An optimization system and method for commercial inspection of air conditioning refrigeration systems

By constructing a dynamically calibrated digital simulation model and combining gray box modeling and data-driven methods, the problem of model deviation in the commercial inspection of air conditioning refrigeration systems was solved, achieving highly accurate commercial inspection judgment and deterioration trend identification.

CN123069716APending Publication Date: 2026-11-10FOSHAN SHUNDE WEILONG HEATING EQUIP CO LTD
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202611559096.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-09
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

The existing air conditioning refrigeration system inspection and evaluation optimization system cannot dynamically track the actual operating status of the unit under test, resulting in a deviation between the model output and the actual system, making it difficult to give accurate inspection conclusions.

Method used

A digital simulation model is constructed to simulate commodity inspection scenarios by integrating a benchmark architecture with multi-dimensional performance parameters. A two-dimensional hierarchical judgment method of performance analysis and trend analysis is adopted, combined with gray box modeling and data-driven methods, to dynamically calibrate the model to reflect individual differences and working condition fluctuations.

Benefits of technology

It improves the accuracy and relevance of commodity inspection judgments, avoids misjudgments, can identify potential deterioration trends, and achieves an optimized effect from passive judgment to proactive prevention.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123069716A_ABST
    Figure CN123069716A_ABST
Patent Text Reader

Abstract

This invention relates to the field of air conditioning inspection technology, and discloses an optimized system and method for inspection judgment of air conditioning refrigeration systems. The system includes: a parameter fusion module for constructing a digital simulation model; a scenario simulation module for fusing the digital simulation model with the inspection scenario; a parameter model module for simulating multi-dimensional performance parameters; a performance analysis module for determining the performance level of the unit under test; and a trend analysis module for determining the stability level of the unit under test and generating an inspection report. This invention, by fusing a baseline architecture with multi-dimensional performance parameters into a digital simulation model, transforms the digital simulation model from a static physical simulation model into a virtual-real fusion model that can be dynamically calibrated according to the actual operating state of the unit under test. This avoids the problem of deviation between the simulation model and the actual system, provides high-quality data sources for performance analysis and trend analysis, and improves the pertinence and accuracy of inspection judgment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of air conditioning commercial inspection technology, and more specifically, to an optimized system and method for commercial inspection judgment of air conditioning refrigeration systems. Background Technology

[0002] With the widespread application of air conditioning and refrigeration systems in industrial, commercial, and residential fields, manufacturer inspection, performance evaluation, and energy efficiency assessment have become crucial aspects of ensuring product quality. In order to accurately and comprehensively identify potential problems in the performance of air conditioning and refrigeration systems and prevent substandard equipment from entering the market, it is necessary to conduct commercial inspections on air conditioning and refrigeration systems.

[0003] The patent application with publication number CN116066955A discloses a general intelligent commercial inspection control system and method, which includes acquiring the instantaneous value of refrigerant flow rate at the inspection site, current operating condition data, instantaneous value of condensing high pressure of the tested unit, and instantaneous value of evaporating low pressure of the tested unit, and sending the acquired data to a host computer; identifying the identity information of the tested unit; calculating the average refrigerant flow rate, average condensing high pressure, and average evaporating low pressure of the tested unit based on the identity information, the instantaneous value of refrigerant flow rate at the inspection site, current operating condition data, the instantaneous value of condensing high pressure, and the instantaneous value of evaporating low pressure of the tested unit; calculating the confidence level of the average refrigerant flow rate, average condensing high pressure, and average evaporating low pressure of the tested unit relative to the commercial inspection test, and judging the commercial inspection status based on the confidence level.

[0004] Existing commercial inspection judgment optimization systems typically perform offline modeling based on fixed design parameters and ideal operating conditions when constructing simulation models of air conditioning and refrigeration systems. However, in actual commercial inspection processes, the units under test are affected by various factors such as individual differences, performance degradation, and fluctuations in operating conditions. The offline-built models cannot dynamically track the actual operating status of the units under test, nor can they reflect the dynamic changes brought about by these factors. This leads to deviations between the model output and the actual system, which can easily cause misjudgments in the commercial inspection of air conditioning and refrigeration systems, making it difficult to provide reliable and accurate commercial inspection conclusions for specific conditions.

[0005] In view of this, the present invention proposes an optimization system and method for commercial inspection judgment of air conditioning refrigeration system to solve the above problems. Summary of the Invention

[0006] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: an air conditioning refrigeration system commodity inspection judgment optimization system, applied to a commodity inspection server, comprising: The parameter fusion module is used to collect the baseline parameters and multi-dimensional performance parameters of the unit under test, construct a baseline architecture based on the baseline parameters, and fuse the baseline architecture with the multi-dimensional performance parameters to construct a digital simulation model. The scenario simulation module is used to set the commercial inspection conditions and parameters of the unit under test, simulate the commercial inspection scenario, and integrate the digital simulation model with the commercial inspection scenario. The parameter model module is used to determine the inspection cycle based on time parameters, mark the inspection time within the inspection cycle, and simulate the multi-dimensional performance parameters of the digital simulation model at the inspection time with the inspection cycle as the time limit. The performance analysis module is used to input simulated multidimensional performance parameters into the performance learning model, identify performance indicators, including cooling capacity and cooling power consumption, and determine the performance level of the unit under test based on the fusion analysis results of the performance indicators. The trend analysis module is used to draw commercial inspection curves based on the inspection time and multi-dimensional performance parameters. Based on the trend analysis results of the commercial inspection curves, the stability level of the unit under test is determined, and a commercial inspection report of the unit under test is generated.

[0007] Furthermore, the baseline parameters include geometric data, structural data, and logical parameters; the multidimensional performance parameters include refrigerant-side parameters, air-side parameters, electrical-side parameters, and environmental-side parameters. The method for constructing a digital simulation model is as follows: The ten-coefficient performance curves of the logical parameters, the heat transfer coefficient and pressure drop coefficient, and the flow characteristic curves are embedded into the corresponding component entity structure models in the baseline architecture to generate physical mechanism sub-models. In chronological order, multidimensional performance parameters are sequentially input into the physical mechanism sub-model. The operating status of the unit under test is systematically identified using the gray box modeling method to obtain the data-driven sub-model. The data-driven sub-model indirectly calibrates the physical mechanism sub-model through model coupling. The refrigerant flow estimate output by the data-driven sub-model is used as the boundary condition input of the physical mechanism sub-model to construct a digital simulation model.

[0008] Furthermore, the inspection conditions include cooling, heating, and dehumidification; the inspection parameters include inspection temperature, inspection wind speed, and inspection wind direction. The method for scene model fusion is as follows: A virtual scene with two virtual workstations is simulated in the commodity inspection server, and scene boundaries and scene parameter fields are set in the virtual scene; Import the commodity inspection conditions and parameters into one of the virtual workstations, and establish a mapping channel between the scene parameter field and the parameter workstation; The digital simulation model is imported into another virtual workstation, and the scene parameter field of the commodity inspection scenario is mapped to the boundary conditions of the digital simulation model, so that the virtual scenario is transformed into a commodity inspection scenario.

[0009] Furthermore, the time parameters include temperature stabilization time, cooling response time, and baseline inspection time; The method for determining the commodity inspection cycle is as follows: Using the model and specifications of the unit under test as the baseline model and specifications, air conditioning units with the same baseline model and specifications are selected from the database and recorded as candidate units. Key parameters with the keywords "stability maintenance", "response" and "baseline" were extracted from the commercial inspection logs of the candidate units, and the numerical parts of the key parameters were recorded as the first duration, the second duration and the third duration, respectively. The average of all the first, second, and third durations is calculated to determine the temperature stabilization duration, cooling response duration, and baseline inspection duration. The inspection cycle is calculated by substituting the temperature stabilization time, cooling response time, and benchmark inspection time into the cycle fusion function.

[0010] Furthermore, when marking the inspection time, the starting time is the moment when the digital simulation model begins to simulate in the inspection scenario, and the ending time is the moment after one inspection cycle. Based on the calibrated parameter update duration, the inspection times are marked at intervals within the inspection cycle.

[0011] Furthermore, the performance levels include high energy efficiency, medium energy efficiency, and low energy efficiency. The method for determining the performance level is as follows: Under A inspection time, divide the cooling capacity by the cooling power consumption to calculate the energy efficiency ratio for each inspection time. When the energy efficiency ratio is greater than or equal to the preset first energy efficiency threshold, the inspection time is the high-performance time; when the energy efficiency ratio is less than the preset first energy efficiency threshold, but the energy efficiency ratio is greater than or equal to the preset second energy efficiency threshold, the inspection time is the medium-performance time; when the energy efficiency ratio is less than the preset second energy efficiency threshold, the inspection time is the low-performance time. Divide the number of high-performance time, medium-performance time and low-performance time by the number of commercial inspection time respectively to calculate the high-performance ratio, medium-performance ratio and low-performance ratio. When the proportion of high performance is greater than or equal to the preset high proportion threshold, the performance level is high energy efficiency level; when the proportion of low performance is greater than or equal to the preset low proportion threshold, the performance level is low energy efficiency level; when the proportion of medium performance is within the preset medium proportion threshold range, the performance level is medium energy efficiency level.

[0012] Furthermore, the method for drawing the commodity inspection curve is as follows: A basic coordinate system is drawn with the inspection time as the horizontal axis, the performance values ​​of multi-dimensional performance parameters as the vertical axis, and a standard of one multi-dimensional performance parameter corresponding to one inspection curve. Within the basic coordinate system, the points corresponding to the performance values ​​of the multidimensional performance parameters are recorded as curve points, and the commercial inspection curve is drawn by connecting the curve points in sequence. Find the standard performance parameters corresponding to the multidimensional performance parameters, and mark the points corresponding to the performance values ​​of the standard performance parameters in the basic coordinate system, and record them as standard points. After connecting all the standard points in sequence, a standard curve is drawn to obtain the commercial inspection curve of the unit under test.

[0013] Furthermore, the stability level includes high stability level and low stability level, and the method for determining the stability level is as follows: In the commodity inspection curve graph, for the same multidimensional performance parameter, the performance difference is calculated by taking the absolute value of the difference between the performance value of the commodity inspection curve and the performance value of the standard curve at the same commodity inspection time. The inspection time when the performance difference is less than or equal to the preset performance threshold is recorded as the approach time, and the inspection time when the performance difference is greater than the preset performance threshold is recorded as the distance time. Divide the number of near-times by the number of inspection times to calculate the near-time percentage. Then, sum the near-time percentages of all multidimensional performance parameters and take the average to calculate the stationary percentage. When the percentage of smooth operation is greater than or equal to the preset smoothness threshold, the smoothness level is high smoothness; when the percentage of smooth operation is less than the preset smoothness threshold, the smoothness level is low smoothness.

[0014] Furthermore, the commodity inspection report consists of commodity inspection conditions, commodity inspection parameters, multi-dimensional performance parameters, performance indicators, performance level, stability level, and commodity inspection conclusion; The inspection results include "passed inspection" and "failed inspection". When the performance level of the unit under test is high energy efficiency or medium energy efficiency, and the stability level is high smoothness, the commercial inspection conclusion is that the commercial inspection is qualified; when the performance level of the unit under test is low energy efficiency, or the stability level is low smoothness, the commercial inspection conclusion is that the commercial inspection is unqualified.

[0015] An optimization method for commodity inspection judgment of air conditioning refrigeration systems, applied to a commodity inspection server, is implemented based on an optimization system for commodity inspection judgment of air conditioning refrigeration systems, including: SS1: Collect the baseline parameters and multi-dimensional performance parameters of the unit under test, construct a baseline architecture based on the baseline parameters, and integrate the baseline architecture with the multi-dimensional performance parameters to construct a digital simulation model; SS2: Set the commercial inspection conditions and parameters of the unit under test, simulate the commercial inspection scenario, and merge the digital simulation model with the commercial inspection scenario. SS3: Based on time parameters, determine the inspection cycle, mark the inspection time within the inspection cycle, and use the inspection cycle as a time limit to simulate the multi-dimensional performance parameters of the digital simulation model at the inspection time. SS4: Input the simulated multidimensional performance parameters into the performance learning model, identify the performance indicators, and determine the performance level of the unit under test based on the fusion analysis results of the performance indicators; SS5: Plot the commercial inspection curve using the inspection time and multi-dimensional performance parameters. Based on the trend analysis results of the commercial inspection curve, determine the stability level of the unit under test and generate the commercial inspection report for the unit under test.

[0016] The technical advantages of the present invention, a commercial inspection judgment optimization system and method for air conditioning refrigeration systems, are as follows: (1): This invention integrates the benchmark architecture with multi-dimensional performance parameters into a digital simulation model, so that the digital simulation model is no longer a static physical simulation model, but a virtual-real fusion model that can be dynamically calibrated according to the actual operating state of the unit under test. This ensures the rationality of the model output at the physical level, avoids the problem of deviation between traditional simulation models and actual systems, provides a high-quality data source for subsequent performance analysis and trend analysis, and improves the pertinence and accuracy of commodity inspection judgment.

[0017] (2): This invention adopts a two-dimensional classification judgment method of performance analysis and trend analysis, realizing the optimization transformation from a single binary judgment to a multi-dimensional classification judgment. It avoids the misjudgment phenomenon caused by the occasional fluctuation at a single inspection time, and can also reflect the severity of the deviation of the performance parameters of the unit under test from the standard value during the inspection cycle. It helps to identify the unit under test that has qualified performance level but poor stability level, so as to discover potential deterioration trends before the performance parameters exceed the qualified threshold, and achieve the judgment optimization effect from passive judgment to active prevention. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of a commercial inspection judgment optimization system for an air conditioning refrigeration system provided in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the architecture of an air conditioning refrigeration system commercial inspection judgment optimization system provided in Embodiment 1 of the present invention; Figure 3 This is a flowchart illustrating an optimization method for commercial inspection of an air conditioning refrigeration system provided in Embodiment 2 of the present invention. Detailed Implementation

[0019] 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.

[0020] Example 1: Please refer to Figures 1-2 As shown in this embodiment, an air conditioning refrigeration system commodity inspection judgment optimization system is applied to a commodity inspection server and includes: The parameter fusion module collects the baseline parameters and multi-dimensional performance parameters of the unit under test, constructs a baseline architecture based on the baseline parameters, and integrates the baseline architecture with the multi-dimensional performance parameters to build a digital simulation model. The unit under test refers to the air conditioning unit whose air conditioning refrigeration system needs to be inspected and optimized in this test. The unit under test includes, but is not limited to, split air conditioning units, multi-split air conditioning units, and chiller units.

[0021] Baseline parameters are hardware parameters used to provide a basic description of the physical structure and operating logic of the unit under test; Specifically, the baseline parameters include geometric data, structural data, and logical parameters.

[0022] Geometric data refers to the three-dimensional spatial shape and size information of each component of the unit under test, which is obtained by querying the design drawings or CAD model of the unit under test; geometric data includes, but is not limited to, the tube diameter, tube spacing, fin type, fin parameters and loop design parameters of the heat exchanger.

[0023] Structural data refers to the assembly relationships and physical structure information of the components of the unit under test, which is obtained by consulting the assembly drawings of the unit under test; structural data includes, but is not limited to, the cylinder diameter, piston stroke, theoretical displacement and connection relationships of the compressor, etc.

[0024] Logical parameters refer to the parameterized representation of the physical laws and control logic followed by each component of the unit under test during operation. They are obtained by consulting the performance data sheets of the compressor manufacturer and the calibration test data of the heat exchanger. Logical parameters include, but are not limited to, the ten-coefficient curve of the compressor, the heat transfer coefficient and pressure drop coefficient of the heat exchanger, and the flow characteristic curve of the expansion valve.

[0025] The baseline architecture is a three-dimensional model used to represent the physical structural components of the unit under test, serving as the skeleton basis for subsequent integration with multi-dimensional performance parameters; In this embodiment, when constructing the baseline architecture, based on geometric and structural data, a solid structure model of each component in the unit under test is constructed using a 3D modeling engine (e.g., SolidWorks). The solid structure models of each component are then assembled according to the actual assembly relationship to construct the baseline architecture at the solid structure level.

[0026] Multidimensional performance parameters are performance-related parameters of the unit under test in multiple dimensions during the commercial inspection process, and serve as data inputs for the dynamic calibration of the subsequent digital simulation model; specifically, multidimensional performance parameters include refrigerant-side parameters, air-side parameters, electrical-side parameters, and environmental-side parameters.

[0027] Refrigerant-side parameters refer to parameters that reflect the state changes of refrigerant in a thermodynamic cycle. Refrigerant-side parameters include, but are not limited to, compressor discharge pressure, compressor suction pressure, compressor discharge temperature, compressor suction temperature, and refrigerant flow rate. Among them, refrigerant flow rate is an optional parameter, and an estimated value is used when direct measurement is unavailable.

[0028] Air-side parameters refer to parameters that reflect the changes in the state of air before and after passing through the heat exchanger. Air-side parameters include, but are not limited to, outlet dry-bulb temperature, outlet wet-bulb temperature, outlet air volume, and coil temperature.

[0029] Electrical side parameters refer to parameters that reflect the power consumption and operating status of electrical components of the unit under test. Electrical side parameters include, but are not limited to, compressor operating current, operating frequency, input power and winding temperature.

[0030] Environmental parameters refer to parameters that reflect the conditions of the commodity inspection scenario. Environmental parameters include, but are not limited to, ambient dry-bulb temperature, ambient wet-bulb temperature, and atmospheric pressure.

[0031] In this embodiment, the method for collecting multidimensional performance parameters is consistent with the method for collecting corresponding parameters in the existing commodity inspection process. For example, power, voltage, and current are collected using a power meter, heat exchanger coil pressure is collected using a pressure gauge, and exhaust temperature, inlet temperature, outlet temperature, and ambient temperature of the commodity inspection room are collected using a thermometer.

[0032] The digital simulation model is a dynamic virtual model of the unit under test in a computer. It includes the physical structure and operating mechanism of the air conditioner, and also incorporates the real-time operating data collected during this commercial inspection. It can simulate the operating status of the unit under test within a specific time period in a virtual commercial inspection scenario, and output multi-dimensional performance parameters at each commercial inspection moment.

[0033] The method for constructing a digital simulation model is as follows: The ten-coefficient performance curves of the logical parameters, the heat transfer coefficient, the pressure drop coefficient, and the flow characteristic curves are embedded into the corresponding component entity structure models in the baseline architecture to generate physical mechanism sub-models. Specifically, the evaporator in the baseline architecture is divided into a two-phase region and a superheated region along the refrigerant flow direction, and the condenser is divided into a superheated region, a two-phase region, and a subcooled region. In each region, refrigerant-side mass conservation equations, energy conservation equations, and pipe wall energy conservation equations are established. The compressor adopts a ten-coefficient semi-empirical model, with evaporation temperature and condensation temperature as independent variables, to fit the refrigerant mass flow rate and compressor input power. Among them, the refrigerant-side mass conservation equations, energy conservation equations, and pipe wall energy conservation equations are all existing technologies in this field, and their specific forms can be referred to the heat exchanger models in the Modelica standard library. Following the chronological order, multidimensional performance parameters are sequentially input into the physical mechanism sub-model. The operating state of the unit under test is then systematically identified using the gray-box modeling method, resulting in a data-driven sub-model. The system identification process of the gray-box modeling method is as follows: Thermal resistance, heat capacity, and evaporative heat transfer coefficient are set as the parameter vector to be identified; a gray-box model is established using compressor status indication and ambient temperature as inputs, and refrigerant mass flow rate and compressor power as outputs; a genetic algorithm is used to optimize the parameter vector to be identified, with the objective function being minimizing the root mean square error (RMSE) between the simulation output and the measured output; the solution is iteratively solved until the RMSE converges to a preset accuracy threshold, completing the system identification. When the refrigerant flow rate of the unit under test lacks direct measurement values, the data-driven sub-model indirectly calibrates the physical mechanism sub-model through model coupling, using the refrigerant flow rate estimate output by the data-driven sub-model as the boundary condition input of the physical mechanism sub-model. The model coupling method indirectly calibrates the physical mechanism sub-model through the data-driven sub-model. In this embodiment, the calibration target is that the average absolute percentage error of the flow rate estimate does not exceed 3%. The physical mechanism sub-model and the data-driven sub-model are fused through model coupling to construct a digital simulation model.

[0034] In this embodiment, the gray box modeling method embeds real physical parameters (thermal resistance, heat capacity, evaporation heat transfer coefficient) into the data-driven model, so that the digital simulation model has both the interpretability of physical mechanisms and the ability to capture nonlinear behaviors and individual differences of the unit under test that are difficult to model accurately in the physical model through data-driven methods.

[0035] The scenario simulation module sets the commercial inspection conditions and parameters of the unit under test, simulates the commercial inspection scenario, and integrates the digital simulation model with the commercial inspection scenario. Inspection conditions refer to the types of operating conditions that the unit under test needs to operate during the inspection process. Different inspection conditions correspond to different inspection requirements. Specifically, the inspection conditions include refrigeration, heating, and dehumidification. Refrigeration condition refers to the refrigeration system of the unit under test being in a cooling state, heating condition refers to the refrigeration system of the unit under test being in a heating state, and dehumidification condition refers to the refrigeration system of the unit under test being in a dehumidification and drying state.

[0036] It should be noted that at any given time, only one type of commercial inspection operation can be performed on the unit under test, which can reduce the interference caused by simultaneous testing of different commercial inspection conditions; in this embodiment, the commercial inspection condition is limited to the refrigeration condition.

[0037] The inspection parameters are the operating settings that the unit under test needs to meet under the corresponding inspection conditions. The inspection conditions can be defined from three dimensions: temperature, wind speed, and wind direction. Specifically, the inspection parameters include inspection temperature, inspection wind speed, and inspection wind direction; among them, inspection temperature refers to the temperature required for the unit under test during inspection, inspection wind speed refers to the airflow velocity required for the unit under test during inspection, and inspection wind direction refers to the wind direction required for the unit under test during inspection.

[0038] The commodity inspection scenario is a virtual scenario simulated in the commodity inspection server for simulating and testing the air conditioning and refrigeration system of the unit under test. The virtual workstation is a location constraint in the commodity inspection scenario used to import digital simulation models and corresponding operating parameters, thereby laying a safe and reliable foundation for the integration of digital simulation models with commodity inspection operating conditions and parameters. The method for scene model fusion is as follows: A virtual scene with two virtual workstations is simulated in the commodity inspection server, and scene boundaries and scene parameter fields are set in the virtual scene. The scene boundary is used to provide a closed boundary for the commodity inspection range of the unit under test. The scene boundary includes, but is not limited to, the walls, floors, and ceilings of the commodity inspection room. The scene parameter field is used to limit the external factors in the commodity inspection scene. The scene parameter field includes the initial distribution of temperature field, humidity field, and wind speed field. Import the commodity inspection conditions and parameters into one of the virtual workstations to obtain the parameter workstation, and establish a mapping channel between the scene parameter field and the parameter workstation; the mapping channel can establish a mapping channel between the virtual scene and the parameter workstation, which is used to call the commodity inspection conditions and parameters, so as to provide accurate data parameter support for the commodity inspection operation in the virtual scene; The digital simulation model is imported into another virtual workstation to obtain the model workstation, and the scene parameter field of the commodity inspection scenario is mapped to the boundary conditions of the digital simulation model, so that the virtual scenario is transformed into the commodity inspection scenario.

[0039] In this embodiment, the constructed virtual scene can be understood as a blank commercial inspection scene. Only after the virtual scene is integrated with the digital simulation model, commercial inspection conditions and commercial inspection parameters can it be transformed into a commercial inspection scene that can meet the commercial inspection requirements of the unit under test. Therefore, each commercial inspection scene is unique and independent.

[0040] The parameter model module determines the inspection cycle based on time parameters, marks the inspection time within the inspection cycle, and simulates the multi-dimensional performance parameters of the digital simulation model at the inspection time using the inspection cycle as a time limit. Time parameters refer to different types of time data that affect the performance changes of the air conditioning refrigeration system in the unit under test, and serve as the basis for influencing the specific duration of the digital simulation model in the commodity inspection scenario. Specifically, the time parameters include temperature stabilization time, cooling response time, and baseline inspection time.

[0041] Temperature stabilization time refers to the time required for the compressor of the unit under test to go from a static state to the rated operating temperature. In other words, it can be expressed as the time it takes for the operating temperature of the compressor in the unit under test to change.

[0042] Cooling response time refers to the response time required for the refrigerant flow rate of the unit under test to reach a stable flow rate value from system startup. In other words, it can be expressed as the duration of refrigerant flow rate changes in the unit under test.

[0043] The benchmark inspection duration refers to the benchmark duration for each inspection operation performed on the unit under test, which can serve as the basis for the specific inspection time span of each unit under test.

[0044] The inspection cycle refers to the total time for the digital simulation model to simulate in the inspection scenario, and serves as the time basis for the inspection judgment and optimization of the unit under test. The method for determining the commodity inspection cycle is as follows: Using the model and specifications of the unit under test as the baseline model and specifications, air conditioning units with the same baseline model and specifications are selected from the database and recorded as candidate units. The commercial inspection logs of each candidate unit were analyzed one by one. Key parameters with the keywords "stability maintenance", "response" and "baseline" were extracted from the commercial inspection logs, and the numerical parts of the key parameters were recorded as the first duration, the second duration and the third duration, respectively. The average of all the first, second, and third durations is calculated to determine the temperature stabilization duration, cooling response duration, and baseline inspection duration. The temperature stabilization time, cooling response time, and benchmark inspection time are substituted into the period fusion function to calculate the inspection period. The expression for the periodic fusion function is: ; In the formula, For the commodity inspection cycle, For the duration of temperature stabilization, For cooling response time, Based on the standard commodity inspection time, , , These are the weighting factors for temperature stabilization time, cooling response time, and baseline inspection time, respectively. , , All are greater than 0, and , , The sum of is 1.

[0045] Once the inspection cycle is established, it is a time period consisting of multiple consecutive moments. After a certain period of time within the inspection cycle, the unit under test will generate corresponding multi-dimensional performance parameters. In order to simulate the performance of the unit under test at multiple time points within the inspection cycle in a specific and detailed manner, it is necessary to mark the inspection moments with intervals within the inspection cycle. In this embodiment, when marking the inspection time, the starting time is the time when the digital simulation model begins to simulate in the inspection scenario, and the ending time is the time after one inspection cycle. Based on the calibrated parameter update duration, A interval-distributed inspection times are marked within the inspection cycle.

[0046] The calibrated parameter update duration is a limitation on the time span between two adjacent inspection times, and serves as the parameter recording interval duration corresponding to the unit under test during inspection. In this embodiment, the calibrated parameter update duration can be set based on the minimum value of the parameter acquisition interval duration of all sensors in the unit under test, or it can be obtained by statistically averaging the parameter update durations of a large number of air conditioning units of the same type and specifications in history.

[0047] After marking the A interval distribution of inspection times, the digital simulation model now possesses not only the scenario conditions for simulation but also the time conditions for simulation in the inspection scenario. Therefore, it is necessary to perform simulation operations on the multidimensional performance parameters of the digital simulation model at each inspection time within the inspection cycle. Specifically, after importing the digital simulation model into the virtual workstation of the commodity inspection scenario, the digital simulation model performs simulation in the commodity inspection scenario according to the commodity inspection working conditions and commodity inspection parameters, and simulates the multi-dimensional performance parameters of the digital simulation model at A commodity inspection moments within the commodity inspection cycle one by one.

[0048] It should be noted that the multidimensional performance parameters simulated at A inspection times need to be summarized and arranged in the order of the simulation, so that the multidimensional performance parameters at each inspection time are kept in a summarized state for subsequent query and analysis.

[0049] The performance analysis module inputs the simulated multidimensional performance parameters into the performance learning model, identifies the performance indicators, including cooling capacity and cooling power consumption, and determines the performance level of the unit under test based on the fusion analysis results of the performance indicators. After simulating the multidimensional performance parameters at A commercial inspection times, the multidimensional performance parameters can be input into the performance learning model for deep learning, thereby enabling multi-time-point analysis of the specific situation of the unit under test during the commercial inspection cycle.

[0050] The performance learning model is an artificial intelligence model based on deep learning technology that can analyze and judge the true meaning contained in multi-dimensional performance parameters and output corresponding performance indicators, thereby providing an analysis tool for the specific performance status of the unit under test at each inspection moment during the commercial inspection cycle. Performance indicators are used to measure the specific performance of the unit under test during the commercial inspection process. Specifically, performance indicators include cooling capacity and cooling power consumption. Cooling capacity refers to the total amount of heat removed by the unit under test from a closed space, room, or area per unit time; cooling power consumption refers to the effective input power consumed by the unit under test in cooling mode.

[0051] The performance learning model is not obtained directly in advance, but needs to be trained on the basis of the deep learning model. The performance learning model includes an input layer, a feature extraction layer, a temporal modeling layer, and an output layer. The input layer receives multi-dimensional performance parameters at the time of commodity inspection; the feature extraction layer extracts local features from the multi-dimensional performance parameters through one-dimensional convolutional kernels, and the number and size of the convolutional kernels are determined according to the dimension and sampling density of the multi-dimensional performance parameters; the temporal modeling layer performs bidirectional temporal modeling on the feature sequence output by the feature extraction layer, and the dimension and number of hidden layers are determined according to the number of time points of commodity inspection and the complexity of parameter changes; the output layer outputs the cooling capacity and cooling power consumption of the unit under test.

[0052] The training method for the performance learning model is as follows: Multidimensional performance parameters and performance indicators of M units of the same type and specifications that have completed commercial inspection are collected in advance at various commercial inspection times in the commercial inspection scenario. The multidimensional performance parameters are used as input samples, and the cooling capacity and cooling power consumption are used as label data. The input samples and label data are combined to form a training dataset. Set the number and size of convolutional kernels for the feature extraction layer, set the dimension and number of hidden layers for the temporal modeling layer, and set the loss function of the output layer to the mean squared error loss function. The performance learning model is iteratively trained using the Adam optimizer until the mean squared error loss function converges to below the preset loss threshold, thus obtaining a well-trained performance learning model.

[0053] It should be noted that the Adam optimizer is existing technology in this field and will not be described in detail here. For example, the preset loss threshold is 0.01.

[0054] In this embodiment, the multidimensional performance parameters of A inspection times are input into the performance learning model one by one, and the corresponding performance indicators of A inspection times can be output.

[0055] After obtaining the performance indicators of the unit under test, it is necessary to conduct a comprehensive analysis of the performance indicators and, based on the results of the comprehensive analysis, determine the specific performance of the unit under test in this commercial inspection cycle, thereby determining the performance level of the unit under test in this commercial inspection. The performance rating is used to indicate the overall energy efficiency of the air conditioning and refrigeration system of the unit under test, and is one of the dimensions of the commercial inspection results of the unit under test. Specifically, the performance levels include high energy efficiency, medium energy efficiency, and low energy efficiency; the energy efficiency performance corresponding to the high, medium, and low energy efficiency levels is from high to low.

[0056] The method for determining performance levels is as follows: Under A inspection time, divide the cooling capacity by the cooling power consumption to calculate the energy efficiency ratio for each inspection time. The energy efficiency ratio is compared with a preset first energy efficiency threshold and a preset second energy efficiency threshold, wherein the preset first energy efficiency threshold is greater than the preset second energy efficiency threshold. The preset first energy efficiency threshold and the preset second energy efficiency threshold refer to the critical values ​​at which the energy efficiency performance corresponding to the energy efficiency ratio changes at the time of commercial inspection, thereby providing a precise numerical basis for determining the performance level. The preset first energy efficiency threshold and the preset second energy efficiency threshold are determined according to the rated energy efficiency ratio of the unit under test. For example, the preset first energy efficiency threshold is 90% of the rated energy efficiency ratio, and the preset second energy efficiency threshold is 80% of the rated energy efficiency ratio. When the energy efficiency ratio is greater than or equal to the preset first energy efficiency threshold, it indicates that the energy efficiency performance of the unit under test is strong, and the commercial inspection time is recorded as the high performance time. When the energy efficiency ratio is less than the preset first energy efficiency threshold and the energy efficiency ratio is greater than or equal to the preset second energy efficiency threshold, it indicates that the energy efficiency performance of the unit under test is medium, and the inspection time is recorded as the medium performance time. When the energy efficiency ratio is less than the preset second energy efficiency threshold, it indicates that the energy efficiency performance of the unit under test is weak, and the test time is recorded as a low performance time. Divide the number of high-performance time, medium-performance time and low-performance time by the number of commercial inspection time respectively to calculate the high-performance ratio, medium-performance ratio and low-performance ratio. When the high performance ratio is greater than or equal to the preset high ratio threshold, it indicates that the unit under test has maintained high energy efficiency for a relatively long period during the commercial inspection cycle, and the performance level is determined to be high energy efficiency level; the preset high ratio threshold refers to the minimum value of the high performance ratio corresponding to being recorded as high energy efficiency level. When the percentage of low performance is greater than or equal to the preset low percentage threshold, it indicates that the unit under test has maintained low energy efficiency for a relatively long period during the commercial inspection cycle, and the performance level is determined to be low energy efficiency level; the preset low percentage threshold refers to the minimum value of low performance percentage corresponding to being recorded as low energy efficiency level. When the performance percentage falls within the preset medium percentage threshold range, it indicates that the unit under test maintains medium energy efficiency for a moderate period during the commercial inspection cycle, and its performance level is determined to be medium energy efficiency. The preset medium percentage threshold range refers to the coverage area of ​​the medium performance percentage corresponding to being recorded as medium energy efficiency level; It should be noted that the performance level is a holistic representation of all inspection moments within the inspection cycle, not the result of a single inspection moment. Therefore, the unit under test has one and only one performance level within an inspection cycle.

[0057] The trend analysis module plots the inspection curve using the inspection time and multi-dimensional performance parameters. Based on the trend analysis results of the inspection curve, it determines the stability level of the unit under test and generates an inspection report for the unit under test. The inspection curve is a curve distribution diagram formed by each inspection time and the corresponding performance value within the inspection cycle. It can continuously represent the changes of multi-dimensional performance parameters of the unit under test along the time line within the inspection cycle. The method for drawing the commodity inspection curve is as follows: Using the inspection time as the horizontal axis, the performance values ​​of the multi-dimensional performance parameters as the vertical axis, and taking the inspection curve corresponding to one multi-dimensional performance parameter as the standard, B basic coordinate systems are drawn. Within the basic coordinate system, the points corresponding to the performance values ​​of the multidimensional performance parameters are recorded as curve points, and the commercial inspection curve is drawn by connecting the curve points in sequence. The standard performance parameters corresponding to the multidimensional performance parameters are retrieved from the technical database, and the points corresponding to the performance values ​​of the standard performance parameters are marked in the basic coordinate system and recorded as standard points. After connecting all the standard points in sequence, a standard curve is drawn to obtain the commercial inspection curve of the unit under test during the commercial inspection cycle.

[0058] In this embodiment, the standard performance parameters are multi-dimensional performance parameters that are pre-stored in the technical database and are corresponding to the unit under test under normal and fault-free conditions, so that the standard performance parameters can be used as the standard values ​​of the multi-dimensional performance parameters.

[0059] The stability rating is used to indicate the overall stability performance of the air conditioning and refrigeration system of the unit under test, serving as another dimension of the commercial inspection results for the unit under test. Specifically, the smoothness levels include high smoothness level and low smoothness level; among them, the smoothness performance corresponding to the high smoothness level and low smoothness level is from high to low.

[0060] The method for determining the stability level is as follows: In the commodity inspection curve graph, for the same multidimensional performance parameter, the performance difference is calculated by taking the absolute value of the difference between the performance value of the commodity inspection curve and the performance value of the standard curve at the same commodity inspection time. The performance difference is compared with a preset performance threshold. The inspection time when the performance difference is less than or equal to the preset performance threshold is recorded as the approach time, and the inspection time when the performance difference is greater than the preset performance threshold is recorded as the distance time. The preset performance threshold is the critical value of the performance difference corresponding to the inspection time being divided into the approach time and the distance time, thereby distinguishing the stable performance of different inspection times. Divide the number of near-times by the number of inspection times to calculate the near-time percentage. Then, sum the near-time percentages of all multidimensional performance parameters and take the average to calculate the stationary percentage. When the smoothness ratio is greater than or equal to the preset smoothness threshold, it indicates that the performance parameters of the unit under test have strong smoothness stability during the commercial inspection cycle, and the smoothness level is determined to be a high smoothness level. When the smoothness percentage is less than the preset smoothness threshold, it indicates that the smoothness stability of the performance parameters of the unit under test is weak during the commercial inspection cycle, and the smoothness level is determined to be low smoothness level.

[0061] The preset smoothness threshold is a critical value used to distinguish the smoothness ratio corresponding to high smoothness level and low smoothness level. The preset smoothness threshold is obtained by averaging the critical values ​​of the smoothness ratio corresponding to a large number of historical high smoothness levels and low smoothness levels; for example, the preset smoothness threshold is 75%.

[0062] The inspection report is a summary report of all inspection information of the unit under test during the current inspection cycle, and can be used as the information basis for the inspection judgment and subsequent optimization of the unit under test. Specifically, the commodity inspection report consists of commodity inspection conditions, commodity inspection parameters, multi-dimensional performance parameters, performance indicators, performance level, stability level, and commodity inspection conclusion.

[0063] The inspection conclusion is the final judgment of the unit under test in this inspection, and thus serves as the direct basis for determining whether the unit under test meets the factory requirements. The inspection results include "passed inspection" and "failed inspection". Specifically, when the performance level of the unit under test is high energy efficiency or medium energy efficiency, and the stability level is high smoothness, the commercial inspection conclusion of the unit under test is qualified. When the performance level of the unit under test is low energy efficiency or the stability level is low smoothness, the commercial inspection conclusion of the unit under test is that it fails the commercial inspection.

[0064] In this embodiment, by judging the commercial inspection conclusion of the unit under test, a final conclusion can be provided for the current commercial inspection of the unit under test, and it also serves as the basis for whether to optimize and adjust the unit under test in the future.

[0065] Example 2: Please refer to Figure 3 As shown, parts not described in detail in this embodiment are described in Embodiment 1. This embodiment provides an optimization method for commodity inspection judgment of an air conditioning refrigeration system, applied to a commodity inspection server, and implemented based on an optimization system for commodity inspection judgment of an air conditioning refrigeration system, including: SS1: Collect the baseline parameters and multi-dimensional performance parameters of the unit under test, construct a baseline architecture based on the baseline parameters, and integrate the baseline architecture with the multi-dimensional performance parameters to construct a digital simulation model; SS2: Set the commercial inspection conditions and parameters of the unit under test, simulate the commercial inspection scenario, and merge the digital simulation model with the commercial inspection scenario. SS3: Based on time parameters, determine the inspection cycle, mark the inspection time within the inspection cycle, and use the inspection cycle as a time limit to simulate the multi-dimensional performance parameters of the digital simulation model at the inspection time. SS4: Input the simulated multidimensional performance parameters into the performance learning model, identify the performance indicators, and determine the performance level of the unit under test based on the fusion analysis results of the performance indicators; SS5: Plot the commercial inspection curve using the inspection time and multi-dimensional performance parameters. Based on the trend analysis results of the commercial inspection curve, determine the stability level of the unit under test and generate the commercial inspection report for the unit under test.

[0066] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. An optimization system for commodity inspection judgment of air conditioning refrigeration systems, applied to a commodity inspection server, characterized in that, include: The parameter fusion module is used to collect the baseline parameters and multi-dimensional performance parameters of the unit under test, construct a baseline architecture based on the baseline parameters, and fuse the baseline architecture with the multi-dimensional performance parameters to construct a digital simulation model. The scenario simulation module is used to set the commercial inspection conditions and parameters of the unit under test, simulate the commercial inspection scenario, and integrate the digital simulation model with the commercial inspection scenario. The parameter model module is used to determine the inspection cycle based on time parameters, mark the inspection time within the inspection cycle, and simulate the multi-dimensional performance parameters of the digital simulation model at the inspection time with the inspection cycle as the time limit. The performance analysis module is used to input simulated multidimensional performance parameters into the performance learning model, identify performance indicators, including cooling capacity and cooling power consumption, and determine the performance level of the unit under test based on the fusion analysis results of the performance indicators. The trend analysis module is used to draw commercial inspection curves based on the inspection time and multi-dimensional performance parameters. Based on the trend analysis results of the commercial inspection curves, the stability level of the unit under test is determined, and a commercial inspection report of the unit under test is generated.

2. The air conditioning refrigeration system commercial inspection judgment optimization system according to claim 1, characterized in that, The baseline parameters include geometric data, structural data, and logical parameters; the multidimensional performance parameters include refrigerant-side parameters, air-side parameters, electrical-side parameters, and environmental-side parameters. The method for constructing a digital simulation model is as follows: The ten-coefficient performance curves of the logical parameters, the heat transfer coefficient and pressure drop coefficient, and the flow characteristic curves are embedded into the corresponding component entity structure models in the baseline architecture to generate physical mechanism sub-models. In chronological order, multidimensional performance parameters are sequentially input into the physical mechanism sub-model. The operating status of the unit under test is systematically identified using the gray box modeling method to obtain the data-driven sub-model. The data-driven sub-model indirectly calibrates the physical mechanism sub-model through model coupling. The refrigerant flow estimate output by the data-driven sub-model is used as the boundary condition input of the physical mechanism sub-model to construct a digital simulation model.

3. The air conditioning refrigeration system commercial inspection judgment optimization system according to claim 2, characterized in that, The inspection conditions include cooling, heating, and dehumidification; the inspection parameters include inspection temperature, inspection fan speed, and inspection air direction. The method for scene model fusion is as follows: A virtual scene with two virtual workstations is simulated in the commodity inspection server, and scene boundaries and scene parameter fields are set in the virtual scene; Import the commodity inspection conditions and parameters into one of the virtual workstations, and establish a mapping channel between the scene parameter field and the parameter workstation; The digital simulation model is imported into another virtual workstation, and the scene parameter field of the commodity inspection scenario is mapped to the boundary conditions of the digital simulation model, so that the virtual scenario is transformed into a commodity inspection scenario.

4. The air conditioning refrigeration system commercial inspection judgment optimization system according to claim 3, characterized in that, The time parameters include temperature stabilization time, cooling response time, and baseline inspection time; The method for determining the commodity inspection cycle is as follows: Using the model and specifications of the unit under test as the baseline model and specifications, air conditioning units with the same baseline model and specifications are selected from the database and recorded as candidate units. Key parameters with the keywords "stability maintenance", "response" and "baseline" were extracted from the commercial inspection logs of the candidate units, and the numerical parts of the key parameters were recorded as the first duration, the second duration and the third duration, respectively. The average of all the first, second, and third durations is calculated to determine the temperature stabilization duration, cooling response duration, and baseline inspection duration. The inspection cycle is calculated by substituting the temperature stabilization time, cooling response time, and benchmark inspection time into the cycle fusion function.

5. The air conditioning refrigeration system commercial inspection judgment optimization system according to claim 4, characterized in that, When marking inspection times, the starting time is the moment when the digital simulation model begins to simulate in the inspection scenario, and the ending time is the moment after one inspection cycle. Based on the calibrated parameter update duration, the inspection times are marked at intervals within the inspection cycle.

6. The air conditioning refrigeration system commercial inspection judgment optimization system according to claim 5, characterized in that, Performance levels include high energy efficiency, medium energy efficiency, and low energy efficiency. The method for determining the performance level is as follows: Under inspection time A, divide the cooling capacity by the cooling power consumption to calculate the energy efficiency ratio for each inspection time. When the energy efficiency ratio is greater than or equal to the preset first energy efficiency threshold, the inspection time is the high-performance time; when the energy efficiency ratio is less than the preset first energy efficiency threshold, but the energy efficiency ratio is greater than or equal to the preset second energy efficiency threshold, the inspection time is the medium-performance time; when the energy efficiency ratio is less than the preset second energy efficiency threshold, the inspection time is the low-performance time. Divide the number of high-performance time, medium-performance time and low-performance time by the number of commercial inspection time respectively to calculate the high-performance ratio, medium-performance ratio and low-performance ratio. When the proportion of high performance is greater than or equal to the preset high proportion threshold, the performance level is high energy efficiency level; when the proportion of low performance is greater than or equal to the preset low proportion threshold, the performance level is low energy efficiency level; when the proportion of medium performance is within the preset medium proportion threshold range, the performance level is medium energy efficiency level.

7. The air conditioning refrigeration system commercial inspection judgment and optimization system according to claim 6, characterized in that, The method for drawing the commodity inspection curve is as follows: A basic coordinate system is drawn with the inspection time as the horizontal axis, the performance values ​​of multi-dimensional performance parameters as the vertical axis, and a standard of one multi-dimensional performance parameter corresponding to one inspection curve. Within the basic coordinate system, the points corresponding to the performance values ​​of the multidimensional performance parameters are recorded as curve points, and the commercial inspection curve is drawn by connecting the curve points in sequence. Find the standard performance parameters corresponding to the multidimensional performance parameters, and mark the points corresponding to the performance values ​​of the standard performance parameters in the basic coordinate system, and record them as standard points. After connecting all the standard points in sequence, a standard curve is drawn to obtain the commercial inspection curve of the unit under test.

8. The air conditioning refrigeration system commercial inspection judgment optimization system according to claim 7, characterized in that, The stability level includes high smoothness level and low smoothness level. The method for determining the stability level is as follows: In the commodity inspection curve graph, for the same multidimensional performance parameter, the performance difference is calculated by taking the absolute value of the difference between the performance value of the commodity inspection curve and the performance value of the standard curve at the same commodity inspection time. The inspection time when the performance difference is less than or equal to the preset performance threshold is recorded as the approach time, and the inspection time when the performance difference is greater than the preset performance threshold is recorded as the distance time. Divide the number of near-times by the number of inspection times to calculate the near-time percentage. Then, sum the near-time percentages of all multidimensional performance parameters and take the average to calculate the stationary percentage. When the percentage of smooth operation is greater than or equal to the preset smoothness threshold, the smoothness level is high smoothness; when the percentage of smooth operation is less than the preset smoothness threshold, the smoothness level is low smoothness.

9. The air conditioning refrigeration system commercial inspection judgment optimization system according to claim 8, characterized in that, The commodity inspection report consists of commodity inspection conditions, commodity inspection parameters, multi-dimensional performance parameters, performance indicators, performance level, stability level, and commodity inspection conclusion; The inspection results include "passed inspection" and "failed inspection". When the performance level of the unit under test is high energy efficiency or medium energy efficiency, and the stability level is high smoothness, the commercial inspection conclusion is that the commercial inspection is qualified; when the performance level of the unit under test is low energy efficiency, or the stability level is low smoothness, the commercial inspection conclusion is that the commercial inspection is unqualified.

10. A method for optimizing the inspection of an air conditioning refrigeration system, applied to an inspection server, implemented based on the air conditioning refrigeration system inspection optimization system according to any one of claims 1-9, characterized in that... include: SS1: Collect the baseline parameters and multi-dimensional performance parameters of the unit under test, construct a baseline architecture based on the baseline parameters, and integrate the baseline architecture with the multi-dimensional performance parameters to construct a digital simulation model; SS2: Set the commercial inspection conditions and parameters of the unit under test, simulate the commercial inspection scenario, and merge the digital simulation model with the commercial inspection scenario. SS3: Based on time parameters, determine the inspection cycle, mark the inspection time within the inspection cycle, and simulate the multi-dimensional performance parameters of the digital simulation model at the inspection time using the inspection cycle as the time limit; SS4: Input the simulated multidimensional performance parameters into the performance learning model, identify the performance indicators, and determine the performance level of the unit under test based on the fusion analysis results of the performance indicators. SS5: Plot the commercial inspection curve using the commercial inspection time and multi-dimensional performance parameters. Based on the trend analysis results of the commercial inspection curve, determine the stability level of the unit under test and generate the commercial inspection report of the unit under test.

Citation Information

Patent Citations

  • Universal intelligent commodity inspection control system and method

    CN116066955A