An area corrosion environment inversion method based on air conditioner operation data

Abstract

The application provides a regional corrosion environment inversion method based on air conditioner operation data, follows a technical route of data collection, preprocessing, sensitive parameter identification, mapping relationship establishment, inversion model construction, regional inversion and grade output, and specifically comprises the following steps: air conditioner operation parameter collection, data preprocessing and standardization, corrosion sensitive parameter identification, operation parameter and corrosion feature mapping relationship establishment, corrosion environment inversion model construction, regional corrosion environment parameter inversion, corrosion grade division and output. The application has the following advantages: significantly reduces the system cost and operation and maintenance burden of corrosion monitoring, improves data acquisition efficiency, and can realize corrosion environment evaluation of a large range of regions. The widely deployed air conditioner equipment is used as a sensing node, which can realize corrosion environment coverage evaluation of cities, parks and even larger scale regions. Continuous monitoring and dynamic updating are supported, the trend of corrosion environment change is continuously tracked, and the application has strong engineering popularization value.

Description

Technical Field

[0001] This invention relates to the fields of corrosion environment assessment and industrial big data analysis, and particularly to a method for regional corrosion environment inversion based on air conditioning operation data. Specifically, it is an indirect measurement method that utilizes operating parameters collected by existing air conditioning IoT systems to establish a mapping relationship between equipment response and environmental corrosivity, thereby achieving large-scale, low-cost corrosion environment assessment. Background Technology

[0002] Accurate identification of the corrosive environment is crucial for material service life assessment, corrosion protection design, and equipment operation and maintenance decisions. In air conditioning systems, the corrosion rate of heat exchangers and other metal components is directly influenced by factors such as temperature, humidity, salt spray concentration, and atmospheric pollutants in the surrounding environment. Therefore, accurate assessment of the corrosivity of the equipment's operating environment is essential for predicting equipment corrosion status and optimizing maintenance strategies.

[0003] In existing technologies, obtaining information about corrosive environments primarily relies on direct measurement methods. Common techniques include: deploying standard corrosion test pieces in the target area and periodically weighing them to obtain corrosion weight loss data, thereby estimating the corrosion rate; deploying electrochemical corrosion sensors to monitor parameters such as corrosion current or polarization resistance in real time; and using environmental monitoring equipment to measure environmental parameters such as temperature, relative humidity, and chloride ion deposition, and then indirectly assessing corrosivity based on empirical models. However, these methods have significant limitations in practical applications. First, the number of monitoring points is limited, making it difficult to cover large areas. Due to the cost of sensor deployment and maintenance, direct measurement methods typically only obtain corrosion data from a limited number of points, failing to reflect the distribution characteristics of the corrosive environment at a regional scale. Second, sensor deployment and maintenance costs are high. Corrosion sensors themselves are expensive, and in highly corrosive environments, they also face the problem of failure, requiring periodic calibration and replacement, resulting in high long-term operating costs. Third, it is difficult to achieve long-term continuous monitoring. Direct measurement methods are mostly intermittent sampling (such as the test piece method) or finite-life online monitoring, making it difficult to achieve continuous environmental tracking throughout the entire service life.

[0004] On the other hand, during operation, air conditioning equipment's IoT system continuously collects a large amount of environmental and operational status data, including inlet air temperature, outlet air temperature, relative humidity, compressor operating frequency, operating time, and system energy consumption. This data, to some extent, reflects the thermal and humidity characteristics and operating load of the environment in which the equipment operates. However, current technology has not yet established an effective method to deduce environmental corrosion characteristics from this existing equipment operating data, particularly lacking a systematic mapping relationship between equipment operating status and the corrosive environment. This valuable IoT data resource has not yet been used for environmental corrosion assessment, resulting in a waste of data value.

[0005] Therefore, there is an urgent need to provide a new technical solution that can indirectly acquire and dynamically assess the regional corrosive environment by utilizing the massive operational data of existing air conditioning equipment without increasing additional hardware investment. Summary of the Invention

[0006] The purpose of this invention is to overcome the technical shortcomings of existing technologies that rely on the deployment of additional sensors for obtaining corrosion environment data, which are costly and have limited coverage. This invention provides a method for regional corrosion environment inversion based on air conditioning operation data. This method establishes a mapping relationship between air conditioning operating parameters and material corrosion characteristics, and uses operating data collected by existing air conditioning IoT systems to infer the corrosion environment characteristics of the area where the equipment is located. This achieves the goal of assessing the corrosion environment of a large area without the need for additional corrosion monitoring devices.

[0007] To achieve the above-mentioned objectives, this invention provides a method for regional corrosion environment inversion based on air conditioning operation data. This method follows a technical route of data acquisition, preprocessing, sensitive parameter identification, mapping relationship establishment, inversion model construction, regional inversion, and graded output, specifically including the following steps: S1: Air Conditioner Operating Parameter Collection The system continuously collects multi-source data generated by the air conditioner during operation via its built-in IoT system or an external data acquisition terminal, and uploads this data to the data processing platform in real time. The air conditioner operating parameters include at least one or more of the following categories: Environmental condition parameters: These reflect the thermal and humidity characteristics of the environment in which the equipment is located, including but not limited to outdoor ambient temperature, outdoor relative humidity, indoor return air temperature, and indoor return air humidity. Operating condition parameters: These represent the operating status and load level of the equipment, including but not limited to compressor operating frequency, fan speed, running time, and number of start-stop cycles. Heat exchange process parameters: These reflect the heat exchange efficiency of the heat exchanger, including but not limited to the evaporator inlet and outlet air temperature difference, condenser inlet and outlet air temperature difference, refrigerant pressure, and refrigerant temperature. Electrical and control parameters: These reflect the energy consumption and control behavior of the equipment, including but not limited to compressor current, system input power, electronic expansion valve opening, and control mode.

[0008] The above parameters are collected by the Internet of Things system at a preset frequency (such as minute-level or hour-level) to form a continuous time series dataset.

[0009] S2: Data Preprocessing and Standardization The multi-source operating parameters collected in step S1 are preprocessed to eliminate noise and outliers, thereby improving data quality. The preprocessing includes: Noise reduction: Filtering algorithms (such as moving average filtering and median filtering) are used to eliminate sensor measurement noise; Missing value handling: For data missing due to communication interruption or sensor failure, methods such as linear interpolation, time series prediction or adjacent value filling are used to fill in the missing data. Time synchronization: Aligning data from different sources and with different sampling frequencies to a unified timeline to ensure data time consistency.

[0010] After preprocessing, the data is standardized to convert parameters with different dimensions and numerical ranges into uniform, dimensionless standardized values, which facilitates subsequent analysis.

[0011] S3: Corrosion-Sensitive Parameter Identification Based on corrosion mechanism analysis and statistical data methods, a set of sensitive parameters that significantly influence material corrosion behavior is selected from the operating parameters processed in step S2. The identification of these corrosion-sensitive parameters can be achieved through the following methods: Mechanism analysis and screening: Based on the principles of corrosion electrochemistry, identify environmental factors (such as temperature, humidity, and condensation time) that are closely related to the corrosion rate, as well as parameters that reflect the operating status of the equipment (such as operating time and start-up / shutdown frequency). Correlation analysis: Using Pearson correlation coefficient, Spearman rank correlation coefficient or mutual information method, calculate the correlation between each operating parameter and known corrosion data, and screen out parameters with significant correlation. Principal component analysis or feature importance assessment: Machine learning methods (such as random forest feature importance) are used to identify the operating parameters that have the greatest impact on the corrosion state.

[0012] The set of corrosion-sensitive parameters is the core input feature for the subsequent inversion model construction.

[0013] S4: Establishing the mapping relationship between operating parameters and corrosion characteristics A quantitative mapping relationship between air conditioning operating parameters and material corrosion characteristics is established using experimental or historical data. These corrosion characteristics include, but are not limited to, corrosion rate (corrosion weight loss or corrosion depth per unit time), corrosion morphology characteristics (such as pitting density, corrosion area ratio), or corrosion level (slight corrosion, moderate corrosion, severe corrosion). The mapping relationship can be established using the following methods: Laboratory accelerated corrosion test: Under controlled environmental conditions, accelerated corrosion tests are conducted on samples of different materials, and environmental parameters (temperature, humidity, etc.) and corrosion rate data are recorded simultaneously to establish a benchmark relationship model between environmental factors and corrosion rate. Data correlation of field exposure tests: Material exposure tests are conducted at natural exposure test stations, and operating parameters of air conditioning equipment near the stations are collected simultaneously to establish a statistical correlation model between operating parameters and measured corrosion data; Historical Operation and Maintenance Data Mining: By utilizing the historical operation and maintenance records of existing air conditioning equipment (such as corrosion detection results and maintenance records) and the operating data of the same period, a mapping relationship is established through data mining.

[0014] The mapping relationship can be expressed in the form of regression model, neural network or physical information machine learning model, so as to realize the transformation from the operating parameter space to the corrosion feature space.

[0015] S5: Construction of Corrosion Environment Inversion Model Based on the mapping relationship between operating parameters and corrosion characteristics established in step S4, a corrosion environment inversion model is constructed. The core function of this model is to indirectly measure environmental corrosivity by inferring it from the operating parameters of the air conditioning equipment as input and the corrosion characteristic parameters of the environment in which the equipment is located as output. The inversion model can take the following form: Parametric regression models, such as multiple linear regression and logistic regression, are suitable for scenarios where the mapping relationship is relatively linear. Machine learning models, such as support vector regression, random forest, gradient boosting tree, or deep neural networks, are suitable for complex nonlinear mapping relationships. Hybrid model: Combining physical mechanism model and data-driven model to improve the accuracy and generalization ability of inversion.

[0016] The corrosion environment inversion model outputs corrosion environment parameters that are not directly measured, such as equivalent corrosion rate, corrosion level index, or corrosion environment classification.

[0017] S6: Inversion of Regional Corrosion Environment Parameters The operating parameters of multiple air conditioning devices within the area to be evaluated are input into the inversion model constructed in step S5 to obtain the corrosion environment parameters at each device location. For regional-scale corrosion environment assessment, the following method can be used: Single-point inversion: Invert each device independently to obtain the corrosion environment parameters of the device's location; Regional interpolation: Based on the inversion results of multiple equipment points, spatial interpolation methods (such as Kriging interpolation and inverse distance weighted interpolation) are used to generate a corrosion environment distribution map of the entire region; Regional aggregation: Statistical analysis is performed on the inversion results of multiple devices within a region to obtain the statistical characteristics of the corrosion environment in that region (such as average value, maximum value, coefficient of variation).

[0018] S7: Corrosion Level Classification and Output Based on the corrosion environment parameters obtained from step S6, and combined with a preset corrosion level classification standard, the degree of regional corrosion is determined. The corrosion level classification standard can be preset according to material type, service requirements, or industry standards; for example, corrosion rates can be divided into multiple levels such as slight corrosion (<0.1 mm / year), moderate corrosion (0.1-0.5 mm / year), and severe corrosion (>0.5 mm / year). The final output is the regional corrosion environment assessment result, including corrosion level, corrosion parameter distribution map, and analysis of key sensitive parameters.

[0019] Advantages of this invention: This invention provides a method for regional corrosion environment inversion based on air conditioning operation data, which eliminates the need for additional corrosion sensors and significantly reduces system costs. This invention fully utilizes the existing Internet of Things system and operation data of air conditioning equipment to acquire and assess the corrosion environment without increasing any hardware investment, thus significantly reducing the system cost and maintenance burden of corrosion monitoring.

[0020] This invention leverages existing equipment to achieve environmental corrosion sensing, improving data acquisition efficiency: Air conditioning equipment is widely distributed and numerous, and its operational data can be collected in real time and continuously through an Internet of Things (IoT) system. This invention transforms these existing data resources into corrosion environment sensing capabilities, enabling in-depth mining of data value.

[0021] This invention enables comprehensive corrosion environment assessment over large areas: Traditional sensor deployment methods are limited by cost and maintenance, resulting in a limited number of monitoring points. This invention utilizes widely deployed air conditioning equipment as sensing nodes, enabling comprehensive corrosion environment assessment across cities, industrial parks, and even larger areas.

[0022] Supports continuous monitoring and dynamic updates: Air conditioning operation data is collected continuously in real time, and the corrosion environment assessment results based on the method of this invention can be dynamically refreshed as the operation data is updated, so as to continuously track the trend of corrosion environment changes.

[0023] It has good scalability and engineering application value: This invention does not depend on specific models or brands of air conditioning equipment. As long as it has basic operational data collection capabilities, it can be applied. It is suitable for large-scale deployment of various air conditioning systems and has strong engineering promotion value. Attached Figure Description

[0024] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments: Figure 1 This is a schematic diagram of the overall process of the regional corrosion environment inversion method based on air conditioning operation data described in this invention, which shows the complete technical route from data acquisition to corrosion level output. Figure 2This is a schematic diagram illustrating the mapping relationship between operating parameters and corrosion characteristics described in this invention, demonstrating the process of constructing a mapping model through laboratory experiments and data correlation. Figure 3 This is a schematic diagram of the regional corrosion environment inversion described in this invention, illustrating the process of inputting the operating data of multiple air conditioning devices into the inversion model and generating a regional corrosion environment distribution map through spatial interpolation. Detailed Implementation

[0025] The present invention will be further explained below with reference to specific implementation schemes, but it is not limited to the present invention. The structures, proportions, sizes, etc. shown in the accompanying drawings are only used to complement the content disclosed in the specification, so as to enable those skilled in the art to understand and read, and are not intended to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any modification of the structure, change of the proportion relationship or adjustment of the size, without affecting the effect and purpose that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0026] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below with reference to specific embodiments.

[0027] Example 1: Inversion of Corrosion Environment in a Single Region S1: Air Conditioner Operating Parameter Collection In a coastal industrial park, 30 outdoor air conditioning units located in different positions were selected as sensing nodes. Operating data was continuously collected for one year through an air conditioning IoT platform, with a collection frequency of once per hour. The collected parameters included: outdoor ambient temperature, outdoor relative humidity, compressor operating frequency, system runtime, and condenser inlet and outlet air temperature difference.

[0028] S2: Data Preprocessing and Standardization The collected data underwent preprocessing. Moving average filtering was used to eliminate sensor noise; for the approximately 3% data loss due to network interruption, linear interpolation was used to fill in the missing data; all parameters were aligned according to timestamps. Subsequently, the min-max normalization method was used to transform each parameter to the [0,1] interval.

[0029] S3: Corrosion-Sensitive Parameter Identification Based on corrosion mechanism analysis and correlation analysis, sensitive parameters that significantly affect aluminum fin corrosion were selected. The analysis results showed that the cumulative condensation time (estimated by the combination of temperature and humidity), average relative humidity, and average daily compressor operating time had the highest correlation with the known corrosion rate and were selected as the set of sensitive parameters.

[0030] S4: Establishing the mapping relationship between operating parameters and corrosion characteristics Under laboratory conditions, a series of accelerated corrosion experiments were conducted on 3003 aluminum alloy samples. Different temperature and humidity combinations and wet-dry cycle periods were controlled. The corrosion weight loss and corrosion current density of the samples under each condition were measured, and a baseline relationship model between environmental parameters and corrosion rate was established. Simultaneously, a correlation analysis was performed using historical exposure test data (standard aluminum test pieces already deployed) from the industrial park and concurrent air conditioning operation data to establish a statistical mapping model between operating parameters (cumulative condensation duration, average relative humidity, and operating time) and measured corrosion rate. A multivariate nonlinear regression method was used to fit the mapping equation.

[0031] S5: Construction of Corrosion Environment Inversion Model Based on the mapping relationship established in step S4, a corrosion environment inversion model is constructed. The model input consists of sensitive parameters (cumulative condensation duration, average relative humidity, and average daily compressor operating time), and the output is the equivalent corrosion rate (unit: mm / year). The model adopts an artificial neural network structure with 3 neurons in the input layer, 8 neurons in the hidden layer, and 1 neuron in the output layer. It is jointly trained using laboratory experimental data and field exposure data.

[0032] S6: Inversion of Regional Corrosion Environment Parameters The annual operating data of 30 air conditioning units were input into the inversion model to obtain the equivalent corrosion rate at each unit's location. The calculation results show that the average equivalent corrosion rate for units near the coastline is 0.42 mm / year, for units far from the coastline it is 0.18 mm / year, and in the central area of ​​the park it is 0.28 mm / year.

[0033] S7: Corrosion Level Classification and Output Based on the preset corrosion level classification standards (slight corrosion <0.1 mm / year, moderate corrosion 0.1-0.3 mm / year, severe corrosion >0.3 mm / year), the coastal area of ​​the industrial park is classified as severely corroded, the central area as moderately corroded, and the inland area as moderately corroded. A corrosion level distribution map of the output areas is provided as a basis for differentiated operation and maintenance of air conditioning equipment.

[0034] Example 2: Comparative Evaluation of Corrosion Environments in Multiple Regions S1 to S2: Data Acquisition and Preprocessing Three regions with different characteristics were selected: Region A (arid inland region, annual average relative humidity 55%), Region B (humid inland region, annual average relative humidity 75%), and Region C (high humidity coastal region, annual average relative humidity 85%, high salt spray). Twenty air conditioning units of the same model were selected in each region, and one year of operating data was collected and preprocessed.

[0035] S3 to S5: Sensitive parameter identification and inversion model construction Using the same sensitive parameter identification method and inversion model construction process as in Example 1, a unified inversion model is established.

[0036] S6: Inversion of Regional Corrosion Environment Parameters The operational data of 60 devices across three regions were input into the inversion model to obtain the statistical results of the equivalent corrosion rate for each region: • Region A: Average equivalent corrosion rate 0.08 mm / year, maximum 0.12 mm / year; • Region B: Average equivalent corrosion rate 0.22 mm / year, maximum 0.31 mm / year; • Region C: Average equivalent corrosion rate 0.45 mm / year, maximum 0.58 mm / year.

[0037] S7: Corrosion Level Classification and Output Based on corrosion level standards, the following classifications were applied: Area A was classified as slightly corroded; Area B was classified as moderately corroded, with some high-humidity areas approaching the level of severe corrosion; and Area C was classified as severely corroded. A corrosion level comparison report for the three areas was generated, which largely matched the results of the field survey (based on measurements using standard test pieces), verifying the effectiveness of the method described in this invention.

[0038] Technical effect verification In Examples 1 and 2, the equivalent corrosion rate obtained by the inversion method of the present invention was compared with the measured corrosion weight loss data of standard aluminum specimens placed at the same location. The average relative error was 12.8%, indicating that the method of the present invention has good accuracy. Furthermore, compared with traditional sensor deployment methods, the present invention requires no additional hardware investment and can complete the evaluation using only existing air conditioning IoT data, significantly reducing the cost and implementation difficulty of obtaining corrosive environment data.

[0039] Matters not covered in this invention are common knowledge.

[0040] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for regional corrosion environment inversion based on air conditioning operation data, characterized in that, Includes the following steps: S1: Collect air conditioning operating parameters through the Internet of Things (IoT) system of the air conditioning equipment. The air conditioning operating parameters include at least one of the following: environmental status parameters, operating condition parameters, heat exchange process parameters, or electrical and control parameters. S2: Perform data preprocessing and standardization on the collected air conditioning operating parameters. The preprocessing includes noise reduction, missing value processing, and time synchronization. S3: Based on corrosion mechanism analysis and data statistics methods, a set of corrosion-sensitive parameters that have a significant impact on the corrosion behavior of materials are identified and screened from the pre-processed operating parameters; S4: Establish a mapping relationship between air conditioning operating parameters and material corrosion characteristics by correlating data from laboratory accelerated corrosion experiments, field exposure tests, or historical operation and maintenance data mining. The corrosion characteristics include corrosion rate, corrosion morphology, or corrosion level. S5: Based on the mapping relationship established in step S4, construct a corrosion environment inversion model. The inversion model takes air conditioning operating parameters as input and non-directly measured corrosion environment parameters as output. S6: Input the operating parameters of multiple air conditioning devices in the area to be evaluated into the inversion model, invert to obtain the corrosion environment parameters of each device location, and obtain the regional scale corrosion environment characteristics through spatial interpolation or regional aggregation. S7: Based on the corrosion environment parameters obtained from the inversion and combined with the preset corrosion level classification standards, determine the level of regional corrosion and output the evaluation results.

2. The regional corrosion environment inversion method based on air conditioning operation data according to claim 1, characterized in that, The environmental status parameters mentioned in step S1 include outdoor ambient temperature, outdoor relative humidity, indoor return air temperature or indoor return air humidity; the operating condition parameters include compressor operating frequency, fan speed, running time or number of start-stop cycles. The heat exchange process parameters include the evaporator inlet and outlet air temperature difference, the condenser inlet and outlet air temperature difference, and the refrigerant pressure or refrigerant temperature; the electrical and control parameters include the compressor current, system input power, electronic expansion valve opening degree, or control mode.

3. The regional corrosion environment inversion method based on air conditioning operation data according to claim 1, characterized in that, The denoising process in step S2 uses moving average filtering or median filtering, the missing value processing uses linear interpolation, time series prediction or adjacent value filling, and the time synchronization aligns data from different sampling frequencies to a unified time axis.

4. The regional corrosion environment inversion method based on air conditioning operation data according to claim 1, characterized in that, The corrosion-sensitive parameter identification in step S3 uses Pearson correlation coefficient, Spearman rank correlation coefficient, mutual information method or random forest feature importance method to calculate the correlation between each operating parameter and corrosion data, and screens out parameters with significant correlation as the sensitive parameter set.

5. The regional corrosion environment inversion method based on air conditioning operation data according to claim 1, characterized in that, The laboratory accelerated corrosion experiment described in step S4 involves conducting corrosion experiments on material samples under controlled environmental conditions, simultaneously recording environmental parameters and corrosion rate data, and establishing a benchmark relationship model between environmental factors and corrosion rate. The field exposure test data correlation involves conducting material exposure tests at a natural exposure test station, while simultaneously collecting operating parameters of air conditioning equipment near the station, and establishing a statistical correlation model between operating parameters and measured corrosion data.

6. The regional corrosion environment inversion method based on air conditioning operation data according to claim 1, characterized in that, The inversion model described in step S5 uses multiple linear regression, support vector regression, random forest, gradient boosting tree, deep neural network or physical information machine learning model to realize the transformation from the operating parameter space to the erosion feature space.

7. The regional corrosion environment inversion method based on air conditioning operation data according to claim 1, characterized in that, The spatial interpolation in step S6 uses Kriging interpolation or inverse distance weighted interpolation to generate a corrosion environment distribution map of the region; the regional aggregation includes calculating the average value, maximum value, or coefficient of variation of corrosion environment parameters within the region.

8. The regional corrosion environment inversion method based on air conditioning operation data according to claim 1, characterized in that, The corrosion level classification standard mentioned in step S7 divides the corrosive environment into multiple levels such as slight corrosion, moderate corrosion and severe corrosion based on material type, service requirements or industry standard preset.

9. The regional corrosion environment inversion method based on air conditioning operation data according to claim 1, characterized in that, The method utilizes the existing Internet of Things system and operational data of the air conditioning equipment to indirectly acquire and dynamically assess the regional corrosion environment without increasing additional hardware investment.

Images