An energy-saving variable frequency coordinated control method for oil fume purification equipment

By using a multi-dimensional sensor data acquisition and multi-variable collaborative control optimization platform, combined with a collaborative temperature control model and robust optimization algorithm, the shortcomings of oil fume purification equipment in multi-parameter collaborative control are solved, and the equipment achieves high efficiency, energy saving and stable purification under dynamic operating conditions.

CN122281335APending Publication Date: 2026-06-26SHANDONG JINGXI KITCHEN EQUIP TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610408837.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-03-31
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

Existing fume purification equipment lacks a precise control mechanism that coordinates multiple parameters, and the control model and optimization algorithm are not well adapted, resulting in a disconnect between frequency conversion regulation and purification effect, high energy consumption, and unstable purification effect.

Method used

By collecting data such as oil fume concentration, airflow rate, and equipment operating power through multi-dimensional sensing modules, a multi-source data acquisition matrix is ​​constructed. Abnormal data is eliminated using a multi-variable collaborative control optimization platform, and a variable frequency and purification collaborative temperature control model is established. Combined with an oil fume load variable frequency prediction model and an energy-saving variable frequency robust optimization algorithm, the optimal variable frequency control parameter set is generated to achieve the collaborative operation of the variable frequency module and the purification module.

Benefits of technology

It achieves efficient and energy-saving operation of the fume purification equipment under dynamic working conditions, improves the stability of the purification effect and the adaptability of the equipment, and meets the increasing environmental protection standards and market demand for coordinated control of equipment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122281335A_ABST
    Figure CN122281335A_ABST
Patent Text Reader

Abstract

This invention discloses an energy-saving variable frequency collaborative control method for oil fume purification equipment, comprising: collecting multi-source operating data such as oxygen flux, coal quality characteristics, combustion load, furnace temperature, and pollutant generation through multi-dimensional sensing modules; inputting the processed data into a multi-variable collaborative optimization platform; and sequentially executing core steps such as collaborative temperature control model construction, combustion load prediction and simulation, and energy-saving robust optimization algorithm calculation, forming a closed-loop analysis mechanism. This method integrates multi-dimensional operating parameters, establishes a dynamic correlation between oxygen flux and combustion judgment indicators, accurately captures load change characteristics, and scientifically optimizes control parameters to adapt to complex combustion conditions. It breaks through the limitations of traditional single-variable analysis, significantly improves the pertinence and reliability of analysis results, and provides strong data support for the precise control of combustion systems. It is applicable to the operation status analysis and optimization control scenarios of various coal combustion equipment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of energy-saving frequency conversion technology for purification equipment, and in particular to an energy-saving frequency conversion coordinated control method for oil fume purification equipment. Background Technology

[0002] In the catering industry and industrial fume emission scenarios, fume purification equipment, as a key environmental protection device, must simultaneously meet the dual requirements of high-efficiency purification and energy-saving operation. With increasingly stringent environmental standards, the fixed-frequency operation mode of traditional equipment is no longer adequate to adapt to the dynamic changes in fume load, leading to a growing contradiction between unstable purification efficiency and energy waste. The current market demands increasingly intelligent and collaborative control of equipment, necessitating the integration of multi-dimensional operating parameters to construct a control mechanism that deeply coordinates frequency conversion and purification. This will achieve precise matching between equipment operating status and real-time conditions, solving the dynamic control problem under different fume concentrations, airflow rates, and equipment wear conditions, and driving the development of fume purification equipment towards high efficiency, energy saving, stability, and reliability.

[0003] Existing control technologies for oil fume purification equipment have two main drawbacks: First, they lack a precise control mechanism that coordinates multiple parameters, relying solely on oil fume concentration or operating power for frequency conversion adjustment. This fails to fully consider the dynamic correlation between key parameters such as purification chamber temperature and filter clogging degree and frequency conversion parameters, leading to a disconnect between frequency conversion adjustment and purification effect, and preventing the formation of closed-loop coordinated control. Second, the adaptability of control models and optimization algorithms is insufficient. Predictive models do not effectively integrate load variation characteristics under multiple operating conditions, and optimization algorithms do not balance energy efficiency and equipment operational stability. This results in a lack of scientific basis for frequency conversion adjustment range correction, making it difficult to generate optimal control parameters that adapt to complex and variable oil fume emission conditions. Consequently, the equipment operates at high energy consumption and experiences significant fluctuations in purification effect. Summary of the Invention

[0004] In order to overcome the shortcomings and deficiencies of the existing technology, the present invention provides an energy-saving variable frequency collaborative control method for oil fume purification equipment.

[0005] The technical solution adopted in this invention is an energy-saving variable frequency coordinated control method for oil fume purification equipment, comprising the following steps: S1, collecting relevant sensor data such as oil fume concentration, airflow rate, equipment operating power, purification chamber temperature, variable frequency module output frequency, and filter clogging degree through a multi-dimensional sensing module mounted on the oil fume purification equipment, and establishing a multi-source data acquisition matrix; S2, inputting the collected multi-source data into a multi-variable coordinated control optimization platform, removing abnormal data through the platform's built-in data filtering mechanism, and extracting calibration feature parameters directly related to energy-saving variable frequency and purification efficiency; S3, constructing a variable frequency and purification coordinated temperature control model based on the calibration feature parameters, and using this model... S4. The dynamic correlation between temperature change and frequency conversion parameters is mapped to determine the initial frequency conversion adjustment range; S5. The oil fume load frequency conversion prediction model is used to deduce the oil fume load change trend under different working conditions, output the load dynamic change sequence, and correct the initial frequency conversion adjustment range; S6. The corrected frequency conversion adjustment range is optimized using an energy-saving frequency conversion robust optimization algorithm to generate the optimal frequency conversion control parameter set adapted to the real-time working conditions; S7. The optimal frequency conversion control parameter set is transmitted to the actuator of the oil fume purification equipment to enable the coordinated operation of the frequency conversion module and the purification module. The operating data is fed back in real time through the multi-variable collaborative control optimization platform to dynamically adjust the control parameters.

[0006] Furthermore, the expression for the frequency conversion and purification synergistic temperature control model is: ,in, To achieve optimal temperature control within the purification chamber. This is the frequency weighting coefficient for the frequency conversion. The frequency output of the inverter module, The concentration of cooking fumes, For operating power weighting coefficient, For equipment operating power, The degree of filter clogging This is the airflow rate weighting coefficient. For airflow rate, This is the cavity temperature weighting coefficient. To purify the current temperature of the cavity, For adjustment coefficients, This is the amplitude of frequency conversion adjustment.

[0007] Furthermore, the expression for the variable frequency prediction model of the fume load is: ,in, To predict the oil fume load, For frequency load factor, The frequency output of the inverter module. The concentration of cooking fumes, The airflow blockage coefficient, For airflow rate, The degree of filter clogging For power temperature coefficient, For equipment operating power, To purify the current temperature of the cavity, This is the temperature fluctuation coefficient. To adjust the load factor, This is the amplitude of frequency conversion adjustment.

[0008] Furthermore, the expression for the energy-saving variable frequency robust optimization algorithm is: ,in, To achieve the optimal frequency conversion, As the load deviation weight, To predict the oil fume load, For the rated oil fume load, For power frequency weighting, For equipment operating power, The frequency output of the inverter module. To block the adjustment weight, The degree of filter clogging For frequency conversion adjustment amplitude, As a weight for temperature deviation, To purify the current temperature of the cavity, To achieve optimal temperature control for the purification chamber.

[0009] Furthermore, the parameter coordination model expression of the multivariable collaborative control optimization platform is as follows: ,in, For multivariate coordination coefficients, The temperature coherence coefficient, To achieve optimal temperature control within the purification chamber. This is the load coordination factor. To predict the oil fume load, For frequency conversion coordination coefficient, To achieve the optimal frequency conversion, The concentration synergy coefficient, The concentration of cooking fumes, The airflow coordination coefficient, For airflow rate, For blocking coordination coefficient, The degree of filter clogging The power attenuation coefficient, This refers to the operating power of the equipment.

[0010] Furthermore, the comprehensive parameter model expression for the energy-saving variable frequency coordinated control of the fume purification equipment is as follows: ,in, For the comprehensive indicators of coordinated control, The variable frequency temperature coordination coefficient, To achieve the optimal frequency conversion, To achieve optimal temperature control within the purification chamber. The load congestion coordination coefficient, To predict the oil fume load, The degree of filter clogging The power concentration synergy coefficient, For equipment operating power, The concentration of cooking fumes, For airflow rate, The coefficients represent multivariate coordination coefficients.

[0011] Further, step S3 includes the following sub-steps: S31, selecting the purification chamber temperature, inverter module output frequency, oil fume concentration, and equipment operating power as calibration input parameters, constructing a four-dimensional parameter correlation matrix, and clarifying the priority of each parameter in temperature control; S32, based on the parameter correlation matrix, establishing an initial correlation model between temperature and inverter parameters using a nonlinear mapping method, calibrating the initial parameters of the model through sensor data, and reducing parameter coupling interference; S33, introducing filter blockage degree and airflow rate as correction factors, dynamically adjusting the initial correlation model, and optimizing the model's adaptability to different blockage conditions; S34, verifying the model output accuracy through a multi-variable collaborative control optimization platform, fine-tuning the model weight coefficients based on the verification results, and forming the final inverter and purification collaborative temperature control model.

[0012] Further, step S4 includes the following sub-steps: S41, collecting historical oil fume load data, corresponding frequency conversion parameters, and operating condition parameters, dividing the training dataset and validation dataset according to time series to ensure that the dataset covers different operating conditions; S42, inputting the training dataset into the initially constructed prediction model framework, optimizing the internal parameters of the model through gradient descent, and establishing a nonlinear relationship between oil fume load and input parameters; S43, using the validation dataset to test the model's prediction performance, identifying operating condition intervals with large prediction deviations, and adjusting the model structure and parameters accordingly; S44, combining the real-time collected operating condition data to perform online correction of the model's prediction results, generating an oil fume load frequency conversion prediction model that can reflect real-time changing trends.

[0013] Further, S5 includes the following sub-steps: S51, determining energy efficiency, frequency conversion stability, purification effect, and equipment loss as optimization objectives, establishing a multi-objective optimization function, and clarifying the constraints and boundary ranges of each objective; S52, using frequency conversion adjustment amplitude, frequency adjustment step size, and power allocation ratio as optimization variables, constructing a variable search space, and excluding variable combinations that do not meet the equipment operation requirements; S53, using an iterative optimization strategy to evaluate each variable combination in the search space, calculating the optimization objective value corresponding to each combination, and selecting the set of non-dominated solutions; S54, based on the parameter coordination requirements of the multi-variable collaborative control optimization platform, performing a second screening of the set of non-dominated solutions, determining the optimal variable combination, and forming an energy-saving frequency conversion robust optimization algorithm.

[0014] An energy-saving variable frequency coordinated control method for oil fume purification equipment is disclosed. This method is implemented through different units, including: a multi-dimensional sensor data acquisition unit, a multi-source data filtering and processing unit, a variable frequency and purification coordinated temperature control model construction unit, an oil fume load variable frequency prediction model derivation unit, an energy-saving variable frequency robust optimization algorithm calculation unit, and a multi-variable coordinated control execution unit. The multi-dimensional sensor data acquisition unit and the multi-source data filtering and processing unit are unidirectionally connected, used to collect multi-source operating data such as oil fume concentration and airflow rate and transmit them to the filtering and processing unit. The multi-source data filtering and processing unit is connected to the variable frequency and purification coordinated temperature control model construction unit and the oil fume load variable frequency prediction model derivation unit. The system features bidirectional connections between units, filtering the collected data and providing it to two model units while simultaneously receiving parameter adjustment requests from the models. Both model units are unidirectionally connected to the energy-saving variable frequency robust optimization algorithm calculation unit, outputting model results for the algorithm calculation unit to use. The energy-saving variable frequency robust optimization algorithm calculation unit is unidirectionally connected to the multivariable collaborative control execution unit, transmitting the optimized control parameters to the execution unit. The multivariable collaborative control execution unit is bidirectionally connected to the multi-dimensional sensor data acquisition unit, executing control commands and feeding back operating data to form a closed-loop control. Each unit interacts with parameters via a data bus to ensure the continuity and accuracy of collaborative control.

[0015] Beneficial Effects: This invention proposes an energy-saving variable frequency coordinated control method for oil fume purification equipment. By integrating multi-dimensional operating parameters to construct a multi-variable coordinated control optimization platform, with a variable frequency and purification coordinated temperature control model, an oil fume load variable frequency prediction model, and an energy-saving variable frequency robust optimization algorithm as the core, it addresses the problem of lacking a multi-parameter coordinated control mechanism by comprehensively incorporating key parameters such as oil fume concentration, airflow rate, purification chamber temperature, and filter clogging degree into the control system. It establishes a dynamic correlation between each parameter and variable frequency adjustment, achieving deep coordination between the variable frequency module and the purification module to form a complete closed-loop control. Addressing the problem of insufficient adaptability between the control model and the optimization algorithm, it accurately captures load change characteristics under multiple operating conditions through a load prediction model, and utilizes a robust optimization algorithm to balance energy efficiency and operational stability, scientifically correcting the variable frequency adjustment range and generating optimal control parameters to adapt to complex and variable emission conditions. Meanwhile, the multi-unit collaborative control system architecture ensures the continuity of parameter interaction and command execution, significantly improving the equipment's adaptability to dynamic working conditions. While enhancing the stability of purification effect, it reduces operating energy consumption, promotes the intelligent, efficient, and energy-saving operation of fume purification equipment, and meets the core needs of environmental protection standards and the market for collaborative equipment control. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the overall process of the method of the present invention.

[0017] Figure 2 This is a flowchart of method step S3 of the present invention;

[0018] Figure 3 This is a flowchart of method step S4 of the present invention;

[0019] Figure 4 This is a flowchart of step S5 of the method of the present invention;

[0020] Figure 5 This is a diagram showing the unit composition for implementing the method of the present invention. Detailed Implementation

[0021] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] like Figure 1As shown, an energy-saving variable frequency coordinated control method for an oil fume purification device includes the following steps: S1, collecting relevant sensor data such as oil fume concentration, airflow rate, equipment operating power, purification chamber temperature, variable frequency module output frequency, and filter clogging degree through a multi-dimensional sensing module mounted on the oil fume purification device, and establishing a multi-source data acquisition matrix; S2, inputting the collected multi-source data into a multi-variable coordinated control optimization platform, removing abnormal data through the platform's built-in data filtering mechanism, and extracting calibration feature parameters directly related to energy-saving variable frequency and purification efficiency; S3, constructing a variable frequency and purification coordinated temperature control model based on the calibration feature parameters, and mapping temperature through this model. The dynamic correlation between changes and frequency conversion parameters determines the initial frequency conversion adjustment range; S4, the frequency conversion prediction model for oil fume load is used to deduce the trend of oil fume load changes under different operating conditions, outputs the dynamic change sequence of load, and corrects the initial frequency conversion adjustment range; S5, the energy-saving frequency conversion robust optimization algorithm is used to perform multi-objective optimization on the corrected frequency conversion adjustment range to generate the optimal frequency conversion control parameter set adapted to real-time operating conditions; S6, the optimal frequency conversion control parameter set is transmitted to the actuator of the oil fume purification equipment to enable the coordinated operation of the frequency conversion module and the purification module. The operating data is fed back in real time through the multi-variable collaborative control optimization platform to dynamically adjust the control parameters.

[0023] Step S1 is the data acquisition stage, which forms the foundation of the entire analysis method. Multi-dimensional sensing modules are deployed at key locations such as the furnace inlet and outlet, burner perimeter, critical flue sections, and equipment control box of the coal combustion equipment to comprehensively collect core operating parameters. The sensing modules include high-precision gas sensors, flow rate sensors, temperature sensors, power sensors, and resistance sensors. The acquisition frequency is set to 20 times per second, with a continuous acquisition duration of no less than one hour, ensuring coverage of different operating conditions such as combustion start-up, stable operation, and load adjustment. The acquired parameters specifically include oxygen flux, coal quality characteristics, combustion load, furnace temperature, frequency converter output frequency, and resistance parameters corresponding to filter blockage. Each parameter acquisition is accompanied by a timestamp and operating condition identifier, ultimately forming a three-dimensional multi-source data acquisition matrix encompassing time, parameter, and operating condition dimensions. This matrix provides complete and continuous raw data support for subsequent data processing and model construction. During the acquisition process, a sensor self-calibration mechanism reduces errors caused by environmental interference, ensuring the accuracy and consistency of the data acquisition.

[0024] Step S2 involves data filtering and core feature extraction of the three-dimensional multi-source data acquisition matrix obtained in step S1. This is specifically achieved through the multi-level data processing mechanism built into the multi-variable collaborative optimization platform. First, an outlier removal strategy based on statistical distribution is adopted, setting normal fluctuation ranges for each parameter. The oxygen flux fluctuation range is controlled within ±15% of the set baseline value, the furnace temperature fluctuation range is ±20℃, and the combustion load fluctuation range is ±10%. Data exceeding these ranges are marked as outliers and removed. Second, through data continuity verification, missing data segments caused by sensor malfunctions during the acquisition process are deleted, and short-term missing data are supplemented using linear interpolation. After data screening, the correlation between each parameter and the analysis results of oxygen flux impact are calculated based on the correlation analysis algorithm. Core parameters with correlation coefficients higher than 0.7 with oxygen flux and combustion judgment indicators are retained, including oxygen flux, combustion load, furnace temperature, filter clogging degree, and frequency converter output frequency. At the same time, derived features such as the rate of change, peak value, and mean value of each core parameter are extracted, and finally a feature dataset including the original core parameters and derived features is formed. This dataset effectively reduces data dimensionality and redundant information, reducing the computational load for subsequent model construction and optimization algorithm calculations, while ensuring the relevance and effectiveness of the input data.

[0025] Step S3, based on the feature dataset extracted in step S2, constructs a variable frequency and purification coordinated temperature control model. The specific implementation process revolves around three core stages: parameter correlation modeling, model calibration, and dynamic correction. First, furnace temperature is selected as the core output parameter, while oxygen flux, variable frequency module output frequency, combustion load, and filter clogging degree are used as input parameters. A nonlinear mapping model framework is constructed. Initial parameters of the model are trained using matching data from over 1000 sets of historically collected data under different operating conditions, clarifying the influence weight of each input parameter on furnace temperature. After model training, calibration is performed using 50 sets of real-time collected continuous operating condition data to adjust parameter weight coefficients and reduce coupling interference between parameters. Subsequently, the dynamic change rate of filter clogging degree is introduced as a correction factor, establishing a time-series-based dynamic adjustment mechanism. When the change rate of filter clogging degree exceeds 5% per hour, model parameter fine-tuning is automatically triggered. Simultaneously, the multivariate collaborative optimization platform receives the model output results and the actual monitored furnace temperature data in real time, calculates the deviation value and feeds it back to the model, continuously optimizes the model's mapping accuracy, and finally forms a collaborative temperature control model that can accurately reflect the dynamic correlation between input parameters and furnace temperature. Through this model, the initial frequency conversion adjustment range under different operating conditions is output, providing a basic adjustment range for subsequent load prediction and parameter optimization.

[0026] Step S4 employs a variable frequency drive (VFD) prediction model for fume load to extrapolate the load variation trend of oxygen flux under different operating conditions. Specifically, a training dataset and a validation dataset are first divided. Historically collected multi-condition data are allocated in a 7:3 ratio. The training dataset is used for model training, and the validation dataset is used for performance testing. During model construction, historical data on oxygen flux, combustion load, furnace temperature, and filter blockage level are used as inputs, and load changes over the next 10, 30, and 60 minutes are used as outputs. A time-series prediction algorithm is used to construct the model framework, and iterative training optimizes the time step parameters and feature extraction dimensions within the model. After model training, the prediction accuracy is tested using the validation dataset to ensure that short-term prediction deviation is controlled within 8% and medium- to long-term prediction deviation is controlled within 12%. For high-load, low-oxygen-flux operating conditions with large prediction deviations, 200 sets of specific operating condition data are added for model reinforcement training. In actual simulation, the real-time core feature parameters extracted in step S2 are input into the trained prediction model, which outputs a dynamic load change sequence for different time periods in the future. This sequence includes key information such as load change amplitude, change rate, and peak occurrence time. Based on this sequence, the initial frequency conversion adjustment range output in step S3 is specifically modified to expand the adjustment range under high load conditions and reduce the fluctuation amplitude of the adjustment range under low load conditions, so that the adjustment range is more in line with the actual load change requirements.

[0027] Step S5 utilizes an energy-saving variable frequency robust optimization algorithm to perform multi-objective optimization on the corrected variable frequency control range, generating an optimal set of variable frequency control parameters adapted to real-time operating conditions. During implementation, the optimization objectives are first defined, including four core objectives: reducing operating energy consumption, improving regulation stability, ensuring combustion efficiency, and reducing equipment losses. Constraints are set for each objective: the variable frequency control frequency change step size does not exceed 2 Hz per second, the maximum regulation frequency does not exceed 120% of the rated frequency, and the minimum regulation frequency is not lower than 60% of the rated frequency. Subsequently, the variable frequency control amplitude, frequency control step size, and power allocation ratio are used as optimization variables to construct a multi-dimensional variable search space. A hierarchical iterative optimization strategy is employed to evaluate each combination of variables within the search space, calculating the comprehensive score of the four optimization objectives for each variable combination. The evaluation process incorporates a weighting mechanism, adjusting the weights of each objective based on the real-time operating condition. Under high-load conditions, the weight of combustion efficiency is increased, while under low-load conditions, the weight of energy saving is increased. Through more than 100 iterations of calculations, the set of non-dominated solutions with the highest comprehensive score is selected. Then, combined with the parameter coordination requirements of the multi-variable collaborative control optimization platform, the set of non-dominated solutions is further screened to finally determine the unique optimal set of variable frequency control parameters. This set of parameters not only meets the load requirements of the real-time operating condition but also balances multiple objectives, providing a core basis for precise equipment control.

[0028] Step S6, serving as the execution and feedback link of the entire control method, first transmits the optimal variable frequency control parameter set determined in step S5 to the actuator of the coal combustion equipment via the data transmission module. The actuator includes a variable frequency controller, a combustion regulator, and a temperature control module. Each module coordinates its actions according to the instructions in the parameter set. The variable frequency controller adjusts its output frequency to the optimal value, the combustion regulator adjusts the fuel supply and air intake accordingly, and the temperature control module adjusts the operating status of the furnace heat exchange device in conjunction with the control, achieving synchronous and coordinated operation of the variable frequency module and the purification module. During equipment operation, the multi-dimensional sensing module continuously collects real-time operating data, including adjusted oxygen flux, furnace temperature, combustion load, and equipment operating power, feeding back the data to the multi-variable collaborative control optimization platform every 5 seconds. After receiving the data, the platform compares the deviation between the actual operating parameters and the optimal parameter set in real time. When the oxygen flux deviation exceeds ±5%, the furnace temperature deviation exceeds ±10℃, or the combustion load deviation exceeds ±8%, the parameter adjustment process is automatically triggered. The deviation data is fed back to the collaborative temperature control model in step S3 and the load prediction model in step S4, and the model parameter fine-tuning and adjustment range correction are initiated to form a complete closed-loop control loop. Through continuous parameter feedback and dynamic adjustment, the equipment is ensured to always operate in the optimal state, and the precise control and dynamic adaptation of the oxygen flux on the combustion process and judgment indicators are achieved.

[0029] Preferably, the expression for the frequency conversion and purification synergistic temperature control model is: ,in, To achieve optimal temperature control within the purification chamber. This is the frequency weighting coefficient for the frequency conversion. The frequency output of the inverter module. The concentration of cooking fumes, For operating power weighting coefficient, For equipment operating power, The degree of filter clogging This is the airflow rate weighting coefficient. For airflow rate, This is the cavity temperature weighting coefficient. To purify the current temperature of the cavity, For adjustment coefficients, This is the amplitude of frequency conversion adjustment.

[0030] Specifically, the variable frequency and purification synergistic temperature control model is based on a multi-parameter coupling influence mechanism. It analyzes the nonlinear correlation between core parameters and the purification chamber temperature through system analysis and employs a step-by-step construction strategy to form the formula. The core driving factors for temperature regulation are identified as the synergistic effect between variable frequency output parameters and oil fume concentration, and the coupling effect between operating power and filter blockage degree. These two are set as numerators to reflect the positive driving relationship. Simultaneously, airflow rate and the current chamber temperature are used as denominators to reflect their inhibitory effect on temperature regulation. The ratio of the numerator and denominator is used to initially establish the correlation framework between temperature and key parameters. Subsequently, a logarithmic function term is introduced to nonlinearly correct the influence of the variable frequency adjustment amplitude, adapting to the nonlinear characteristics of temperature response under different adjustment amplitudes. Each weighting coefficient and adjustment coefficient is calibrated through more than 1000 sets of multi-condition tests. The weighting coefficients are set between 0.3 and 0.8 based on the sensitivity of the parameters to temperature influence, and the adjustment coefficients are between 0.5 and 1.2, ensuring that the model output highly matches the actual temperature change trend. The implementation of this model utilizes a multivariable collaborative control optimization platform. Real-time collected parameter data is input into the platform's built-in calculation module to complete the calculations and output the optimal control temperature of the purification chamber. This provides a precise target benchmark for subsequent frequency conversion regulation. The model's logic lies in fully considering the dynamic coupling relationship between various parameters, avoiding temperature control deviations caused by single-parameter driving, and enhancing the model's adaptability to complex working conditions through nonlinear functions and coefficient calibration, thereby ensuring the accuracy and stability of temperature control.

[0031] Preferably, the expression for the frequency conversion prediction model of the fume load is: ,in, To predict the oil fume load, For frequency load factor, The frequency output of the inverter module. The concentration of cooking fumes, The airflow blockage coefficient, For airflow rate, The degree of filter clogging For power temperature coefficient, For equipment operating power, To purify the current temperature of the cavity, This is the temperature fluctuation coefficient. To adjust the load factor, This is the amplitude of frequency conversion adjustment.

[0032] Specifically, the variable frequency load prediction model for cooking fumes is constructed step by step, based on time-series data correlation analysis and combined with the dynamic characteristics of load changes and the comprehensive influence of multiple parameters. First, by mining the correlation between historical load data and various parameters, the square term of the variable frequency output parameters and the power term of the cooking fume concentration are identified as the core driving terms for load prediction, reflecting the nonlinear influence of parameters on the load. The product of airflow rate and filter blockage degree is used as a secondary driving term, reflecting the synergistic effect of fluid characteristics and equipment status on the load. A product term of operating power and cavity temperature sinusoidal function is introduced to adapt to the periodic influence of temperature fluctuations on the load. Simultaneously, a linear term of the variable frequency adjustment amplitude is added to correct short-term load fluctuation deviations. Parameter values ​​are calibrated through extensive operating condition tests. The load correlation coefficient is set between 0.2 and 0.9 based on the load response sensitivity under different operating conditions, the temperature fluctuation coefficient is determined between 0.1 and 0.5 based on common temperature fluctuation cycles, and other coefficients are set between 0.3 and 1.0, ensuring that the model's short-term prediction deviation is controlled within 8% and the medium- and long-term prediction deviation is controlled within 12%. The implementation of this model requires first dividing historical data into training and validation sets in a 7:3 ratio. The optimal values ​​of each coefficient are determined through training. After successful validation, the model is embedded into a multivariate collaborative control optimization platform. It receives core parameter data in real time and substitutes it into the model for calculation, outputting the dynamic load change sequence for different time periods. The model is based on the fact that load changes are affected by multiple parameters and exhibit nonlinear and periodic characteristics. By combining different forms of parameter terms and function types, the model comprehensively covers the influence mechanism of various factors, improving the accuracy and reliability of the model's prediction of load changes under complex working conditions.

[0033] Preferably, the expression for the energy-saving variable frequency robust optimization algorithm is: ,in, To achieve the optimal frequency conversion, As the load deviation weight, To predict the oil fume load, For the rated oil fume load, For power frequency weighting, For equipment operating power, The frequency output of the inverter module. To block the adjustment weight, The degree of filter clogging For frequency conversion adjustment amplitude, As a weight for temperature deviation, To purify the current temperature of the cavity, To achieve optimal temperature control for the purification chamber.

[0034] Specifically, the energy-saving variable frequency robust optimization algorithm is based on multi-objective optimization theory, constructing an optimization function with four core objectives: minimizing load deviation, optimizing energy consumption, controlling equipment losses, and ensuring temperature stability. Each optimization objective is determined as follows: the square of the deviation between the predicted load and the rated load is taken as the primary objective, reflecting the requirement for precise load matching; the ratio of operating power to relevant parameters of the variable frequency output is taken as the energy consumption optimization objective, reflecting the core energy-saving requirement; the product of the filter blockage degree and the variable frequency adjustment amplitude is taken as the equipment loss control objective; and the absolute value of the temperature deviation is taken as the temperature stability objective. A multi-objective optimization function is constructed through weighted summation. The weight coefficients are dynamically adjusted according to the optimization priority of different operating conditions: the load deviation weight is between 0.4 and 0.7, the energy consumption weight is between 0.2 and 0.5, and the blockage adjustment weight and temperature deviation weight are both between 0.1 and 0.3, ensuring balanced optimization across multiple objectives. The implementation of this algorithm requires the use of the optimization module of a multivariable collaborative control optimization platform. First, the objective constraints and parameter value ranges are input. The corrected frequency conversion adjustment range is used as the variable search space. The optimal frequency conversion frequency is selected through more than 100 iterations. Its establishment logic is to take into account the needs of multiple objectives and avoid the overall performance degradation caused by the optimization of a single objective. The weight allocation mechanism adapts to the optimization focus of different working conditions. The robust design ensures that the algorithm can still output stable and reliable optimization results under parameter fluctuations, providing a scientific and accurate parameter basis for frequency conversion control.

[0035] Preferably, the parameter coordination model expression of the multivariable collaborative control optimization platform is: ,in, For multivariate coordination coefficients, The temperature coherence coefficient, To achieve optimal temperature control within the purification chamber. This is the load coordination factor. To predict the oil fume load, For frequency conversion coordination coefficient, To achieve the optimal frequency conversion, The concentration synergy coefficient, The concentration of cooking fumes, The airflow coordination coefficient, For airflow rate, For blocking coordination coefficient, The degree of filter clogging The power attenuation coefficient, This refers to the operating power of the equipment.

[0036] Specifically, the parameter coordination model of the multivariable collaborative control optimization platform is based on the interactive coupling mechanism between the output results of each core model and multi-source parameters. It constructs a collaborative correlation framework by integrating the key outputs of preceding models and the original core parameters. The core components of the coordination coefficient are determined to be the optimal control temperature, predicted load, and optimal frequency conversion. These three factors, as numerators, reflect the positive driving effect of each core model output on collaborative control. Fume concentration, airflow rate, and filter blockage degree are used as denominators to reflect the constraining influence of basic operating condition parameters on the coordination effect. The correlation logic between the coordination coefficient and each parameter is initially established through the ratio of the numerator and denominator. Subsequently, an exponential function term is introduced to attenuate and correct the impact on operating power, adapting to the weakening characteristics of the coordination effect under high-power conditions. Parameter values ​​are calibrated through more than 500 sets of multi-condition cross-tests. The temperature coordination coefficient, load coordination coefficient, and frequency conversion coordination coefficient are between 0.3 and 0.8; the concentration coordination coefficient, airflow coordination coefficient, and blockage coordination coefficient are between 0.2 and 0.6; and the power attenuation coefficient is between 0.1 and 0.4. The implementation of this model requires embedding the collaborative analysis module of the multivariable collaborative control optimization platform. This module receives the output results of the preceding model and the basic parameters collected in real time. The platform's built-in computing unit calculates the collaborative coefficients, quantifies the collaborative strength of each core model and basic parameter, and provides a quantitative basis for multivariable collaborative control. During implementation, the parameter interaction weights of each module are dynamically adjusted through the collaborative coefficients to ensure deep collaboration between frequency conversion regulation and purification operation, and avoid control disconnection caused by independent operation of each module.

[0037] Preferably, the comprehensive parameter model expression for the energy-saving variable frequency coordinated control of the fume purification equipment is as follows: ,in, For the comprehensive indicators of coordinated control, The variable frequency temperature coordination coefficient, To achieve the optimal frequency conversion, To achieve optimal temperature control within the purification chamber. The load congestion coordination coefficient, To predict the oil fume load, The degree of filter clogging The power concentration synergy coefficient, For equipment operating power, The concentration of cooking fumes, For airflow rate, The coefficients represent multivariate coordination coefficients.

[0038] Specifically, the comprehensive parameter model for energy-saving variable frequency coordinated control of fume purification equipment is based on the overall goal of coordinated control. It constructs a comprehensive evaluation framework by integrating the core model output results, basic operating parameters, and coordination coefficients. The core components of the comprehensive index are determined as follows: the product of the optimal variable frequency and the optimal control temperature is used as the variable frequency temperature coordination term; the product of the predicted load and the filter blockage compensation value is used as the load blockage coordination term; and the product of the operating power, the ratio of fume concentration and airflow rate, and the coordination coefficient is used as the power concentration coordination term. These three terms are superimposed to form the comprehensive index expression. Parameter values ​​are calibrated through multi-condition optimization tests. The variable frequency temperature coordination coefficient and the load blockage coordination coefficient are set between 0.4 and 0.9, and the power concentration coordination coefficient is set between 0.3 and 0.7, ensuring that the weights of each coordination term are balanced and conform to actual control requirements. The implementation of this model requires the use of the comprehensive evaluation module of the multivariable collaborative control optimization platform to collect the core parameters and the output results of the preceding models in real time, and substitute them into the model to complete the calculation of comprehensive indicators. Its establishment is based on the comprehensive quantification of the overall effect of collaborative control. The comprehensive indicators reflect the collaborative level of frequency conversion regulation, temperature control, load adaptation, and energy consumption optimization. During the implementation process, the goal is to maximize the comprehensive indicators, and the parameter weights of each core model are adjusted in reverse to ensure that the collaborative control reaches the optimal state. This provides a unified quantitative evaluation standard for the collaborative control effect. Through the dynamic monitoring and feedback of this indicator, the operating performance of the entire control system is continuously optimized to ensure the comprehensive operating effect of the equipment under complex working conditions.

[0039] Preferred, such as Figure 2 As shown, step S3 includes the following sub-steps: S31, selecting the purification chamber temperature, inverter module output frequency, oil fume concentration, and equipment operating power as calibration input parameters, constructing a four-dimensional parameter correlation matrix, and clarifying the priority of each parameter in temperature control; S32, based on the parameter correlation matrix, establishing an initial correlation model between temperature and inverter parameters using a nonlinear mapping method, calibrating the initial parameters of the model through sensor data, and reducing parameter coupling interference; S33, introducing filter blockage degree and airflow rate as correction factors, dynamically adjusting the initial correlation model, and optimizing the model's adaptability to different blockage conditions; S34, verifying the model output accuracy through a multi-variable collaborative control optimization platform, fine-tuning the model weight coefficients based on the verification results, and forming the final inverter and purification collaborative temperature control model.

[0040] Specifically, steps S31 to S34 of step S3 involve constructing a variable frequency and purification-coordinated temperature control model. S31 first selects the purification chamber temperature, variable frequency output parameters, oil fume concentration, and equipment operating power as four-dimensional input parameters from the core parameters extracted in step S2. The priority of each parameter's influence is determined through correlation ranking, with the correlation weight between the variable frequency output parameters and the purification chamber temperature set to the highest. Then, a four-dimensional parameter correlation matrix is ​​constructed, including parameter name, acquisition time, and numerical range, providing a basic parameter framework for subsequent modeling. S32 uses this matrix to build an initial correlation model using a nonlinear mapping method. 1000 sets of historical matching data are selected to train the initial model parameters. Simultaneously, the parameters are initially calibrated using real-time sensor data. A deviation correction mechanism is used to reduce coupling interference between parameters, keeping the output deviation of the initial model within 10%. S33 introduces filter blockage degree and airflow rate as dynamic correction factors. A 5% change in filter blockage degree triggers a model parameter adjustment, and a change in airflow rate exceeding 10% initiates factor weight optimization. Continuous iterative adjustments optimize the model's adaptability to different blockage conditions. S34 transmits the model output results to the multivariable collaborative control optimization platform. The platform's built-in accuracy verification module compares the model output values ​​with the actual monitored values, calculates the deviation rate, and fine-tunes the model weight coefficients according to the deviation rate level. When the deviation rate is below 5%, the model construction is considered complete, ultimately forming a frequency conversion and purification collaborative temperature control model that can accurately reflect the dynamic correlation of parameters, providing reliable model support for the subsequent determination of the adjustment range.

[0041] Preferred, such as Figure 3 As shown, step S4 includes the following sub-steps: S41, collecting historical oil fume load data, corresponding frequency conversion parameters, and operating condition parameters, dividing the training dataset and validation dataset according to time series to ensure that the dataset covers different operating conditions; S42, inputting the training dataset into the initially constructed prediction model framework, optimizing the internal parameters of the model through gradient descent, and establishing a nonlinear relationship between oil fume load and input parameters; S43, using the validation dataset to test the model prediction performance, identifying operating condition intervals with large prediction deviations, and adjusting the model structure and parameters accordingly; S44, combining the real-time collected operating condition data to perform online correction of the model prediction results, generating an oil fume load frequency conversion prediction model that can reflect real-time changing trends.

[0042] Specifically, steps S41 to S44 of step S4 involve constructing and optimizing the variable frequency prediction model for oil fume load. S41 first collects historical oil fume load data, corresponding variable frequency parameters, and operating condition parameters under different operating conditions over the past six months. This data is then divided into training and validation datasets according to the time series, with a strict 7:3 ratio. This ensures the training dataset covers high, medium, and low load conditions, with each condition's data volume not less than 25% of the total training data. The validation dataset includes typical fluctuation range data for each condition, ensuring comprehensive model validation. S42 inputs the training dataset into the initially constructed prediction model framework and iteratively optimizes the model's internal parameters using gradient descent. The iteration step size is set to 0.01, and the number of iterations is no less than 500. By minimizing the error between the predicted and actual values, a nonlinear relationship between the oil fume load and the input parameters is established, ensuring the model's fit on the training dataset reaches above 0.9. S43 utilizes a validation dataset to test the model's predictive performance, calculating prediction deviations for different load ranges. For operating conditions with deviations exceeding 12%, 200 sets of specialized operating condition data are added for model reinforcement training. The model's feature extraction dimensions and time step parameters are adjusted to optimize its adaptability to complex and fluctuating operating conditions. S44 establishes a real-time correction mechanism, collecting current operating condition data every 10 minutes, comparing it with the model's prediction results, calculating the deviation value, and substituting it into a preset correction formula to adjust the model output. This generates a variable frequency prediction model for fume load that can reflect load change trends in real time, ensuring that the output load dynamic change sequence accurately adapts to actual operating conditions.

[0043] Preferred, such as Figure 4 As shown, step S5 includes the following sub-steps: S51, determining energy efficiency, frequency conversion stability, purification effect, and equipment loss as optimization objectives, establishing a multi-objective optimization function, and clarifying the constraints and boundary ranges of each objective; S52, using frequency conversion adjustment amplitude, frequency adjustment step size, and power allocation ratio as optimization variables, constructing a variable search space, and excluding variable combinations that do not meet the equipment operation requirements; S53, using an iterative optimization strategy to evaluate each variable combination in the search space, calculating the optimization objective value corresponding to each combination, and selecting the set of non-dominated solutions; S54, based on the parameter coordination requirements of the multi-variable collaborative control optimization platform, performing a second screening of the set of non-dominated solutions, determining the optimal variable combination, and forming an energy-saving frequency conversion robust optimization algorithm.

[0044] Specifically, steps S51 to S54 of step S5 involve constructing and implementing an energy-saving variable frequency robust optimization algorithm. S51 clarifies the four optimization objectives of the algorithm: energy efficiency, variable frequency stability, purification effect, and equipment loss. Specific constraints are set for each objective: energy efficiency must be at least 15% higher than traditional algorithms; variable frequency fluctuations must not exceed 2Hz per second; the pollutant removal rate corresponding to the purification effect must be no less than 90%; and the equipment loss rate must be controlled within 5%. The boundary ranges of each objective are also defined to ensure that the optimization process does not exceed the equipment's safe operating threshold. S52 uses the variable frequency adjustment amplitude, frequency adjustment step size, and power allocation ratio as core optimization variables to construct a three-dimensional variable search space. The search range for the variable frequency adjustment amplitude is set to 50% to 150% of the rated amplitude; the frequency adjustment step size is 0.5Hz to 3Hz; and the power allocation ratio is defined as 3:7 to 7:3 according to the operating requirements of the variable frequency module and the purification module. At the same time, variable combinations that do not meet the equipment's operating requirements are excluded, reducing the size of the search space. S53 employs a hierarchical iterative optimization strategy. First, it divides the system into layers based on power allocation ratios. Within each layer, it evaluates the combination of variable frequency control amplitude and frequency adjustment step size, performing at least 20 evaluations per layer. It calculates the comprehensive score for each variable combination corresponding to the four optimization objectives, using a weighted summation method to determine the score. Weights are dynamically allocated according to the operating condition type, ultimately selecting the set of non-dominated solutions. S54, based on the parameter coordination requirements of the multi-variable collaborative control optimization platform, performs a second screening of the non-dominated solution set, eliminating solutions that conflict with the platform's parameter interaction requirements. It then determines the unique optimal variable combination, solidifying it as the core parameters of the energy-saving variable frequency robust optimization algorithm. This ensures that the optimal control parameter set output by the algorithm satisfies both the multi-objective optimization requirements and adapts to the platform's collaborative control mechanism.

[0045] like Figure 5As shown, an energy-saving variable frequency coordinated control method for oil fume purification equipment is implemented through different units, including: a multi-dimensional sensor data acquisition unit, a multi-source data filtering and processing unit, a variable frequency and purification coordinated temperature control model construction unit, an oil fume load variable frequency prediction model derivation unit, an energy-saving variable frequency robust optimization algorithm calculation unit, and a multi-variable coordinated control execution unit. The multi-dimensional sensor data acquisition unit and the multi-source data filtering and processing unit are unidirectionally connected to collect multi-source operating data such as oil fume concentration and airflow rate and transmit them to the filtering and processing unit. The multi-source data filtering and processing unit is connected to the variable frequency and purification coordinated temperature control model construction unit and the oil fume load variable frequency prediction model derivation unit. The model unit is bidirectionally connected, filtering the collected data and providing it to two model units, while simultaneously receiving parameter adjustment requests from the models. Both model units are unidirectionally connected to the energy-saving variable frequency robust optimization algorithm operation unit, outputting model results for the algorithm operation unit to use. The energy-saving variable frequency robust optimization algorithm operation unit is unidirectionally connected to the multivariable collaborative control execution unit, transmitting the optimized control parameters to the execution unit. The multivariable collaborative control execution unit is bidirectionally connected to the multi-dimensional sensor data acquisition unit, executing control commands and feeding back operating data to form closed-loop control. Each unit interacts with parameters through a data bus to ensure the continuity and accuracy of collaborative control.

[0046] The formula in this invention incorporates different scalar and vector parameters into the same calculation system, achieving adaptation through parameter standardization, dimensional unification transformation, and quantitative modeling of coupling relationships. First, for scalar parameters such as fume concentration, equipment operating power, and filter clogging degree, and vector parameters such as airflow rate and frequency converter adjustment amplitude, dimensionless processing is used to eliminate dimensional differences, mapping all parameters to the 0-1 range to ensure consistency in numerical range. Second, based on the actual needs of multi-variable collaborative control, the coupling relationships of different types of parameters are quantified. For example, airflow rate, as a vector, directly affects the diffusion efficiency of fume concentration (scalar), and changes in frequency converter adjustment amplitude (vector) drive dynamic adjustments to equipment operating power (scalar). The formula uses weighting coefficients... By assigning reasonable contribution values ​​to different types of parameters, the directional and magnitude effects of vectors and the numerical effects of scalars can be accurately reflected. Finally, the correlation strength of parameters is calibrated using multi-condition test data. For example, the magnitude of the airflow rate vector is multiplied by the scalar of oil fume concentration to reflect the synergistic effect of the two. The nonlinear correlation between vectors and scalars is adapted by using logarithmic, exponential and other function forms to ensure that different types of parameters form a logically self-consistent calculation chain in the formula. This preserves the essential characteristics of each parameter and realizes the fusion calculation of multi-dimensional information, providing a comprehensive and accurate quantitative basis for collaborative control.

[0047] An energy-saving variable frequency collaborative control method for oil fume purification equipment integrates dynamic monitoring of oxygen flux, multi-dimensional parameter acquisition of the combustion process, and intelligent analysis models through a multi-variable collaborative optimization framework, forming a closed-loop analysis system for the entire process. This significantly improves the accuracy and adaptability of coal combustion process analysis. Its core advantage lies in fully exploring the deep correlation between oxygen flux and judgment indicators such as combustion efficiency, pollutant generation, and furnace temperature distribution. Multi-parameter collaborative modeling breaks through the limitations of single-variable analysis, enabling precise capture of the influence of oxygen flux under different coal types and combustion loads. Simultaneously, intelligent optimization algorithms dynamically adjust analysis parameters to adapt to complex and changing combustion conditions, providing reliable data support for the precise control of the combustion system.

[0048] This method addresses the issue that traditional analytical methods do not fully consider the dynamic changes in oxygen flux and the synergistic correlation of multiple decision indicators. By constructing a multi-dimensional parameter acquisition and collaborative analysis mechanism, it comprehensively incorporates oxygen flux, coal quality characteristics, combustion condition parameters, and other factors into the analysis system, establishing dynamic mapping relationships between various factors and realizing the transformation from single-variable analysis to multi-variable collaborative analysis. Furthermore, to address the shortcomings of traditional methods in terms of poor model adaptability and inability to cope with complex combustion conditions, this method introduces an intelligent optimization algorithm to optimize model parameters, enhancing the model's adaptability to different oxygen flux ranges and combustion states. This effectively reduces the impact of operating condition fluctuations on the analysis results, improves the stability and reliability of the analysis results, and provides strong support for the efficient and low-consumption operation of the combustion system.

[0049] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," "link," and "fix" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0050] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various equivalent changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for energy-saving variable frequency coordinated control of an oil fume purification device, characterized in that, Includes the following steps: S1 collects relevant sensor data such as fume concentration, airflow rate, equipment operating power, purification chamber temperature, inverter module output frequency, and filter clogging degree through the multi-dimensional sensing module mounted on the fume purification equipment, and establishes a multi-source data acquisition matrix. S2 inputs the collected multi-source data into the multi-variable collaborative control optimization platform, and uses the platform's built-in data filtering mechanism to remove abnormal data and extract calibration characteristic parameters that are directly related to energy-saving frequency conversion and purification efficiency. S3. Based on the calibration characteristic parameters, a variable frequency and purification coordinated temperature control model is constructed. This model maps the dynamic correlation between temperature changes and variable frequency parameters to determine the initial variable frequency adjustment range. S4 uses a variable frequency prediction model for fume load to deduce the trend of fume load change under different working conditions, outputs a dynamic load change sequence, and corrects the initial variable frequency adjustment range. S5 utilizes an energy-saving variable frequency robust optimization algorithm to perform multi-objective optimization on the corrected variable frequency regulation range, generating an optimal variable frequency control parameter set that adapts to real-time operating conditions; S6 transmits the optimal set of variable frequency control parameters to the actuator of the fume purification equipment to enable the coordinated operation of the variable frequency module and the purification module. The operating data is fed back in real time through the multi-variable collaborative control optimization platform, and the control parameters are dynamically adjusted.

2. The energy-saving variable frequency coordinated control method for oil fume purification equipment according to claim 1, characterized in that, The expression for the frequency conversion and purification synergistic temperature control model is: , in, To achieve optimal temperature control within the purification chamber. This is the frequency weighting coefficient for the frequency conversion. The frequency output of the inverter module. The concentration of cooking fumes, For operating power weighting coefficient, For equipment operating power, The degree of filter clogging This is the airflow rate weighting coefficient. For airflow rate, This is the cavity temperature weighting coefficient. To purify the current temperature of the cavity, For adjustment coefficients, This is the amplitude of frequency conversion adjustment.

3. The energy-saving variable frequency coordinated control method for oil fume purification equipment according to claim 1, characterized in that, The expression for the frequency conversion prediction model of the fume load is: , in, To predict the oil fume load, For frequency load factor, The frequency output of the inverter module. The concentration of cooking fumes, This is the airflow blockage coefficient. For airflow rate, The degree of filter clogging For power temperature coefficient, For equipment operating power, To purify the current temperature of the cavity, This is the temperature fluctuation coefficient. To adjust the load factor, This is the amplitude of frequency conversion adjustment.

4. The energy-saving variable frequency coordinated control method for oil fume purification equipment according to claim 1, characterized in that, The expression for the energy-saving variable frequency robust optimization algorithm is: , in, To achieve the optimal frequency conversion, As the load deviation weight, To predict the oil fume load, For the rated oil fume load, For power frequency weighting, For equipment operating power, The frequency output of the inverter module. To block the adjustment weight, The degree of filter clogging For frequency conversion adjustment amplitude, As a weight for temperature deviation, To purify the current temperature of the cavity, To achieve optimal temperature control for the purification chamber.

5. The energy-saving variable frequency coordinated control method for oil fume purification equipment according to claim 1, characterized in that, The parameter coordination model expression of the multivariable collaborative control optimization platform is as follows: , in, For multivariate coordination coefficients, The temperature coherence coefficient, To achieve optimal temperature control within the purification chamber. This is the load coordination factor. To predict the oil fume load, For frequency conversion coordination coefficient, To achieve the optimal frequency conversion, The concentration synergy coefficient, The concentration of cooking fumes, The airflow coordination coefficient, For airflow rate, For blocking coordination coefficient, The degree of filter clogging The power attenuation coefficient, This refers to the operating power of the equipment.

6. The energy-saving variable frequency coordinated control method for oil fume purification equipment according to claim 1, characterized in that, The comprehensive parameter model expression for the energy-saving variable frequency coordinated control of the fume purification equipment is as follows: , in, For the comprehensive indicators of coordinated control, The variable frequency temperature coordination coefficient, To achieve the optimal frequency conversion, To achieve optimal temperature control within the purification chamber. The load congestion coordination coefficient, To predict the oil fume load, The degree of filter clogging The power concentration synergy coefficient, For equipment operating power, The concentration of cooking fumes, For airflow rate, The coefficients represent multivariate coordination coefficients.

7. The energy-saving variable frequency coordinated control method for an oil fume purification equipment according to claim 1, characterized in that, S3 includes the following steps: S31, select the purification chamber temperature, the output frequency of the frequency converter module, the oil fume concentration and the equipment operating power as calibration input parameters, construct a four-dimensional parameter correlation matrix, and clarify the priority of each parameter in temperature control; S32, based on the parameter correlation matrix, uses a nonlinear mapping method to establish an initial correlation model between temperature and frequency conversion parameters, and calibrates the initial parameters of the model through sensor data to reduce parameter coupling interference; S33 introduces filter blockage degree and airflow rate as correction factors to dynamically adjust the initial correlation model and optimize the model's adaptability to different blockage conditions. S34 verifies the model output accuracy through a multivariable collaborative control optimization platform, and fine-tunes the model weight coefficients based on the verification results to form the final frequency conversion and purification collaborative temperature control model.

8. The energy-saving variable frequency coordinated control method for oil fume purification equipment according to claim 1, characterized in that, S4 includes the following sub-steps: S41: Collect historical oil fume load data, corresponding frequency conversion parameters and operating condition parameters, and divide the training dataset and validation dataset according to the time series to ensure that the dataset covers different operating conditions. S42, input the training dataset into the initially constructed prediction model framework, optimize the internal parameters of the model through gradient descent, and establish a nonlinear relationship between the oil fume load and the input parameters; S43, use the validation dataset to test the model's prediction performance, identify the operating conditions with large prediction deviations, and adjust the model structure and parameters accordingly; S44 combines real-time collected operating data to perform online corrections on the model prediction results, generating a frequency conversion prediction model for oil fume load that can reflect real-time changing trends.

9. The energy-saving variable frequency coordinated control method for an oil fume purification equipment according to claim 1, characterized in that, S5 includes the following steps: S51, energy efficiency, frequency conversion stability, purification effect and equipment loss are determined as optimization objectives, a multi-objective optimization function is established, and the constraints and boundary ranges of each objective are clarified; S52 uses the frequency conversion regulation amplitude, frequency regulation step size, and power distribution ratio as optimization variables to construct a variable search space and eliminate variable combinations that do not meet the equipment operation requirements. S53, adopts an iterative optimization strategy to evaluate each combination of variables in the search space, calculates the optimization target value corresponding to each combination, and selects the set of non-dominated solutions; S54, based on the parameter coordination requirements of the multivariable collaborative control optimization platform, performs secondary screening on the non-dominated solution set to determine the optimal variable combination and form an energy-saving variable frequency robust optimization algorithm.

10. A method for energy-saving variable frequency coordinated control of an oil fume purification device according to any one of claims 1-9, characterized in that, This method is implemented through different units, including: a multi-dimensional sensor data acquisition unit, a multi-source data filtering and processing unit, a variable frequency and purification coordinated temperature control model construction unit, an oil fume load variable frequency prediction model deduction unit, an energy-saving variable frequency robust optimization algorithm calculation unit, and a multi-variable coordinated control execution unit; The multi-dimensional sensor data acquisition unit is unidirectionally connected to the multi-source data filtering and processing unit to collect multi-source operating data such as fume concentration and airflow rate and transmit them to the filtering and processing unit. The multi-source data filtering and processing unit is bidirectionally connected to the frequency conversion and purification collaborative temperature control model construction unit and the fume load frequency conversion prediction model derivation unit. After filtering the collected data, it provides it to the two model units and receives parameter adjustment requests from the models. Both model units are unidirectionally connected to the energy-saving frequency conversion robust optimization algorithm operation unit and output model results for the algorithm operation unit to call. The energy-saving frequency conversion robust optimization algorithm operation unit is unidirectionally connected to the multi-variable collaborative control execution unit and transmits the optimized control parameters to the execution unit. The multi-variable collaborative control execution unit is bidirectionally connected to the multi-dimensional sensor data acquisition unit, executes control commands and feeds back operating data to form a closed-loop control. Each unit interacts with parameters through a data bus to ensure the continuity and accuracy of collaborative control.