithm is generated based on the deviation features and historical control data. The algorithm combines genetic algorithm, neural network and fuzzy logic control, and the specific implementation is as follows:

[0094] Genetic algorithm optimizes control parameters: dynamic fitness function is introduced, evaluation criteria are adjusted according to real-time monitoring data (such as vibration, noise, temperature). For example, when the load increases by 20%, the fitness function increases the vibration control weight (from 0.3 to 0.5) and reduces the energy consumption weight (from 0.4 to 0.2). Adaptive crossover and mutation operations are designed, with initial crossover rate 0.9 and mutation rate 0.2, and search later stage respectively reduced to 0.6 and 0.05. For example, when optimizing the damper coefficient of the vibration isolator, the crossover rate is adjusted to 0.7 according to the vibration deviation of 0.2mm / s, and the optimal damper coefficient (such as 1.15) is generated.

[0095] Neural network extracts features: deep convolutional neural network (CNN) is used to process vibration signals, extract local features (such as abnormal amplitude of high frequency vibration), and long short-term memory network (LSTM) is used to process time series characteristics of noise signals (such as periodic fluctuations). Incremental training is implemented, new data (about 1000 data points) is collected every 12 hours, and the neural network model is updated. For example, CNN identifies abnormal 200Hz frequency component in vibration signal, LSTM detects periodic increase of noise level by 3dB within 10 minutes, and outputs feature vector for control parameter adjustment.

[0096] Fuzzy logic control multi-dimensional reasoning: automatically generate fuzzy rules by genetic algorithm, optimize membership function parameters. For example, based on vibration deviation 0.2mm / s and temperature 42℃, adjust the membership function, generate the rule "if vibration deviation is high and temperature is high, then damping coefficient increases significantly". Fuzzy logic controller considers vibration amplitude, frequency, noise intensity and temperature to generate control instructions. For example, when vibration deviation is 0.3mm / s and temperature exceeds 40℃, the controller outputs instructions to adjust the damping coefficient from 1.0 to 1.2, and the sound-absorbing material thickness increases to 15cm.

[0097] In an optional embodiment of the present application, a multi-dimensional feedback matrix can also be constructed by collecting real-time sensor data of vibration, noise and temperature based on the actual effect of the first control and the second control. Principal component analysis (PCA) is used to reduce the dimensionality of the data, retaining the main features to reduce computational complexity. Combined with the physical coupling model of vibration, noise and temperature, the data is smoothed by Kalman filter to generate a high-precision feedback matrix. Support vector machine (SVM) is used to classify the data in the feedback matrix to identify the deviation characteristics of the control effect. SVM maps high-dimensional features through kernel functions (such as radial basis functions) to extract the nonlinear coupling characteristics of vibration, noise and temperature. Combined with time series analysis (such as ARIMA model), the time dynamic characteristics of the deviation are extracted. Reinforcement learning (RL) framework (such as deep Q network, DQN) is used to generate adaptive control algorithm to replace the combination of genetic algorithm, neural network and fuzzy logic.

[0098] By defining a reward function (based on deviation characteristics and control objectives), the control parameters are optimized. DQN uses deep neural networks to approximate the state-action value function to dynamically adjust the control strategy. Convolutional neural network (CNN) is used to extract local features from vibration, noise and temperature signals, and long short-term memory network (LSTM) is used to capture time series characteristics. Based on Markov decision process (MDP), multi-dimensional reasoning is performed to generate control instructions, achieving similar multivariate control effect as fuzzy logic.

[0099] It should be noted that in the prior art, vibration, noise and temperature control are mostly dependent on static parameters or single feedback mechanism, making it difficult to effectively handle multi-dimensional coupling effects and dynamic working conditions, resulting in unstable control effect. For example, traditional methods may only adjust the damping coefficient based on vibration deviation, ignoring the coupling effects of noise and temperature. The preferred solution of the present application is to generate a feedback matrix and deviation characteristics based on a closed-loop feedback mechanism, combined with the coupling characteristics of vibration, noise and temperature, and to generate an adaptive control algorithm through the combination of genetic algorithm, neural network and fuzzy logic control. This method can respond to the device operating state and environmental changes in real time, ensuring the accuracy and stability of the control.

[0100] The method of this embodiment significantly improves the control accuracy and adaptability of pump and fan equipment by generating an equipment optimization control model. Based on the feedback data of the first control and the second control, an adaptive control algorithm is generated in combination with the coupling characteristics, which can quickly respond to abnormalities, thereby improving the equipment operation efficiency and reliability and reducing the failure rate.

[0101] S106: Through the equipment optimization control model, a time-varying performance optimization model is generated based on long-term operation data, and vibration, noise and temperature of the equipment are continuously optimized and controlled.

[0102] In S106, the long-term performance characteristics of the equipment operation state are generated through the equipment optimization control model; based on the long-term operation data, the performance decay correction term is generated in combination with the environmental changes and the equipment aging characteristics; and the performance decay correction term and the real-time control data are generated.

[0103] In a preferred embodiment of the present application, the long-term performance characteristics of the equipment operation state are generated through the equipment optimization control model. The equipment optimization control model is based on the adaptive control algorithm of vibration, noise and temperature generated in step S105 (combined with genetic algorithm, neural network and fuzzy logic control), uses a central processing unit to process real-time data from an intelligent monitoring module, including data collected by vibration sensors, sound level meters and temperature sensors. At least 6 months of operation data (1000 data points collected every 24 hours, covering vibration, noise and temperature) are extracted from the local database, and long-term performance characteristics are extracted through time series analysis. These characteristics include monthly change rate of vibration amplitude (e.g. 5% increase per month), periodic fluctuation of noise level (e.g. 2dB increase every 3 months) and trend change of temperature (e.g. 1℃ increase every quarter). For example, analysis shows that after 6 months of operation, the vibration amplitude increases from 2.5mm / s to 2.8mm / s, indicating that there may be equipment aging or working condition changes.

[0104] Based on the long-term operation data, the performance decay correction term is generated in combination with the environmental changes and the equipment aging characteristics. Environmental changes are detected by real-time monitoring of parameters such as load, operating speed and ambient temperature, for example, a load increase of more than 20% within 10 minutes is marked as "working condition change". Equipment aging is detected by trend analysis, for example, a monthly change rate of vibration amplitude of more than 10% is marked as "possible aging". In combination with historical data and user feedback, the aging feature library and the working condition change feature library are updated to generate the performance decay correction term. The performance decay correction term is represented as an adjustment factor for the control parameters, for example:

[0105] Vibration decay correction term: if the vibration amplitude increases by 10% due to aging, the correction term is to increase the damping coefficient by 0.2 (e.g. from 1.0 to 1.2).

[0106] Noise attenuation correction term: If the noise level increases by 3dB due to operating condition changes, the correction term is to increase the thickness of the sound-absorbing material by 5cm (e.g. from 10cm to 15cm).

[0107] Temperature attenuation correction term: If the temperature increases by 2℃ due to aging, the correction term is to increase the cooling power by 15%.

[0108] Based on the performance attenuation correction term and real-time control data, a time-varying performance optimization model is generated. The performance optimization model uses an incremental learning mechanism, and the machine learning model (such as online support vector machine or incremental neural network) is updated every 7 days. The update process includes:

[0109] Incremental data preparation: Extract the data of the last 7 days (about 1000 data points / day) from the database, including vibration, noise, temperature and environmental parameters.

[0110] Feature extraction: Extract frequency domain features through fast Fourier transform (FFT, 1024 sampling points, frequency resolution 1Hz), extract time-frequency features through wavelet transform (Morlet wavelet, 5 layers of decomposition), and generate feature vectors combined with time domain features (such as mean, variance).

[0111] Model update: Input the performance attenuation correction term (such as damping coefficient adjustment 0.2) and real-time control data (such as current vibration 2.9mm / s) into the model to optimize the control parameters. For example, the model adjusts the damping coefficient to 1.2 according to the vibration attenuation correction term, and the micro-pore density of the sound-absorbing material increases to 200 per square centimeter.

[0112] Performance evaluation: Use the test set (30% of historical data) to evaluate the performance of the model, focusing on accuracy, recall rate and F1 score. If the performance decreases by more than 10%, trigger an alarm and prompt the user to intervene.

[0113] The performance optimization model continuously monitors the control effect through a closed-loop feedback mechanism, compares the actual output (such as vibration, noise, temperature) with the expected target (vibration <2.5mm / s, noise <75dB, temperature <35℃), and automatically adjusts the control parameters.

[0114] In an optional embodiment of the present application, in step S106, when generating the long-term performance characteristics of the device running state, a digital twin technology can also be used instead of the device optimization control model. By constructing a digital twin model of the device, the running state of the device is simulated in real time and the long-term performance characteristics are predicted; based on the long-term running data, combined with environmental changes and device aging characteristics, a performance attenuation correction term is generated using a statistical probability model; through the performance attenuation correction term and real-time control data, the optimization control of the device running state is realized.

[0115] It should be noted that in the prior art, the control of vibration, noise and temperature is usually based on static parameters or short-term feedback, which is difficult to effectively cope with performance degradation caused by equipment aging and environmental changes. For example, the traditional method may only adjust the control parameters by a fixed threshold (such as vibration 3.5 mm / s), ignoring the aging effect and working condition changes in long-term operation. The preferred solution of the present application is to generate a time-varying performance optimization model based on the device optimization control model, and to dynamically adjust the control parameters by using the performance degradation correction term to adapt to the device aging and environmental changes. This method can continuously optimize the control effect and ensure the stability and reliability of the device in long-term operation.

[0116] The method of the present embodiment significantly improves the long-term operation efficiency and reliability of pump and fan equipment by generating a time-varying performance optimization model. Based on long-term performance characteristics and performance degradation correction terms, the control parameters are dynamically adjusted to quickly respond to aging or working condition changes, prolonging the service life of the equipment and reducing maintenance costs.

[0117] Through the user interaction interface, the device state, monitoring data and control effect are displayed in real time; based on user feedback, the control strategy and parameters are adjusted to generate a user optimized control model.

[0118] In a preferred embodiment of the present application, the system displays the device state, monitoring data and control effect in real time through the user interaction interface. The user interaction interface is designed to be multi-platform compatible, supporting touch screens, computers and mobile devices, ensuring that users can monitor and operate at any time and in any place. The interface displays the following information intuitively:

[0119] Device state: including the running mode of the device (normal operation, slight abnormality, serious failure), prompting the user through color coding (such as green for normal, yellow for slight abnormality, and red for serious failure). For example, when the vibration amplitude exceeds 3.0 mm / s but is lower than 3.5 mm / s, the interface displays "slight abnormality" and marks the specific value (such as 3.2 mm / s).

[0120] Monitoring data: real-time display of vibration, noise and temperature values and trend charts. The trend chart is based on time series analysis, showing the data changes in the last 24 hours (such as vibration amplitude rising from 2.5 mm / s to 2.8 mm / s).

[0121] Control effect: display of current control parameters (such as damper coefficient 1.0, sound-absorbing material thickness 10 cm, fan blade angle 15°) and their effects (such as vibration reduced to below 2.5 mm / s, noise reduced to below 75 dB). The interface also displays control deviation (such as vibration deviation within ±5%) and alarm information (such as "vibration exceeds threshold 3.5 mm / s").

[0122] The interface supports interactive functions, allowing users to view detailed data reports (such as spectrum analysis graphs, correlation analysis results) or trigger fault diagnosis. For example, the user clicks the "Fault Diagnosis" button, and the interface displays possible causes (such as "high-frequency vibration may be caused by bearing wear") and maintenance recommendations (such as "check the lubrication status of the bearing").

[0123] Secondly, the system adjusts the control strategy and parameters based on user feedback, generating a user-optimized control model. Users can manually input feedback through the interactive interface, including:

[0124] Running state feedback: Users record the actual operating conditions of the equipment (such as "the equipment has high vibration under high load") or maintenance records (such as "vibration decreased after replacing the bearing").

[0125] Control parameter adjustment suggestions: Users can manually modify control thresholds or parameters, such as temporarily adjusting the vibration threshold from 3.0 mm / s to 4.0 mm / s under special working conditions, or adjusting the fan blade angle from 15° to 20° to adapt to specific heat dissipation requirements.

[0126] The system processes user feedback through the central processing unit, combines historical data (stored in the local database, containing 6 months of operating data, 1000 data points collected every 24 hours) and real-time monitoring data, and generates a user-optimized control model. The specific implementation is as follows:

[0127] Feedback data processing: The system pre-processes the user input feedback, including denoising and normalization, and extracts key features (such as the user-reported "high-load vibration" corresponding to a vibration amplitude of 3.3 mm / s). These features are compared with real-time data (such as the 3.2 mm / s measured by the vibration sensor) to verify the accuracy of the feedback.

[0128] Control strategy adjustment: The system updates the control strategy based on user feedback. For example, if the user feedback is "high vibration under high load", the system analyzes the historical data to confirm that the vibration increases by 10% when the load increases by 20%, and then adjusts the damper coefficient (such as from 1.0 to 1.2) or the thickness of the sound-absorbing material (such as from 10 cm to 15 cm). The adjusted strategy is optimized by the fuzzy logic controller to generate new fuzzy rules (such as "if the load is high and the vibration is high, then the damping coefficient is significantly increased").

[0129] User optimization control model generation: The system uses an incremental learning mechanism to input user feedback and real-time data into a machine learning model (such as an online support vector machine) to update control parameters. The model is updated every 7 days, extracting time domain, frequency domain, and time-frequency domain features (such as vibration mean deviation 0.2 mm / s, main frequency shift 5 Hz) based on the last 7 days of data (about 1000 data points / day). The updated model generates a user optimization control model, giving priority to special user feedback requirements. For example, if the user requires noise reduction under certain working conditions, the model will increase the noise control weight from 0.3 to 0.5 and optimize the micro-pore density of the sound-absorbing material (from 100 to 200 per square centimeter).

[0130] Example 2, as a second embodiment of the present application, provides a monitoring and feedback pump and fan operating environment control method. In order to verify the beneficial effects of the present application, scientific demonstration is carried out through experiments.

[0131] In the prior art, equipment operating state monitoring and control usually rely on manual inspection, static threshold judgment or single physical field analysis, lacking real-time comprehensive analysis of multi-dimensional data such as vibration, noise and temperature. This leads to difficulties in accurately capturing the dynamic operating state of the equipment in complex industrial environments, and the timeliness and accuracy of abnormality detection are insufficient. In addition, traditional control strategies often use fixed parameter settings, ignoring the influence of environmental parameters (such as load, operating speed, environmental temperature) and equipment aging on operating state, and the control effect is unstable under dynamic working conditions, making it difficult to adapt to long-term operation requirements.

[0132] The present application realizes real-time monitoring and dynamic optimization control of vibration, noise and temperature by the collaborative work of intelligent monitoring module and central processing unit, combined with frequency spectrum analysis, time-frequency analysis and machine learning algorithm, to construct equipment state monitoring model and adaptive control strategy. The system not only can quickly respond to abnormalities, but also can continuously optimize equipment performance according to long-term operation data and user feedback, ensuring operating efficiency and reliability.

[0133] The present scheme adopts fast Fourier transform (FFT) and wavelet transform to perform frequency spectrum analysis and time-frequency analysis on vibration, noise and temperature data, and extract multi-dimensional feature vectors. Specifically, FFT extracts frequency domain features such as the main frequency component of the vibration signal with 1024-point sampling and 1Hz frequency resolution; wavelet transform selects Morlet wavelet for 5-layer decomposition to extract transient features of non-stationary signals such as vibration or noise mutation when the device starts or the load suddenly changes. These features include time domain features (such as mean, variance, peak value), frequency domain features (such as main frequency component) and time-frequency domain features (such as mutation feature), providing a comprehensive data basis for device state monitoring and control strategy generation. For example, in S101, the system generates a three-dimensional feature distribution map based on the feature vectors by support vector machine (SVM), accurately distinguishing between normal operation mode (vibration < 3.0 mm / s), slight abnormality (vibration 3.0 mm / s to 3.5 mm / s) and serious failure (vibration > 3.5 mm / s).

[0134] In S102 to S104, the present application dynamically generates vibration, noise and temperature control strategy models through abnormal point feature frequency and transient feature, combined with environmental parameters such as load, running speed and environmental temperature. The closed-loop feedback mechanism and adaptive control algorithm (combined with genetic algorithm, neural network and fuzzy logic control) generate the device optimization control model in S105. In S106, the system generates a time-varying performance optimization model based on long-term running data, dynamically adjusts the control parameters through performance decay correction terms (such as increasing the damping coefficient by 0.2 when the vibration amplitude increases by 10%), and adapts to device aging and environmental changes.

[0135] The present scheme significantly improves the accuracy and timeliness of anomaly detection and control by real-time acquisition of vibration, noise and temperature data, combined with FFT, wavelet transform and machine learning algorithms, to construct multi-dimensional feature distribution maps and adaptive control strategy models. The system displays the device state, monitoring data and control effect in real time through the user interaction interface, and generates a user optimization control model based on user feedback, significantly improving the running efficiency and reliability of pump and fan devices.

[0136] Embodiment 3 is the third embodiment of the present application, which is different from the first two embodiments:

[0137] If the functions are implemented in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the part of the technical solutions that essentially contribute to the prior art or the part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0138] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered a list of executable instructions for implementing logic functions, and can be specifically embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions, or in conjunction with these instructions execution systems, apparatuses, or devices. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport programs for use by an instruction execution system, apparatus, or device, or in conjunction with these instruction execution systems, apparatuses, or devices.

[0139] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection having one or more wires (electrical devices), a portable computer diskette (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium can even be paper or other suitable medium on which the program can be printed, as the program can be electronically obtained, for example, by optical scanning of the paper or other medium, followed by editing, interpreting, or otherwise processing, if necessary, in other suitable ways to be electronically obtained, and then stored in the computer memory.

[0140] It should be understood that portions of the application can be implemented in hardware, software, firmware, or combinations thereof. In the embodiments described above, the various steps or methods can be implemented, in part, or in whole, by software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, and in another embodiment, any of the following techniques, which are well known in the art, can be used to implement the application: a hybrid of the techniques mentioned above; a combination of one or more of the techniques mentioned above; a combination of one or more of the techniques mentioned above with one or more other techniques not mentioned above; and / or any other techniques for implementing the application.

[0141] Embodiment 4, a fourth embodiment of the present application, which provides a system for monitoring and feedback pump and fan operating environment control, comprising:

[0142] An acquisition module for acquiring device operating parameters and constructing a device state monitoring model;

[0143] A first analysis module for collecting vibration, noise and temperature data in real time through the device state monitoring model, identifying the device operating state, and determining abnormal points;

[0144] A first control model construction module for dynamically generating vibration, noise and temperature control strategy models by combining abnormal point data with environmental parameters, for first control of the device operating state;

[0145] A second control model construction module for combining the first control, analyzing the dynamic characteristics of vibration, noise and temperature, obtaining vibration-noise-temperature multi-dimensional control thresholds, and performing second control of the device operating state;

[0146] An optimization model construction module for generating a device optimization control model based on a closed-loop feedback mechanism and the coupling characteristics of vibration-noise-temperature;

[0147] A performance optimization model construction module for generating a time-varying performance optimization model based on long-term operating data through the device optimization control model, for continuous optimization control of device vibration, noise and temperature.

[0148] It should be noted that the present application, through the cooperative work of the intelligent monitoring module and the central processing unit, combines the fast Fourier transform (FFT), wavelet transform and machine learning algorithm to construct a device state monitoring model, realizes real-time acquisition, feature extraction and state evaluation of pump and fan device operating parameters. Compared with the prior art, the present application significantly improves the accuracy and timeliness of abnormal detection, and can effectively cope with the multi-dimensional data processing demand in complex industrial environment.

[0149] Further, in step S101, the intelligent monitoring module uses a high-precision vibration sensor (measurement accuracy ±5%, frequency response range 0.1 Hz to 1000 Hz), a sound level meter (measurement range 30 dB to 130 dB), and a temperature sensor (accuracy ±1°C) to collect data at a sampling frequency of at least 10 times per second and transmit them to the central processing unit through the ZigBee protocol. The central processing unit generates a feature distribution map of the device operating state using a support vector machine (SVM) or a convolutional neural network (CNN), clearly showing the maximum value area of vibration, noise, and temperature. For example, when the vibration amplitude exceeds 3.5 mm / s, the system is marked as "serious failure", providing an intuitive basis for subsequent control.

[0150] Further, in steps S102 to S104, the system extracts the characteristic frequency and transient characteristics of abnormal points through spectral analysis and time-frequency analysis, and dynamically generates vibration, noise, and temperature control strategy models in combination with environmental parameters (such as load, running speed, and ambient temperature). Specifically, the system adjusts the damper coefficient of the vibration isolator (for example, from 0.8 to 1.2 to cope with high-frequency vibration), optimizes the configuration of sound-absorbing materials (for example, increases the thickness to 15 cm to absorb low-frequency noise), and adjusts the fan blade angle (for example, from 15° to 25° to reduce aerodynamic noise). These dynamic adjustments significantly improve the adaptability and stability of the control strategy.

[0151] It should also be noted that in steps S105 and S106, the system further improves control accuracy and reliability of long-term operation of the device through a closed-loop feedback mechanism and a time-varying performance optimization model. Based on the combination of genetic algorithm, neural network, and fuzzy logic control, the system generates an adaptive control algorithm that can quickly respond to the coupling effects of vibration, noise, and temperature. For example, when the temperature exceeds 40°C, causing the vibration amplitude to increase by 10%, the system automatically adjusts the damping coefficient and cooling power to ensure that the vibration control is below 2.5 mm / s. In addition, the introduction of performance decay correction terms (such as increasing the damping coefficient by 0.2 when the vibration amplitude increases by 10% due to aging) enables the system to adapt to device aging and environmental changes, extending the device's life and reducing maintenance costs.

[0152] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and not to limit it. Although the present application 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 application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and they should be covered in the scope of the claims of the present application.

Claims

1. A monitoring and feedback method for controlling the operating environment of a pump and fan, characterized by: include, Obtain equipment operating parameters and build equipment status monitoring models; The equipment status monitoring model collects vibration, noise and temperature data in real time to identify the equipment operating status and determine abnormal points; Dynamically generate vibration, noise and temperature control strategy models by combining abnormal point data with environmental parameters for primary control of equipment operating status; Combined with the first control, the dynamic characteristics of vibration, noise and temperature are analyzed to obtain the vibration-noise-temperature multi-dimensional control threshold, and the second control of the equipment operation status is carried out; Based on the closed-loop feedback mechanism, the equipment optimization control model is generated by combining the coupling characteristics of vibration, noise and temperature; The equipment optimization control model is used in combination with long-term operation data to generate a time-varying performance optimization model to continuously optimize and control equipment vibration, noise and temperature.

2. A monitoring and feedback method for controlling the operating environment of a pump and fan according to claim 1, characterized in that: The equipment status monitoring model collects vibration, noise, and temperature data in real time to identify equipment operating status and locate abnormal points, including: The equipment status monitoring model collects equipment vibration, noise, and temperature data in real time to generate a characteristic distribution diagram of the equipment's operating status; Identify areas of maximum vibration, noise, or temperature to determine normal operating mode, minor anomalies, or severe fault conditions of the equipment.

3. A monitoring and feedback method for controlling the operating environment of a pump and fan according to claim 2, characterized in that: The method of dynamically generating vibration, noise and temperature control strategy models by combining abnormal point data with environmental parameters includes: Based on spectrum analysis and time-frequency analysis, the characteristic frequency and transient characteristics of abnormal points are extracted; The damping coefficient of the vibration isolator, the configuration of the sound absorbing material and the angle of the fan blade are adjusted according to the characteristic frequency and transient characteristics.

4. A monitoring and feedback method for controlling the operating environment of a pump and fan according to claim 3, characterized in that: Combined with the first control, the dynamic characteristics of vibration, noise, and temperature are analyzed to obtain the vibration-noise-temperature multi-dimensional control threshold, and the second control of the equipment operating status is carried out, including: Generate the vibration, noise and temperature feature matrices of abnormal points through pattern recognition algorithms; Quantify the coupling effects of vibration, noise, and temperature based on historical and real-time data, and generate vibration correction terms; Quantify the impact of noise on vibration control through environmental parameters and equipment operating status, and generate noise correction terms; Based on temperature data and equipment operating status, quantify the impact of temperature on vibration and noise control and generate temperature correction terms; Based on the vibration correction term, the noise correction term and the temperature correction term, and in combination with the reference control parameters, a vibration-noise-temperature multi-dimensional control threshold is generated.

5. The method for controlling the operating environment of a pump and fan with monitoring and feedback according to claim 4, characterized in that: Based on a closed-loop feedback mechanism, the equipment optimization control model is generated by combining the coupling characteristics of vibration, noise, and temperature, including: generating a feedback matrix of vibration, noise and temperature based on actual effects of the first control and the second control; Combine the coupling characteristics of vibration, noise and temperature to extract the deviation characteristics of control effect; generating an adaptive control algorithm based on the deviation characteristics and historical control data; Adaptive control algorithms include a combination of genetic algorithms, neural networks, and fuzzy logic control, where: Optimize control parameters based on dynamic fitness function through genetic algorithm; Extract local features and time series characteristics of vibration, noise and temperature signals through neural networks; Through fuzzy logic control, multi-dimensional reasoning is performed based on automatically generated fuzzy rules to generate control instructions.

6. A monitoring and feedback method for controlling the operating environment of a pump and fan according to claim 5, characterized in that: The equipment optimization control model is combined with long-term operation data to generate a time-varying performance optimization model, including: Generate long-term performance characteristics of equipment operating status through equipment optimization control model; Based on long-term operating data, combined with environmental changes and equipment aging characteristics, performance degradation correction items are generated.

7. A monitoring and feedback method for controlling the operating environment of a pump and fan according to claim 6, characterized in that: Through the user interaction interface, the device status, monitoring data and control effects are displayed in real time; based on user feedback, the control strategy and parameters are adjusted to generate a user-optimized control model.

8. A system for monitoring and feedback of pump and fan operating environment control, applying a method for monitoring and feedback of pump and fan operating environment control according to any one of claims 1 to 7, characterized in that: include: Acquisition module, used to obtain equipment operating parameters and build equipment status monitoring model; A first analysis module is used to collect vibration, noise and temperature data in real time through the equipment status monitoring model, identify the equipment operating status, and determine abnormal points; A first control model building module is used to dynamically generate vibration, noise and temperature control strategy models by combining abnormal point data with environmental parameters for the first control of the equipment operation status; A second control model building module is used to analyze the dynamic characteristics of vibration, noise and temperature in combination with the first control, obtain vibration-noise-temperature multi-dimensional control thresholds, and perform second control of the equipment operating status; An optimization model building module is used to generate an equipment optimization control model based on a closed-loop feedback mechanism and the coupling characteristics of vibration, noise, and temperature; The performance optimization model building module is used to generate a time-varying performance optimization model based on the equipment optimization control model and long-term operation data to continuously optimize and control the equipment vibration, noise and temperature.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of a monitoring and feedback method for controlling the operating environment of a pump and a fan are implemented as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of a monitoring and feedback method for controlling the operating environment of a pump and a fan are implemented as described in any one of claims 1 to 7.

Images