Condensing unit control system

Abstract

The invention provides a condensing unit control system, which relates to the technical field of refrigeration control and comprises a multi-mode sensing module, an edge computing unit, a cooperative control center, an execution module and a man-machine interaction module. The multi-mode sensing module integrates intrusive and non-intrusive sensing, and synchronously collects core component operation parameters, environment parameters, load parameters and component surface temperature / vibration characteristics. The edge calculation unit processes multi-source data and realizes early fault early warning; the cooperative control center generates a global optimal control strategy through a basic energy efficiency library of a pre-stored multi-dimensional reference model and a dynamic correction module; the execution module adjusts the core component according to a hierarchical response mechanism. The problems that a traditional system is single in sensing, lagged in response, fixed in energy efficiency model and free of priority in adjustment are solved, the unit reliability, the full-working-condition energy efficiency and the complex working condition adaptability are improved, and the operation and maintenance cost is reduced.

Description

Technical Field

[0001] This invention relates to the field of refrigeration control technology, and in particular to a condensing unit control system. Background Technology

[0002] Condensing chillers, as core equipment in refrigeration systems, are widely used in cold chain logistics, data centers, and commercial refrigeration. Their operational efficiency and stability directly determine the industry's energy consumption costs and service reliability. Currently, with the advancement of "dual-carbon" policies and the diversification of application scenarios, condensing chillers need to cope with the challenges of complex operating conditions—increased fluctuations in ambient temperature and humidity, and pulsed or periodic changes in terminal cooling demand, making it difficult for traditional control systems to adapt to the high-efficiency operation requirements.

[0003] Existing condensing unit control systems have significant shortcomings: they rely on a single sensing method, either using invasive sensors which lead to complex installation and maintenance, or using only non-invasive sensors which lack core operating parameters; data processing largely depends on the cloud, resulting in response lag, and the energy efficiency model is fixed and cannot adapt to different refrigerant types and environmental conditions; there is no priority division for execution regulation, and when the load changes or the environment changes, local optimization is prone to occur, resulting in a decrease in global energy efficiency, or even causing component failure. Therefore, there is an urgent need for a new control system that integrates multimodal sensing, dynamic model optimization, and hierarchical response. Summary of the Invention

[0004] This invention provides a collaborative optimization control system for condensing units, which aims to solve the problems of traditional condensing unit control systems, such as the single sensing method leading to incomplete acquisition of operating parameters and component status, reliance on cloud for data processing causing lag in dynamic operating condition response, fixed energy efficiency models that cannot adapt to multiple refrigerant types and complex environmental load scenarios, and lack of priority division in execution regulation leading to low overall energy efficiency and high risk of core component failure.

[0005] To solve the above-mentioned technical problems, the present invention provides a condensing unit control system, comprising: The multimodal sensing module integrates invasive and non-invasive sensing units to synchronously collect operating parameters, environmental parameters, and load demand parameters of the core components of the condensing unit. The non-invasive sensing unit obtains the surface temperature distribution and vibration characteristics of the components through infrared thermal imaging and acoustic sensing. The edge computing unit is communicatively connected to the multimodal perception module. It has a built-in adaptive filtering module and feature association engine, which are used to perform spatiotemporal registration, redundancy removal and dynamic association feature extraction on multi-source heterogeneous parameters, and realize early fault warning based on feature deviation. The collaborative control center, which is communicatively connected to the edge computing unit, includes a basic energy efficiency model library and a dynamic correction module. The basic energy efficiency model library pre-stores benchmark optimization models for different refrigerant types and load scenarios. The dynamic correction module iterates the benchmark models online based on real-time feature data and historical energy efficiency deviations to generate a globally optimal collaborative control strategy. The execution module is communicatively connected to the collaborative control center and is equipped with a hierarchical response mechanism for adjusting the compressor capacity, condenser heat exchange efficiency, and throttling element opening in real time according to the priority of the control strategy. The priority is dynamically determined based on the load mutation rate and component safety threshold.

[0006] Preferably, the invasive sensing unit includes a pressure sensor installed at the compressor inlet and outlet, a temperature sensor for the motor windings, and a flow sensor for the condenser piping; the non-invasive sensing unit includes an infrared array sensor and an acoustic wave collector deployed on the unit casing, which are used to acquire the surface temperature field distribution of components and the acoustic wave characteristics of abnormal vibrations, respectively.

[0007] As a preferred embodiment, the feature association engine of the edge computing unit mines parameter associations in the following manner: Establish a multidimensional correlation matrix of operating parameters, environmental parameters, and energy efficiency indicators; Analysis of the phase difference and amplitude coupling of parameter changes based on sliding time window; Extract the parameter response hysteresis characteristics during load mutations.

[0008] Preferably, the early fault warning includes: By comparing the real-time vibration acoustic wave characteristics with the acoustic wave spectrum template under normal operating conditions, abnormal frequency components can be identified. Analyze the gradient anomalies in the temperature field distribution to determine the local scaling or blockage trend of heat exchange components; Based on the fluctuation variance of pressure parameters, the risk of jamming of throttling elements is predicted.

[0009] As a preferred embodiment, the basic energy efficiency model library is categorized according to the following dimensions: Optimization models are categorized according to refrigerant type into Freon-based, natural refrigerant-based, and mixed refrigerant-based models. Control models are categorized based on load characteristics into steady-state load, pulse load, and periodic load. The models are categorized according to environmental conditions, including high-temperature environments, high-humidity environments, and dusty environments.

[0010] Preferably, the online iteration of the dynamic correction module includes: Based on the environmental deviation between the current environmental parameters and the benchmark model, the environmental impact coefficients in the model are corrected. Based on the deviation between recent measured energy efficiency values ​​and model predictions, the parameter weighting factors are dynamically adjusted. A component aging factor is introduced, which is calculated based on the running time and performance degradation trend.

[0011] Preferably, the hierarchical response mechanism of the execution module includes: Level 1 response: When the load change rate exceeds the preset threshold, the opening degree of the compressor's capacity regulation component and throttling element is adjusted first, and the response delay is controlled within the preset range; Level 2 response: When the change in environmental parameters exceeds the preset value, the fan speed and condenser flow rate of the condenser should be adjusted as the main focus. Level 3 response: When the system energy efficiency ratio is lower than the baseline value, the operating parameter combination of all core components is optimized simultaneously.

[0012] As a preferred embodiment, the present invention also includes a human-computer interaction module for displaying real-time operating parameters, energy efficiency curves and fault warning information, and supports manual adjustment of the weight parameters of the collaborative control strategy.

[0013] Compared with related technologies, the condensing unit control system provided by the present invention has the following beneficial effects: 1. This solution significantly improves the operational reliability and fault early warning capability of condensing units through the deep integration of multimodal sensing and edge computing. The invasive sensing unit collects core parameters such as pressure, temperature, and flow rate in real time, while the non-invasive infrared array and acoustic wave collector simultaneously capture the surface temperature field distribution and vibration characteristics. The edge computing unit realizes real-time correlation mining of multi-source data through algorithms such as covariance matrix analysis and phase difference coupling degree calculation. Through the global sensing and local real-time analysis architecture, it can accurately identify early fault characteristics such as abnormal frequency components, abnormal temperature gradients, and pressure fluctuations, and provide early warning of potential risks such as scaling of heat exchange components and jamming of throttling elements, greatly reducing the probability of sudden shutdown and extending the fault-free operation cycle of the equipment.

[0014] 2. The dynamic energy efficiency optimization system endows the system with strong adaptability to operating conditions and energy-saving benefits. The basic energy efficiency model library is classified in multiple dimensions according to refrigerant type, load characteristics, and environmental conditions. Combined with the dynamic correction module, it realizes the self-iteration of energy efficiency throughout the entire life cycle: the influence coefficient is corrected in real time by environmental parameter deviation, the parameter weight is dynamically adjusted according to the deviation between the measured and predicted energy efficiency values, and the component aging coefficient based on the running time is introduced. By introducing a classification model and combining it with an online correction mechanism, it can accurately adapt to the characteristics of different refrigerants such as Freon / natural working fluid, load changes such as steady state / pulse, and environmental differences such as high temperature / high humidity, ensuring that the system always maintains the optimal energy efficiency ratio under complex operating conditions and reduces ineffective energy consumption loss.

[0015] 3. The hierarchical response mechanism achieves a balance between the precision and economy of system regulation. The first-level response prioritizes the control of the compressor and throttling elements in response to load changes, and calculates the change rate and controls the response delay using the time difference method. The second-level response focuses on the adjustment of the condenser fan and medium flow rate based on changes in environmental parameters, and dynamically matches the heat dissipation capacity based on environmental deviations. The third-level response synchronously optimizes the parameters of all components when energy efficiency is low by relying on the corrected weighting factor. The three-level hierarchical regulation logic of this invention not only ensures rapid response under critical operating conditions, but also avoids component damage caused by over-regulation. By rationally allocating regulation resources, the service life of core components is extended while ensuring refrigeration stability, and long-term operation and maintenance costs are reduced.

[0016] In summary, through a collaborative architecture of full-domain perception, intelligent optimization, and precise execution, the reliability, energy efficiency, and adaptability of condensing units have been comprehensively improved. Multimodal perception and edge computing have constructed a three-dimensional fault defense system, while a dynamic energy efficiency model library and correction mechanism ensure energy-saving advantages under all operating conditions. A graded response mechanism balances the adjustment speed and equipment protection requirements. The synergy of these three elements enables the system to operate stably in complex environments, avoid faults in advance, dynamically adapt to changes in operating conditions, and continuously optimize energy consumption. Attached Figure Description

[0017] Figure 1 This invention provides a schematic diagram of a condensing unit control system. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The singular forms “group,” “class,” and “the” as used in this disclosure and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0020] It should be understood that although the terms first, second, third, etc., may be used in this disclosure to describe various information, such information should not be limited to these terms. These terms are used only to distinguish information of the same type from one another. For example, without departing from the scope of this disclosure, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0021] Please refer to the following: Figure 1 A condensing unit control system, comprising: The multimodal sensing module integrates invasive and non-invasive sensing units to simultaneously collect operating parameters, environmental parameters, and load demand parameters of the core components of the condensing unit. The non-invasive sensing unit obtains the surface temperature distribution and vibration characteristics of the components through infrared thermal imaging and acoustic sensing. The edge computing unit communicates with the multimodal perception module and has a built-in adaptive filtering module and feature association engine. It is used to perform spatiotemporal registration, redundancy removal and dynamic association feature extraction of multi-source heterogeneous parameters, and realize early fault warning based on feature deviation. The collaborative control center communicates with the edge computing unit and includes a basic energy efficiency model library and a dynamic correction module. The basic energy efficiency model library pre-stores benchmark optimization models for different refrigerant types and load scenarios. The dynamic correction module iterates the benchmark models online based on real-time feature data and historical energy efficiency deviations to generate the globally optimal collaborative control strategy. The execution module communicates with the collaborative control center and is equipped with a hierarchical response mechanism. It is used to adjust the compressor capacity, condenser heat exchange efficiency and throttling element opening in real time according to the priority of the control strategy. The priority is dynamically determined based on the load change rate and component safety threshold.

[0022] Specifically, the invasive sensing unit includes pressure sensors located at the compressor inlet and outlet, temperature sensors on the motor windings, and flow sensors on the condenser piping. Among them, the compressor discharge pressure collected by the pressure sensor With inhalation pressure Used to calculate compression ratio Combined with the winding temperature collected by the motor winding temperature sensor Assess the compressor's operating status; the flow sensor in the condenser piping collects the heat exchange medium flow rate Q to monitor the medium flow efficiency; The non-invasive sensing unit includes an infrared array sensor and an acoustic wave collector deployed on the unit's outer casing, which are used to acquire the surface temperature field distribution of components and the acoustic wave characteristics of abnormal vibrations, respectively. Among them, the infrared array sensor detects the original temperature of N1 sensing points. Where i is the sensor point number; the surface temperature field of the component is reconstructed using a bicubic interpolation algorithm, i.e., through the formula ,in The interpolation coefficients are obtained by least squares fitting, and the fitting process minimizes the error function. s1 and s2 are the degrees of the interpolation polynomials, ranging from 0 to 3, obtained through least squares fitting. The reconstructed temperature value at coordinates (x, y) on the component surface, where (x, y) represents the two-dimensional coordinates of the component surface, enables the acquisition of the continuous temperature field distribution on the component surface; The acoustic wave acquisition device collects the vibration acoustic wave signal s(t), and converts the time-domain signal into a frequency-domain signal using a fast Fourier transform, as shown in the formula: Where N is the number of sampling points and f is the frequency. This represents the Fast Fourier Transform operator, where u represents the sampling point number of the time-domain signal, and j is the imaginary unit. The spectral deviation is then calculated. ,in This is represented as the reference acoustic spectrum under normal operating conditions. This indicates the actual sound wave signal at a frequency point. The spectral amplitude at point M, where M is the number of sampling points within the frequency range of interest. The k-th frequency point is used to identify the characteristics of abnormal vibration sound waves.

[0023] Specifically, the feature association engine of the edge computing unit mines parameter associations in the following ways: Establish a multidimensional correlation matrix of operating parameters, environmental parameters, and energy efficiency indicators: Let the set of runtime parameters be denoted as The set of environmental parameters is denoted as Energy efficiency label Construct a joint vector containing all parameters. , This indicates a transpose operation, which quantifies the linear correlation between parameters by calculating the covariance matrix. The elements are: ,in Let be the mean of the p-th parameter. It represents the mathematical expectation; the larger the absolute value of the matrix element, the stronger the linear correlation between the corresponding parameter pairs. Analysis of phase difference and amplitude coupling of parameter changes based on sliding time window: The duration is The sliding step size is A sliding window is used to collect discrete sampling sequences within the window for any two parameter sequences A(t) (such as compressor power) and B(t) (such as condenser fan speed). , where k is the number of sampling points within the window; The frequency domain sequence is obtained by performing a Fast Fourier Transform on the sampled sequence. Calculate the cross power spectrum In the formula (represented as complex conjugate), then the phase difference at frequency f is: ; The degree of amplitude correlation is measured by the coherence coefficient: ,in For self-power spectrum, The closer the value is to 1, the stronger the coupling. Extracting parameter response hysteresis characteristics during load abrupt changes: Let the load parameters be The parameters of the unit to be analyzed are: The moment of load mutation is ; Calculate the cross-correlation function: ; in Here, represents the mean values ​​of load and unit parameters, respectively, and T is the analysis duration; the lag corresponding to the maximum correlation value is determined through optimization. This refers to the response lag time of the parameter to sudden load changes.

[0024] Specifically, early fault warnings include: By comparing the real-time vibration acoustic wave characteristics with the acoustic wave spectrum template under normal operating conditions, abnormal frequency components can be identified. Based on the real-time frequency domain acoustic signal S(f) acquired by the acoustic wave acquisition device, and the pre-stored reference acoustic wave spectrum under normal operating conditions. Calculate the k-th frequency point Spectral deviation rate at Set an abnormal frequency threshold. If there exists at least one frequency point that satisfies And the duration of this frequency point Exceeding the preset time If so, the frequency point is determined to be an abnormal frequency component, triggering a vibration anomaly warning. Analyze the gradient anomalies in the temperature field distribution to determine the tendency for localized scaling or blockage in heat exchange components: Based on the component surface temperature field T(x, y) reconstructed from an infrared array sensor, the gradient vector of the temperature field at the two-dimensional coordinates (x, y) is calculated. ,in Let be the partial derivative of temperature along the x-direction. The partial derivatives of temperature along the y-direction are calculated using the central difference method: ,in For a small increment of coordinates; Calculate the magnitude of the gradient vector And pre-store the maximum value of the temperature field gradient modulus under normal operating conditions. ; If there exists a local region that satisfies Where K is the gradient anomaly amplification factor, K>1, then the temperature field gradient in the region is determined to be abnormal, and the corresponding heat exchange components have a tendency to local scaling or blockage, triggering a heat exchange anomaly warning. Based on the variance of pressure parameter fluctuations, predict the jamming risk of throttling elements: Compressor exhaust pressure collected by invasive sensing unit With inhalation pressure Select duration as A sliding time window is used to collect pressure sampling sequences within the window. q is the number of sampling points within the window; Calculate the variance of the fluctuation of this pressure series. ,in The average pressure within the window; a pre-stored upper limit threshold for pressure fluctuation variance under normal operating conditions. If all n1 consecutive sliding windows satisfy If n1 represents the number of consecutive abnormal determinations, then the throttling element is determined to be at risk of jamming, triggering a throttling abnormality warning.

[0025] Specifically, the basic energy efficiency model library is categorized according to the following dimensions: Optimization models are categorized by refrigerant type into Freon-based, natural refrigerant, and mixed refrigerant-based models: Define the refrigerant type identification parameter RefType, with a value of Indicates Freon-like substances, Indicates natural working fluids, This indicates a mixed working fluid type, with pre-stored thermodynamic characteristic coefficients for different types of refrigerants. , Corresponding to Freon-type, Corresponding to natural working fluids Corresponding to mixed working fluids; Based on the unit's preset refrigerant type parameter RefType, the corresponding benchmark energy efficiency model is called. The model satisfies ,in This is a general benchmark energy efficiency value. For the set of running parameters The function associated with the set of environment parameters Env; Control models are categorized based on load characteristics into steady-state load, pulse load, and periodic load. Based on load parameters Calculate the load fluctuation coefficient ,in The load average over a preset time period is calculated using an autocorrelation function. Detect load periodicity; like , To determine the steady-state load threshold, the steady-state load control model is invoked. ;like And there is no significant periodicity (if) This indicates that there is no significant periodicity. (For periodic determination threshold), call the pulse load control model. If significant periodic characteristics exist (if If this indicates the presence of significant periodic characteristics, then the periodic load control model should be invoked. ; Based on environmental conditions, the corresponding adaptation models are divided into high-temperature environment, high-humidity environment, and dusty environment: Extracting ambient temperature from environmental parameters Ambient humidity Dust concentration Set environmental classification thresholds: ( When the temperature threshold is reached, the environment is considered a high-temperature environment. ( When the humidity threshold is reached, the environment is considered high humidity. ( When the threshold for determining dust is reached, the environment is classified as a dusty environment. Corresponding to the call of adaptation models for different environments Environmental impact factors are introduced into the model. ( Corresponding to high temperature environment, Corresponding to high humidity environments (For sandstorm environments), the adapted model meets the requirements. .

[0026] It should be noted that the thermodynamic characteristic coefficients of the various models in the basic energy efficiency model library, such as the refrigerant type optimization model, are taken from publicly available thermodynamic data, and the correlation function is obtained through regression analysis of multi-condition energy efficiency experiments of the corresponding refrigerant units. The load characteristic control model uses mature signal processing technology for load feature identification, and the control strategy is tuned by referring to existing refrigeration variable load control algorithms and combining them with unit operating data. The environmental impact law of the environmental condition adaptation model is derived from existing research in refrigeration and environmental engineering, and the environmental impact factor is obtained through fitting experiments of energy efficiency testing in an environmental simulation chamber. Overall, it is a classification, integration, and adaptation of existing technologies, industry experimental data, and theoretical methods.

[0027] Specifically, the online iteration of the dynamic correction module includes: Based on the environmental deviation between the current environmental parameters and the baseline model, the environmental impact coefficients in the model are corrected: Extract ambient temperature from current environmental parameters. Ambient humidity Call the baseline ambient temperature corresponding to the baseline model. Reference ambient humidity (Pre-stored in the basic energy efficiency model library); Calculate environmental deviations: ; Environmental impact coefficient (The environmental adaptation model, taken from the basic energy efficiency model library, is corrected using the following formula:) ,in , These are the correction coefficients for temperature deviation and humidity deviation (pre-stored in the dynamic correction module). Based on the deviation between recent measured energy efficiency values ​​and model predictions, the parameter weighting factors are dynamically adjusted: Obtain measured energy efficiency values ​​within a recent preset time period. and the model-predicted energy efficiency values ​​for the corresponding time periods ; Calculate energy efficiency deviation: ; For the set of runtime parameters Corresponding weighting factors (Pre-stored in the baseline model), the adjustment formula is as follows ,in This is the weight adjustment coefficient (pre-stored in the dynamic correction module). Let be the normalized value of the i-th running parameter, where , These are the minimum and maximum rated values ​​for this parameter, respectively. Introducing a component aging factor, which is calculated based on the operating time and performance degradation trend: Obtain the cumulative operating time of the core components of the condensing unit (compressor, condenser, throttling element). (Recorded by the system timing module); Pre-stored performance degradation coefficients of each core component. (Determined based on the lifespan degradation curve provided by the component manufacturer); Calculate the aging coefficient of components (when , (for the rated life of the component), if Then fixed ,in (This refers to the aging factor threshold corresponding to the lowest performance of the component). The aging factor is incorporated into the modified model to ensure that the final energy efficiency model meets the requirements. ,in The energy efficiency value is the model value after correction for environmental impact factor.

[0028] Specifically, the hierarchical response mechanism of the execution module includes: Level 1 Response: When the load change rate exceeds a preset threshold, the compressor's capacity regulation component and throttling element opening are adjusted first, and the response delay is controlled within a preset range. Based on the real-time load parameter Load(t) acquired by the multimodal sensing module, the load mutation rate is calculated using the time difference method. Set load mutation threshold ,when At that time, a level 1 response is triggered; Prioritize adjusting the compressor capacity adjustment component; capacity adjustment amount ,in The compressor capacity proportional adjustment coefficient is pre-stored in the execution module; it synchronously adjusts the opening degree of the throttling element, and the adjustment amount is... ,in The coupling coefficient between the throttling element and the capacity is set based on historical adjustment data; By localizing processing through edge computing units, the response latency from triggering to execution of adjustment commands is ensured. ,in A preset delay threshold is set for the Level 1 response; Level 2 Response: When environmental parameters change beyond preset values, the condenser fan speed and condensate flow rate should be adjusted accordingly. Extract the current environmental parameters collected by the multimodal sensing module, including at least the ambient temperature. Ambient humidity Calculate and base environmental parameters in the basic energy efficiency model library ( ) range of change , Set environmental change thresholds ,when or At that time, a level 2 response is triggered; The key adjustment is the condenser fan speed, and the amount of speed adjustment... , This is the fan speed adjustment coefficient. Humidity influence weight; synchronously adjust the condensate flow rate, and adjust the flow rate accordingly. ,in This is the matching coefficient between flow rate and rotational speed; Level 3 Response: When the system energy efficiency ratio is lower than the baseline value, simultaneously optimize the combination of operating parameters for all core components. Acquire real-time measured values ​​of system energy efficiency collected by the multimodal sensing module. Compared with the benchmark energy efficiency ratio in the basic energy efficiency model library Compare and calculate energy efficiency deviation Set energy efficiency deviation threshold ,when At that time, a level three response is triggered; Parameter weighting factors based on the output of the dynamic correction module Simultaneously optimize the operating parameters of all core components: compressor frequency adjustment. Condenser fan speed adjustment Throttling element opening adjustment amount ;in These are the energy efficiency adjustment coefficients for each component. This is the corrected weighting factor for the corresponding parameters.

[0029] Specifically, the present invention also includes a human-computer interaction module for displaying real-time operating parameters, energy efficiency curves and fault warning information, and supports manual adjustment of the weight parameters of the collaborative control strategy.

[0030] All formulas in this solution use dimensionless numerical calculations. Dimensionlessness can be achieved through standardization and other means, which will not be elaborated here. The formulas are obtained by software simulation based on a large amount of condensing unit operating data, which can closely reflect the actual operating conditions. The preset parameters in the formulas are set by those skilled in the art in combination with the actual application scenarios of the units.

[0031] The above embodiments can be implemented by software, hardware, firmware, or any combination thereof. If implemented by software, it can be embodied as a computer program product, the computer instructions contained therein can be stored in a computer-readable storage medium or transmitted between media, and after being loaded and executed, it can realize the control flow and functions described in this solution.

[0032] It should be noted that each program number does not represent the order of execution; the execution order is determined by the functional logic. The method of functional implementation (hardware or a combination of hardware and software) depends on the specific application and design constraints. In the device embodiment, the unit division can be a logical functional division, which can be adjusted as needed. Functional units can be integrated or exist independently. If the function is implemented in the form of a software unit and used independently, it can be stored in a computer-readable storage medium containing program code, and the corresponding control steps can be executed by the device through instructions.

[0033] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the following claims.

[0034] It should be understood that the present invention is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A condensing unit control system, characterized in that, include: The multimodal sensing module integrates invasive and non-invasive sensing units to synchronously collect operating parameters, environmental parameters, and load demand parameters of the core components of the condensing unit. The non-invasive sensing unit obtains the surface temperature distribution and vibration characteristics of the components through infrared thermal imaging and acoustic sensing. The edge computing unit is communicatively connected to the multimodal perception module. It has a built-in adaptive filtering module and feature association engine, which are used to perform spatiotemporal registration, redundancy removal and dynamic association feature extraction on multi-source heterogeneous parameters, and realize early fault warning based on feature deviation. The collaborative control center, which is communicatively connected to the edge computing unit, includes a basic energy efficiency model library and a dynamic correction module. The basic energy efficiency model library pre-stores benchmark optimization models for different refrigerant types and load scenarios. The dynamic correction module iterates the benchmark models online based on real-time feature data and historical energy efficiency deviations to generate a globally optimal collaborative control strategy. The execution module is communicatively connected to the collaborative control center and is equipped with a hierarchical response mechanism for adjusting the compressor capacity, condenser heat exchange efficiency, and throttling element opening in real time according to the priority of the control strategy. The priority is dynamically determined based on the load mutation rate and component safety threshold.

2. The condensing unit control system according to claim 1, characterized in that, The invasive sensing unit includes a pressure sensor installed at the compressor inlet and outlet, a temperature sensor for the motor windings, and a flow sensor for the condenser piping; the non-invasive sensing unit includes an infrared array sensor and an acoustic wave collector deployed on the unit casing, which are used to acquire the surface temperature field distribution of components and the acoustic wave characteristics of abnormal vibrations, respectively.

3. A condensing unit control system according to claim 1, characterized in that, The feature association engine of the edge computing unit mines parameter associations in the following ways: Establish a multidimensional correlation matrix of operating parameters, environmental parameters, and energy efficiency indicators; Analysis of the phase difference and amplitude coupling of parameter changes based on sliding time window; Extract the parameter response hysteresis characteristics during load mutations.

4. A condensing unit control system according to claim 1, characterized in that, The early fault warning includes: By comparing the real-time vibration acoustic wave characteristics with the acoustic wave spectrum template under normal operating conditions, abnormal frequency components can be identified. Analyze the gradient anomalies in the temperature field distribution to determine the local scaling or blockage trend of heat exchange components; Based on the fluctuation variance of pressure parameters, the risk of jamming of throttling elements is predicted.

5. A condensing unit control system according to claim 1, characterized in that, The basic energy efficiency model library is categorized according to the following dimensions: Optimization models are categorized according to refrigerant type into Freon-based, natural refrigerant-based, and mixed refrigerant-based models. Control models are categorized based on load characteristics into steady-state load, pulse load, and periodic load. The models are categorized according to environmental conditions, including high-temperature environments, high-humidity environments, and dusty environments.

6. A condensing unit control system according to claim 1, characterized in that, The online iteration of the dynamic correction module includes: Based on the environmental deviation between the current environmental parameters and the benchmark model, the environmental impact coefficients in the model are corrected. Based on the deviation between recent measured energy efficiency values ​​and model predictions, the parameter weighting factors are dynamically adjusted. A component aging factor is introduced, which is calculated based on the running time and performance degradation trend.

7. A condensing unit control system according to claim 1, characterized in that, The hierarchical response mechanism of the execution module includes: Level 1 response: When the load change rate exceeds the preset threshold, the opening degree of the compressor's capacity regulation component and throttling element is adjusted first, and the response delay is controlled within the preset range; Level 2 response: When the change in environmental parameters exceeds the preset value, the fan speed and condenser flow rate of the condenser should be adjusted as the main focus. Level 3 response: When the system energy efficiency ratio is lower than the baseline value, the operating parameter combination of all core components is optimized simultaneously.

8. A condensing unit control system according to claim 1, characterized in that, It also includes a human-machine interaction module, which displays real-time operating parameters, energy efficiency curves and fault warning information, and supports manual adjustment of the weight parameters of the collaborative control strategy.

Images