System and method for controlling running state of screw compressor based on artificial intelligence

Abstract

The invention relates to the technical field of intelligent control of refrigeration equipment, in particular to an artificial intelligence-based screw compressor running state control system and method, which comprises a signal acquisition unit, a disturbance identification unit and a compressor control unit, the signal acquisition unit acquires a refrigeration house door opening and closing state time sequence signal and three-dimensional temperature field change data, the disturbance identification unit synchronously locks the door opening moment through a timestamp, divides a core influence area and a stable conduction area, analyzes the correlation between the temperature change rate and the door opening angle and the temperature fluctuation amplitude, and determines the temperature change rate of the door opening angle. And the door opening disturbance or equipment abnormity identification factor is output, and the compressor control unit executes differential regulation according to the door opening disturbance or equipment abnormity identification factor, so that the problems of disturbance misjudgment and regulation rigidity in traditional control are solved, and the operation stability and energy consumption control precision of the compressor are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control of refrigeration equipment, and in particular to an artificial intelligence-based screw compressor operating state control system and method. Background Art

[0002] Intelligent control of refrigeration equipment is an important technology. In refrigeration systems such as cold storage and cold chain logistics, screw compressors are core equipment. The stability of their operating status directly affects refrigeration efficiency and energy consumption control. Intelligent control systems are key means to ensure high efficiency and energy saving of refrigeration systems and extend equipment life by sensing changes in operating conditions in real time and dynamically adjusting operating parameters.

[0003] However, traditional screw compressor control technology suffers from insufficient disturbance identification accuracy and rigid control strategies. Existing solutions rely solely on single-point temperature sensor data. When the cold storage door frequently opens and closes, it cannot distinguish between temperature fluctuations caused by door opening and temperature changes caused by equipment anomalies. Door opening disturbances are mistakenly identified as equipment failures, triggering unnecessary frequency reduction protection, resulting in a sudden drop in cooling capacity and a deviation of the internal temperature from the set range. Furthermore, only a fixed threshold is used to trigger control instructions. When the door opening angle is small, the compressor still operates at maximum load, resulting in energy waste. When the equipment is truly abnormal, the response lag causes the fault to expand. In addition, traditional systems lack spatial partitioning analysis, resulting in inaccurate judgment of the temperature field disturbance range. Control instructions are out of sync with actual demand, exacerbating the frequent start-stop of the compressor. This not only shortens equipment life, but also reduces the temperature control accuracy of the cold storage, affecting the quality of stored goods, ultimately leading to increased operation and maintenance costs and reduced system reliability. To address this problem, we provide an artificial intelligence-based screw compressor operating status control system and method. Summary of the Invention

[0004] The purpose of the present invention is to provide an artificial intelligence-based screw compressor operating state control system and method to solve the problems raised in the above background technology.

[0005] 1. Traditional disturbance recognition technology lacks precision and cannot distinguish between door opening disturbances and equipment anomalies, leading to misjudgments and misadjustments. Therefore, in this case, a signal acquisition unit is used to obtain door status and three-dimensional temperature field data. The disturbance recognition unit analyzes the correlation and fluctuation amplitude and outputs the corresponding identification factor, which can accurately identify the disturbance type and reduce misoperation.

[0006] 2. Because traditional control strategies are rigid, use fixed thresholds, and lack zoning analysis, resulting in high energy consumption or delayed fault response, this case divides the compressor control unit into core and stable areas, performing differentiated control to adapt to actual operating conditions, save energy, and handle anomalies in a timely manner.

[0007] To achieve the above objectives, an artificial intelligence-based screw compressor operating status control system is provided. The system includes a signal acquisition unit that acquires the real-time timing signals of the cold storage door's opening and closing status, as well as the three-dimensional temperature field change data generated by the infrared temperature sensor array deployed on the door's inner facade, and executes the following calculation processes in sequence: The disturbance identification unit synchronizes the time stamps of the opening and closing state timing signal with the three-dimensional temperature field change data, locks the triggering moment of the door opening event, divides the core influence area and the stable conduction area with the cold storage door as the center, calculates the average temperature change rate of the core influence area and the stable conduction area within the time window, analyzes the correlation between the average temperature change rate of the core influence area and the opening angle of the cold storage door, obtains the correlation coefficient, and synchronously detects the temperature fluctuation amplitude value of the stable conduction area at the triggering moment of the door opening event. The correlation coefficient and the temperature fluctuation amplitude value are analyzed and compared with their corresponding thresholds, and the comparison results are output. According to the comparison results, the output door opening disturbance identification factor or equipment abnormality identification factor is judged; The compressor control unit performs differentiated regulation on the screw compressor according to the door opening disturbance identification factor or the equipment abnormality identification factor.

[0008] As a further improvement of this technical solution, the opening and closing state timing signal is synchronized with the three-dimensional temperature field change data by time stamp, and the operation of locking the door opening event triggering moment is specifically comprised of the following steps: The door magnetic sensor is used to capture the continuous change trajectory of the cold storage door displacement and generate a displacement time series curve. At the same time, the temperature change gradient data of each detection unit in the infrared temperature sensor array is extracted to generate three-dimensional temperature field change data aligned with the time axis of the displacement time series curve. When the displacement of the cold storage door increases and exceeds the preset displacement threshold within the specified time, it is marked as the potential door opening starting point. The multi-point temperature gradient change trend of the cold storage door area in the three-dimensional temperature field change data is analyzed and compared with the preset judgment rules. The temperature mutation node is marked according to the comparison result. Finally, the time overlap between the potential door opening starting point and the temperature mutation node of the displacement timing curve is cross-validated to confirm the triggering moment of the effective door opening event.

[0009] As a further improvement of this technical solution, a spatial zoning method is used to divide the core influence area and the stable conduction area with the cold storage door as the center, specifically including: Taking the geometric center of the cold storage door as the origin, a three-dimensional fan-shaped space area perpendicular to the door opening direction is constructed as the core influence area. Its radial extension distance is dynamically adjusted according to the volume of the cold storage, and a layered detection grid is set in the core influence area. A spherical space area is delineated with the evaporator unit as the center, which is used as the stable conduction zone. Its radius is adaptively adjusted according to the current operating power of the screw compressor. A transition buffer zone is set at the junction of the stable conduction zone and the core influence zone to suppress the boundary effect error of the temperature field analysis.

[0010] As a further improvement of this technical solution, the operation of calculating the average temperature change rate of the core influence area and the stable conduction area within the time window specifically includes: Sampling points are selected in the layered detection grid of the core impact area. Abnormal temperature data are eliminated according to the sampling points to form valid sampling points. The first-order derivative of the temperature change curve of the valid sampling points is calculated to extract the instantaneous value of the temperature change rate of each valid sampling point. Finally, the weighted average algorithm is used to fuse the instantaneous values ​​of the temperature change rate of each valid sampling point to obtain the average temperature change rate of the core impact area. The full-space temperature field integral operation is performed in the stable conduction zone to generate the overall temperature change trend function. The overall temperature change trend function is smoothed, and its linear fitting slope in the time window is extracted. The linear fitting slope is used as the average temperature change rate in the stable conduction zone.

[0011] As a further improvement of this technical solution, the steps of analyzing the correlation between the average temperature change rate of the core impact area and the opening angle of the cold storage door and obtaining the correlation coefficient include: The real-time opening angle is calculated by integrating the cold storage door displacement time series curve, and an angle-time change relationship table is established. According to the angle-time change relationship table, the angle change curve is subjected to piecewise linear interpolation processing to generate an angle time series sequence; The time series of the average temperature change rate in the core influence area is extracted and input into the dynamic time warping algorithm together with the angle time series to eliminate the time series offset caused by the physical response delay between the two. The covariance matrix is ​​obtained, and the normalized cumulative value of the main diagonal elements in the covariance matrix is ​​extracted as the correlation coefficient.

[0012] As a further improvement of the present technical solution, the operation of synchronously detecting the temperature fluctuation amplitude value of the stable conduction zone at the time of triggering the door opening event specifically includes: During the preset time period before the door opening event is triggered, the three-dimensional temperature field historical data of the stable conduction area is collected, and the theoretical temperature field distribution is predicted by the autoregressive model. The measured temperature field data after the door opening event is triggered are compared point by point with the theoretical temperature field, and the absolute value of the temperature deviation at each spatial point is calculated. The absolute value of the temperature deviation in the entire area is statistically analyzed, and the absolute value of the temperature deviation at the preset quantile is taken as the temperature fluctuation amplitude value.

[0013] As a further improvement to this technical solution, the correlation coefficient and the temperature fluctuation amplitude values ​​are analyzed and compared with their corresponding thresholds, specifically including: According to the current operating mode of the cold storage, the preset threshold reference library is called to obtain the initial threshold parameters. The threshold parameters are corrected in real time in combination with the compressor load rate and the temperature difference between the inside and outside of the cold storage. The first threshold and the second threshold are set according to the corrected threshold parameters. The correlation coefficient and the temperature fluctuation amplitude value are judged according to the first threshold and the second threshold respectively, and the correlation coefficient judgment result and the temperature fluctuation amplitude judgment result are obtained.

[0014] As a further improvement of the present technical solution, the output method of the comparison result in the disturbance identification unit specifically includes: The correlation coefficient determination result and the temperature fluctuation amplitude determination result are combined into a four-digit status code. The first two digits represent the correlation strength, and the last two digits represent the temperature fluctuation amplitude level. The corresponding semantic description text is queried in the preset mapping table based on the four-digit status code. A matching search is performed in the credibility knowledge base based on the semantic description text to generate a confidence score. The confidence score is reversely calculated based on the misjudgment rate of similar four-digit status codes in historical data, and the confidence score is used as the output comparison result.

[0015] As a further improvement to the present technical solution, a method for performing differentiated control on a screw compressor according to a door opening disturbance identification factor or an equipment abnormality identification factor is as follows: When the door opening disturbance identification factor is output, the triggering time of the compressor frequency reduction instruction is delayed. The delay time is dynamically adjusted according to the temperature fluctuation amplitude value, and the defrost program execution is frozen until the temperature field returns to the preset safety range of the stable conduction zone; When the output device abnormality identification factor is detected, the compressor frequency reduction protection is immediately activated and the oil system pressure is simultaneously increased to a safe threshold. An early warning signal containing an abnormality location code is generated and uploaded to the remote operation and maintenance platform through the Internet of Things module.

[0016] A second object of the present invention is to provide a method for implementing an artificial intelligence-based screw compressor operating state control system according to any one of the above, comprising the following steps: S1. The door magnetic sensor captures the cold storage door displacement timing signal in real time, and simultaneously drives the infrared temperature sensor array to collect dynamic three-dimensional temperature field data on the inner facade of the door. Based on the spatiotemporal coupling relationship between the sudden increase in displacement and the radial temperature drop pattern in the door area, the triggering moment of the effective door opening event is verified and locked, completing the initial perception of the cold storage disturbance event. S2. Based on the door opening trigger moment, sector modeling of the core impact area and spherical spatial division of the stable conduction area are performed. The anti-interference weighted average temperature change rate of the two areas within the asynchronous time window is calculated. The hysteresis correlation between the core area temperature change rate and the door opening angle is analyzed using the dynamic time warping algorithm. The temperature fluctuation amplitude value of the stable conduction area is simultaneously detected. Combined with the dynamic correction threshold, a multi-dimensional comparison decision is made, and a four-bit status code is generated with an additional confidence score. Finally, the door opening disturbance identification factor or equipment abnormality identification factor is output; S3. When the door opening disturbance identification factor is output, the compressor frequency reduction response time is delayed and the defrost program is frozen until the temperature in the stable conduction zone returns to the preset safety threshold. When the device abnormality identification factor is output, the compressor frequency reduction protection is immediately triggered and the oil circuit pressure is increased to the safety critical value. At the same time, an early warning signal with positioning code is generated and uploaded to the remote operation and maintenance platform to locate and intervene in the fault.

[0017] Compared with the prior art, the present invention has the following beneficial effects: The present invention synchronously obtains the opening and closing status of the cold storage door and the three-dimensional temperature field data through the signal acquisition unit, accurately locks the door opening event through the disturbance identification unit, divides the core influence area and the stable conduction area, and analyzes the temperature change characteristics, effectively distinguishes the door opening disturbance from the equipment abnormality, and solves the problem of disturbance misjudgment in traditional control. The compressor control unit performs differentiated regulation according to different identification factors, dynamically adjusts the compressor frequency reduction delay time and freezes the defrost program for the door opening disturbance, immediately activates the protection mechanism for the equipment abnormality and uploads the early warning, realizes the precise control of the working condition adaptation, ensures the temperature stability of the cold storage, reduces the operation and maintenance cost, and is suitable for cold chain logistics and other scenarios with high requirements on the reliability of the refrigeration system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is an overall block diagram of the present invention; Figure 2 It is the overall flow chart of the present invention.

[0019] The meaning of each number in the figure is: 1. Signal acquisition unit; 2. Disturbance identification unit; 3. Compressor control unit. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0021] The present invention provides a screw compressor operating state control system based on artificial intelligence, please refer to Figure 1 As shown, the signal acquisition unit 1 acquires the opening and closing status timing signal of the cold storage door in real time and the three-dimensional temperature field change data generated by the infrared temperature sensor array deployed on the inner side of the door, and executes the following calculation process in sequence: The disturbance identification unit 2 synchronizes the time stamps of the opening and closing state timing signal with the three-dimensional temperature field change data, locks the triggering moment of the door opening event, divides the core influence area and the stable conduction area with the cold storage door as the center, calculates the average temperature change rate of the core influence area and the stable conduction area within the time window, analyzes the correlation between the average temperature change rate of the core influence area and the opening angle of the cold storage door, obtains the correlation coefficient, and synchronously detects the temperature fluctuation amplitude value of the stable conduction area at the triggering moment of the door opening event. The correlation coefficient and the temperature fluctuation amplitude value are analyzed and compared with their corresponding thresholds, and the comparison results are output. According to the comparison results, the output door opening disturbance identification factor or the equipment abnormality identification factor is judged; The compressor control unit 3 performs differentiated regulation on the screw compressor according to the door opening disturbance identification factor or the equipment abnormality identification factor.

[0022] In order to accurately lock the actual triggering moment of the door opening event and ensure the temporal correlation between the subsequent temperature field analysis and the door status, it is necessary to synchronize the time stamps of the opening and closing status timing signals with the three-dimensional temperature field change data. The operation of locking the triggering moment of the door opening event includes the following steps: The door magnetic sensor is used to capture the continuous change trajectory of the cold storage door displacement to generate a displacement timing curve. Because the door magnetic sensor can sense the physical position change of the door in real time, the electrical signal it outputs can directly reflect whether the door is open and the degree of opening. The door magnetic sensor is installed at the joint between the door and the door frame. When the door moves, the voltage signal output by the sensor changes with the displacement. The system collects the signal every 10 milliseconds, converts the voltage value into displacement, and arranges it in chronological order to form a displacement timing curve. The horizontal axis of the curve is time and the vertical axis is displacement. It fully records the entire process of the door from closing to opening or from opening to closing. At the same time, the temperature change gradient of each detection unit in the infrared temperature sensor array is extracted. The temperature data is used to generate three-dimensional temperature field change data aligned with the time axis of the displacement timing curve. The infrared temperature sensor array is evenly deployed on the inner facade of the door. Each sensor collects a temperature value every 50 milliseconds. The temperature change gradient data is obtained by calculating the difference between two adjacent collected temperatures, such as the current temperature minus the previous temperature. A positive value indicates a temperature increase, and a negative value indicates a temperature decrease. These gradient data are arranged according to the spatial position of the sensor and the collection time to form three-dimensional temperature field change data. Through a unified timestamp, it is consistent with the time axis of the displacement timing curve to achieve time alignment between the two, ensuring that the door displacement change and the temperature field change correspond to each other in time, laying the foundation for subsequent correlation analysis. When the displacement of the cold storage door increases by more than the preset displacement threshold within the specified time, it is marked as a potential door opening starting point. Because the sudden increase in displacement is a direct physical characteristic of door opening, the specified time is set to 1 second. Based on the normal speed setting of door opening, slow displacement is avoided from being misjudged. The preset displacement threshold is set to 5 cm, that is, the door movement of more than 5 cm is considered to have started to open. The system monitors the displacement timing curve in real time. When the displacement difference between two consecutive collection points with an interval of 10 milliseconds exceeds 5 cm within 1 second, the starting time of this time period is marked as the potential door opening starting point. For example, if the door moves from 0 cm to 6 cm in 0.8 seconds, the first collection point in this time period is marked as the potential door opening starting point. Possible door opening actions are preliminarily screened out, and the temperature gradient change trend of multiple points in the cold storage door area in the three-dimensional temperature field change data is analyzed and compared with the preset judgment rules. The temperature mutation node is marked according to the comparison result. Because the opening of the door will cause external hot air to enter, causing the temperature near the door to rise rapidly, forming a characteristic temperature mutation. The cold storage door area refers to the spatial range with the door as the center and a radius of 1 meter. The temperature change gradients of all sensors in the area are extracted from the three-dimensional temperature field change data, and their change trends are analyzed. The preset judgment rule is "the temperature gradients of three consecutive collection points are all positive, and the cumulative temperature rise exceeds 2 degrees Celsius". When the temperature gradient change at a certain moment meets this rule, the moment is marked as a temperature mutation node. For example, the temperature gradients collected three times by the sensor in the door area within 150 milliseconds are 0.8℃, 1.0℃, and 0.9℃ respectively, and the cumulative temperature rise is 2.7℃. The first collection point in this time period is marked as a temperature mutation node to capture the temperature field change characteristics caused by opening the door. Finally, the potential door opening start point is cross-validated by the displacement timing curve. The time overlap between the point and the temperature mutation node is confirmed as the effective door opening event triggering moment. Because a single displacement change or temperature change may be caused by other factors, the two need to be verified with each other. The judgment standard of time overlap is "the time difference between the two nodes does not exceed 500 milliseconds", that is, the timestamp difference between the potential door opening starting point and the temperature mutation node is within 0.5 seconds. If this condition is met, the earlier moment of the two nodes is confirmed as the effective door opening event triggering moment. If the time difference exceeds 500 milliseconds, it is judged as an invalid event. For example, the displacement change is the vibration of the door body and the temperature change is the heat dissipation of the equipment. The triggering moment is not marked, which provides an accurate time benchmark for the subsequent division of temperature influence areas and analysis of disturbance characteristics, avoiding the control deviation caused by single signal misjudgment.

[0023] In order to accurately analyze the different impact ranges of door opening events on the cold storage temperature field and subsequently calculate the temperature change rate in a targeted manner, a spatial partitioning method is required to divide the core impact area and the stable conduction area with the cold storage door as the center. Specifically, the following methods are used: First, determine the spatial range of the core impact zone. When the door is opened, the external hot air first impacts the area near the door. The temperature change in this area is the most direct and drastic, and needs to be monitored. Take the geometric center of the cold storage door as the origin, that is, the intersection of the midpoint of the door height and the midpoint of the width, and construct a three-dimensional fan-shaped spatial area perpendicular to the door opening direction. Use it as the core impact zone. The two sides of the three-dimensional fan are respectively expanded to both sides along the door opening direction. The expansion angle is set to 120 degrees, covering the main range where hot air may diffuse after the door is opened. The thickness of the fan is consistent with the thickness of the door, and its radial extension distance is dynamically adjusted according to the volume of the cold storage. The specific adjustment rules are as follows: : For every 100 cubic meters increase in cold storage volume, the radial extension distance increases by 0.5 meters. For example, for a cold storage with a volume of 200 cubic meters, the radial extension distance is 1 meter, and for a cold storage with a volume of 300 cubic meters, the radial extension distance is 1.5 meters. This ensures that the core impact area can completely cover the space directly affected by the door opening. At the same time, a layered detection grid is set up in the core impact area, divided into 5 layers at a spacing of 0.2 meters along the radial direction. Virtual detection points are arranged in each layer at a grid density of 0.3 meters × 0.3 meters. Each virtual detection point corresponds to the nearest actual sensor data in the infrared temperature sensor array. Through grid refinement, fine-grained capture of temperature changes in the core area is achieved; Next, the spatial scope of the stable conduction zone is defined. Because the evaporator unit is the core equipment for cold storage temperature control, the temperature changes in the area near it can better reflect the overall temperature stability of the cold storage. Therefore, it needs to be analyzed separately from the core influence zone. A spherical spatial area is delineated with the geometric center of the evaporator unit as the center. This area serves as the stable conduction zone. Its radius is adaptively adjusted according to the current operating power of the screw compressor. The adjustment rule is: for every 10 kilowatt increase in compressor operating power, the radius increases by 0.3 meters. Because higher power increases the evaporator's cooling influence range, the stable conduction zone needs to be expanded accordingly to accurately reflect its surrounding temperature state. A transition buffer zone is set at the junction of the stable conduction zone and the core influence zone to suppress boundary effect errors in temperature field analysis. Because the temperature field change characteristics of the two regions are different, analysis errors are prone to occur at the junction. A smooth transition is required through a buffer zone. The width of the transition buffer zone is fixed at 0.2 meters, covering an area 0.1 meters outward from the boundaries of the two regions. The temperature data within this area is then processed using weighted averaging in subsequent analysis to suppress boundary effect errors in temperature field analysis caused by regional boundary demarcation and ensure more consistent temperature change analysis in the two regions.

[0024] In order to accurately reflect the temperature change speed of different areas after the door opening event and provide data support for the disturbance type judgment, it is necessary to calculate the average temperature change rate of the core impact area and the stable conduction area within the time window respectively. Specifically, the following operations are required: First, calculate the average temperature change rate of the core impact zone, and select sampling points in the layered detection grid of the core impact zone. Because the virtual detection points in the grid already correspond to the actual sensor positions, selecting sampling points can reduce the amount of data processing and retain key information. The selection rule is to evenly select 10 sampling points for each layer of grid, giving priority to positions close to the door body and grid vertices to ensure that different areas of the core impact zone are covered. Abnormal temperature data are eliminated according to the sampling points to form valid sampling points. Abnormal temperature data refers to values ​​with a temperature difference of more than 5 degrees Celsius from other sampling points in the same layer, or values ​​that have not changed for three consecutive collections. The remaining sampling points after elimination are valid sampling points to ensure that the data used for calculation is true and reliable. The first-order derivative of the temperature change curve of the valid sampling points is calculated to extract the instantaneous value of the temperature change rate of each valid sampling point. Finally, the weighted average algorithm is used to fuse the instantaneous value of the temperature change rate of each valid sampling point to obtain the average temperature of the core impact zone. Average temperature change rate. The temperature change curve is formed by connecting the temperature values ​​of each valid sampling point within the time window, such as the temperature value 1 minute after the door opening event is triggered, in chronological order. The first-order derivative calculation is to divide the temperature difference between two adjacent time points by the time interval to obtain the instantaneous value of the temperature change rate at each time point, reflecting the speed of temperature change at that moment. Finally, the weighted average algorithm is used to fuse the instantaneous value of the temperature change rate of each valid sampling point to obtain the average temperature change rate of the core influence area. The instantaneous value of the temperature change rate of each sampling point is multiplied by the corresponding weight and added, and then divided by the sum of the weights to obtain the average temperature change rate of the core influence area within the time window, highlighting the dominant role of temperature changes near the door body. Then the average temperature change rate of the stable conduction zone is calculated, and the full-space temperature field integral operation is performed in the stable conduction zone to generate the overall temperature change trend function. The full-space temperature field integral operation is to summarize the temperature data collected by all infrared temperature sensors in the stable conduction zone, and average them according to the spatial position to obtain the overall temperature value of each time point. These overall temperature values ​​are then arranged in chronological order to form a trend function that reflects the change of the overall temperature of the stable conduction zone over time, reflecting the overall temperature change characteristics of the region. The overall temperature change trend function is smoothed by the sliding average method. The temperature values ​​of 5 consecutive time points are grouped together to calculate The average value is used as the temperature value at the intermediate time point to eliminate the interference of short-term fluctuations on the trend, making the curve smoother and facilitating subsequent analysis. The linear fitting slope within the time window is extracted and the linear fitting slope is used as the average temperature change rate of the stable conduction zone. Linear fitting is to draw a straight line that is closest to the smoothed trend function. The inclination of the straight line reflects the average temperature change rate of the stable conduction zone within the time window. A positive slope indicates heating, and a negative slope indicates cooling. It intuitively presents the speed of the overall temperature change in the area, providing accurate quantitative data for the subsequent analysis of the correlation between the temperature change rate and the door opening angle and the temperature fluctuation amplitude.

[0025] In order to clarify the correlation between the temperature change rate of the core influence area and the door opening angle, and thus determine whether the temperature fluctuation is caused by the door opening action, it is necessary to analyze the correlation between the average temperature change rate of the core influence area and the cold storage door opening angle. The steps to obtain the correlation coefficient include: The real-time opening angle is calculated by integrating the displacement timing curve of the cold storage door body, and an angle-time change relationship table is established. The displacement timing curve records the displacement of the door body at every moment from closing to opening. The method of integrating and calculating the real-time opening angle is: convert the displacement into an angle value. When the door body is fully closed, the angle is 0 degrees. When it is fully open, such as when the door body can rotate 180 degrees, the angle is 180 degrees. The corresponding angle is converted according to the ratio of the actual displacement of the door body to the maximum displacement. For example, if the maximum displacement of the door body is 1 meter and the current displacement is 0.5 meters, the opening angle is 90 degrees. After that, the corresponding opening angle of each moment is recorded in chronological order to form an angle-time change relationship table. The table contains two columns of information, one is time and the other is the time point The opening angle of the door body is clearly presented as a function of time. The angle change curve is subjected to piecewise linear interpolation according to the angle-time change relationship table to generate an angle time series. The angle change curve is a curve that plots the data in the angle-time change relationship table along the time axis. Since there may be data collection intervals during the door body opening process, the curve is not continuous. The piecewise linear interpolation process divides the curve into multiple small segments according to the time interval. In each segment, the angle value of each moment within this period is calculated based on the angle values ​​of the two known time points before and after. After such processing, the generated angle time series is continuous angle data, and each moment has a corresponding angle value, ensuring consistency with the time dimension of the subsequent temperature data. Extract the time series of the average temperature change rate of the core influence area. This sequence is formed by arranging the average temperature change rate of the core influence area at each time point in chronological order. The time interval is the same as that of the angle time series, so as to perform correlation analysis. It is input into the dynamic time warping algorithm together with the angle time series to eliminate the timing offset caused by the physical response delay of the two and obtain the covariance matrix. The physical response delay means that after the door is opened, the temperature change does not occur immediately, but there is a certain lag. For example, the temperature of the core influence area begins to change significantly 1 second after the door is opened. The role of the dynamic time warping algorithm is to stretch or compress the time axis of one of the sequences to better align the two sequences in time. For example, the time of the temperature sequence is shifted back by 0.5 seconds to make it correspond to the changing trend of the angle sequence, thereby accurately reflecting the two sequences. The actual correlation between the two sequences, the covariance matrix is ​​used to describe the correlation between the two sequences. Each element in the matrix represents the degree of correlation between the angle value at a certain moment in the angle sequence and the temperature change rate value at the corresponding moment in the temperature change rate sequence. The larger the value, the more consistent the change trends of the two at that moment. The normalized cumulative value of the main diagonal elements in the covariance matrix is ​​extracted as the correlation coefficient. The main diagonal elements represent the covariance of the two sequences at the same moment. The normalized cumulative value is the sum of the values ​​of these elements, divided by the total number of elements, and then converted to a value between 0 and 1. 0 means no correlation, and 1 means a completely positive correlation. The correlation coefficient obtained in this way can intuitively reflect the overall correlation strength between the average temperature change rate of the core influence area and the opening angle of the cold storage door, providing an important basis for judging whether the temperature fluctuation is caused by the door opening disturbance.

[0026] In order to determine whether the temperature change in the stable conduction zone is caused by the door opening event, it is necessary to synchronously detect the temperature fluctuation amplitude value of the stable conduction zone at the time of the door opening event triggering operation, including: In the preset time period before the door opening event is triggered, the three-dimensional temperature field historical data of the stable conduction area is collected. The preset time period is set to 5 minutes because this time period can cover the regular cycle of temperature changes in the stable conduction area and is sufficient to reflect its normal fluctuation law. When collecting, the temperature data of the infrared temperature sensor corresponding to each grid point in the spherical space of the stable conduction area is extracted during the time period. The data collection interval is 10 seconds to ensure that enough historical samples are obtained. These data will serve as the basis for subsequent prediction of the theoretical temperature field to reflect the temperature characteristics of the area when there is no door opening disturbance. The theoretical temperature is predicted by the autoregressive model. Field distribution, the autoregressive model is a method to predict future trends based on historical data. Here, it is used to predict the theoretical temperature of each spatial point in the stable conduction area within the same length of time after the door opening event is triggered based on the collected 5-minute historical temperature data. In specific operations, the historical data is input into the model in chronological order. The model generates the theoretical temperature value of each spatial point at each moment in the future by analyzing the temperature change trend of each point. These theoretical temperature values ​​are combined to form a theoretical temperature field distribution, which represents the temperature state of the area when there is no door opening disturbance. The measured temperature field data after the door opening event is triggered is compared point by point with the theoretical temperature field to calculate the temperature of each spatial point. After the door opening event is triggered, the measured temperature of each grid point in the stable conduction area is collected for 5 minutes at the same collection interval as the historical data to form the measured temperature field data. Point-by-point comparison means that for each spatial point, the corresponding theoretical temperature is subtracted from the measured temperature, and then the absolute value is taken to obtain the absolute value of the temperature deviation of the point. The larger the deviation value, the more obvious the deviation between the actual temperature of the point and the normal state. The absolute value of the temperature deviation of the entire area is statistically analyzed. Statistical distribution analysis refers to arranging the absolute values ​​of the temperature deviation of all spatial points in order from small to large, observing the distribution of these values, and taking the previous The absolute value of the temperature deviation of the preset percentile is used as the temperature fluctuation amplitude value, and the preset percentile is set to 80%, that is, from the arranged deviation values, the value at the 80% position is selected as the temperature fluctuation amplitude value. If there are 100 spatial points in total, after arranging the deviation values ​​from small to large, the 80th value is the selected value. This value can not only reflect the temperature fluctuation level of most points, but also cover a certain proportion of larger deviations. It objectively reflects the overall temperature fluctuation degree of the stable conduction zone after the door opening event is triggered, and provides a quantitative basis for subsequent judgment of whether it is an equipment abnormality, effectively distinguishes door opening disturbances from equipment abnormalities, and ensures the accuracy of subsequent control decisions.

[0027] In order to determine the correlation between the temperature change rate in the core influence area and the door opening angle, and whether the temperature fluctuation in the stable conduction area is within a reasonable range, the correlation coefficient and the temperature fluctuation amplitude values ​​need to be analyzed and compared with their corresponding thresholds, including: The preset threshold reference library is called according to the current operating mode of the cold storage to obtain the initial threshold parameters. The cold storage operating modes include constant temperature storage mode (such as -18°C constant temperature for storing meat), low temperature quick freezing mode (such as -30°C mode for quick freezing of food), etc. The temperature stability requirements under different modes are different, and the corresponding threshold parameters are also different. The preset threshold reference library is a database that stores threshold data corresponding to various operating modes. The system identifies the current operating mode of the cold storage (such as the mode selection signal on the control panel) and retrieves the corresponding initial threshold parameters from the reference library to provide a basic standard for subsequent judgment. The threshold parameters are corrected in real time based on the compressor load rate and the temperature difference between the inside and outside of the cold storage. The correction rule is: for every 10% increase in the compressor load rate, the correlation coefficient threshold is lowered by 0.05, and the temperature fluctuation amplitude threshold is increased by 0.2 degrees Celsius. For every 10°C increase in the temperature difference between the inside and outside of the cold storage, the correlation coefficient threshold is increased by 0.03, and the temperature fluctuation amplitude threshold is increased by 0.3 degrees Celsius. The first threshold and the second threshold are set according to the corrected threshold parameters to determine whether the correlation between the temperature change rate in the core impact area and the door opening angle is significant. The correlation coefficient and the temperature fluctuation amplitude are respectively corrected according to the first threshold and the second threshold. The second threshold, that is, the corrected temperature fluctuation amplitude threshold, is used to determine whether the temperature fluctuation in the stable conduction zone is within a reasonable range. The two are used as judgment criteria for two dimensions to ensure the pertinence of the analysis and comparison, and to obtain the correlation coefficient judgment results and the temperature fluctuation amplitude judgment results. The judgment rule for the correlation coefficient is: if the correlation coefficient is greater than or equal to the first threshold, the judgment result is "significant correlation"; if it is less than the first threshold, the judgment result is "insignificant correlation"; the judgment rule for the temperature fluctuation amplitude is: if the temperature fluctuation amplitude is less than or equal to the second threshold, the judgment result is "normal fluctuation"; if it is greater than the second threshold, the judgment result is "abnormal fluctuation". These judgment results will serve as an important basis for the subsequent generation of identification factors, ensuring that the judgment of the disturbance type is more in line with the actual operating status, and providing a scientific judgment basis for distinguishing between door disturbances and equipment abnormalities.

[0028] In order to convert the determination results of the correlation coefficient and the temperature fluctuation amplitude into intuitive and reliable comparison information, the disturbance identification unit 2 needs to output the comparison result according to a specific method. The output method of the comparison result in the disturbance identification unit 2 specifically includes: The correlation coefficient and temperature fluctuation results are combined into a four-digit status code. The first two digits represent the strength of the correlation, and the last two digits represent the level of the temperature fluctuation. The correlation coefficient results are categorized as "significant" or "insignificant." The first two digits represent the strength of the correlation: "01" for significant correlation, and "00" for insignificant correlation. The temperature fluctuation results are categorized as "normal" or "abnormal." The last two digits represent the level of fluctuation: "01" for normal fluctuation, and "00" for abnormal fluctuation. For example, if the correlation coefficient is determined to be significant and the temperature fluctuation is normal, the combined result is a four-digit status code of "0101." If the correlation coefficient is insignificant and the temperature fluctuation is abnormal, the status code is "0000." This encoding method condenses the results of the two dimensions into a unified identifier, facilitating subsequent query and analysis. The corresponding semantic description text is retrieved from a pre-set mapping table, a pre-stored comparison table in the system. Each four-digit status code corresponds to a textual description of the combined two results. For example, "0101" corresponds to "the temperature change rate in the core impact area is significantly correlated with the door opening angle, and the temperature fluctuation in the stable conduction area is within the normal range, which meets the characteristics of the door opening disturbance", and "0000" corresponds to "the temperature change rate in the core impact area is not significantly correlated with the door opening angle, and the temperature fluctuation in the stable conduction area is abnormal, which may be an equipment abnormality." The semantic description makes the meaning of the status code easier to understand, providing a clear text basis for subsequent credibility retrieval. Based on the semantic description text, matching retrieval is performed in the credibility knowledge base to generate a confidence score. The credibility knowledge base stores the misjudgment records corresponding to all four-digit status codes in historical operation. Each record contains information such as the status code, semantic description, actual disturbance type, and whether the judgment result is correct. During a matching search, all historical records with the same semantic description are found in the knowledge base. The ratio of correct judgment results to the total number of records is calculated. The confidence score is calculated by reversely calculating the false positive rate for similar four-digit status codes in the historical data. Specifically, if there are 100 historical records for a status code, and 10 of them are false positives, the accuracy rate is 90%, and the confidence score is 90 (out of 100). For example, if the historical records for the status code "0101" show 80 correct judgments and 20 false positives, for an accuracy rate of 80%, the corresponding confidence score is 80. This score directly reflects the reliability of the current judgment result and is used as the output comparison result and synchronously fed back to the subsequent control unit. A higher confidence score indicates a more reliable judgment result based on the correlation coefficient and temperature fluctuation amplitude, providing a reliable decision basis for the compressor control unit to perform differentiated control. This avoids control errors caused by low-confidence judgment results and makes subsequent control decisions more targeted and accurate.

[0029] In order to accurately control the screw compressor according to different disturbance types and avoid unnecessary energy waste or fault expansion, the compressor control unit needs to perform differentiated regulation of the screw compressor according to the door opening disturbance identification factor or the equipment abnormality identification factor. The specific method is as follows: When the disturbance identification unit 2 outputs the door opening disturbance identification factor, it indicates that the current temperature field change is a temporary disturbance caused by the door opening, and there is no need to reduce the compressor load immediately. Therefore, the triggering time of the compressor frequency reduction instruction is delayed, and the delay time is dynamically adjusted according to the temperature fluctuation amplitude value. The specific adjustment rule is: for every 0.5 degrees Celsius increase in the temperature fluctuation amplitude value, the delay time increases by 30 seconds. For example, when the fluctuation amplitude value is 1.0 degrees Celsius, the delay time is set to 60 seconds, and when the fluctuation amplitude value is 1.5 degrees Celsius, the delay time is set to 90 seconds. Through this dynamic adjustment, the compressor has enough time to cope with the temperature shock caused by the door opening, avoiding the decrease in refrigeration efficiency caused by frequent frequency reduction. At the same time, the execution of the defrost program is frozen, because the temperature rise in the warehouse when the door is opened is temporary. Starting defrost at this time will further affect the refrigeration effect. The freezing state continues until the temperature in the stable conduction zone returns to the preset safety range. After the temperature returns to the safety range, the defrost program freezing state is automatically released and the normal defrost cycle is resumed; When disturbance identification unit 2 outputs an equipment anomaly identification factor, it indicates that the temperature field change may be caused by a compressor or related equipment failure, and immediate protective measures are required. Therefore, the compressor frequency reduction protection is immediately activated, reducing the compressor operating frequency to 50% of the rated frequency, reducing the equipment load to prevent the failure from worsening. At the same time, the oil system pressure is simultaneously increased to a safe threshold. The safety threshold is set according to the compressor model. The pressure increase is achieved by increasing the oil pump output power to ensure that the lubricating oil can fully lubricate the equipment components during the frequency reduction process, avoiding mechanical wear caused by insufficient lubrication. In addition, an early warning signal containing an abnormal location code is generated. The abnormal location code is automatically generated by the system based on the fault characteristics and uploaded to the remote operation and maintenance platform through the Internet of Things module.

[0030] In the present invention, the signal acquisition unit 1 obtains the timing signal of the opening and closing status of the cold storage door and the three-dimensional temperature field change data. The disturbance identification unit 2 locks the door opening moment through the timestamp synchronization, divides the core influence area and the stable conduction area, analyzes the correlation between the temperature change rate and the door opening angle and the temperature fluctuation amplitude, and outputs the door opening disturbance or equipment abnormality identification factor. The compressor control unit 3 performs differentiated regulation based on this, which solves the problems of disturbance misjudgment and regulation rigidity in traditional control and improves the compressor operation stability and energy consumption control accuracy.

[0031] A second object of the present invention is to provide a method for implementing any one of the above-mentioned artificial intelligence-based screw compressor operating state control systems, comprising the following steps: S1. The door magnetic sensor captures the cold storage door displacement timing signal in real time, and simultaneously drives the infrared temperature sensor array to collect dynamic three-dimensional temperature field data on the inner facade of the door. Based on the spatiotemporal coupling relationship between the sudden increase in displacement and the radial temperature drop pattern in the door area, the triggering moment of the effective door opening event is verified and locked, completing the initial perception of the cold storage disturbance event. S2. Based on the door opening trigger moment, sector modeling of the core impact area and spherical spatial division of the stable conduction area are performed. The anti-interference weighted average temperature change rate of the two areas within the asynchronous time window is calculated. The hysteresis correlation between the core area temperature change rate and the door opening angle is analyzed using the dynamic time warping algorithm. The temperature fluctuation amplitude value of the stable conduction area is simultaneously detected. Combined with the dynamic correction threshold, a multi-dimensional comparison decision is made, and a four-bit status code is generated with an additional confidence score. Finally, the door opening disturbance identification factor or equipment abnormality identification factor is output; S3. When the door opening disturbance identification factor is output, the compressor frequency reduction response time is delayed and the defrost program is frozen until the temperature in the stable conduction zone returns to the preset safety threshold. When the device abnormality identification factor is output, the compressor frequency reduction protection is immediately triggered and the oil circuit pressure is increased to the safety critical value. At the same time, an early warning signal with positioning code is generated and uploaded to the remote operation and maintenance platform to locate and intervene in the fault.

[0032] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and improvements fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. The screw compressor operating state control system based on artificial intelligence is characterized by: include: The signal acquisition unit (1) acquires the opening and closing status timing signals of the cold storage door body and the three-dimensional temperature field change data generated by the infrared temperature sensor array deployed on the inner side of the door in real time, and executes the following calculation process in sequence: The disturbance identification unit (2) synchronizes the time stamps of the opening and closing state timing signal with the three-dimensional temperature field change data, locks the door opening event triggering moment, divides the core influence area and the stable conduction area with the cold storage door as the center, calculates the average temperature change rate of the core influence area and the stable conduction area within the time window, analyzes the correlation between the average temperature change rate of the core influence area and the opening angle of the cold storage door, obtains the correlation coefficient, synchronously detects the temperature fluctuation amplitude value of the stable conduction area at the door opening event triggering moment, analyzes and compares the correlation coefficient and the temperature fluctuation amplitude value with their corresponding thresholds, outputs the comparison result, and judges and outputs the door opening disturbance identification factor or the equipment abnormality identification factor according to the comparison result; The compressor control unit (3) performs differentiated regulation on the screw compressor according to the door opening disturbance identification factor or the equipment abnormality identification factor.

2. The artificial intelligence-based screw compressor operating state control system according to claim 1, characterized in that: The operation of synchronizing the time stamp of the opening and closing state timing signal with the three-dimensional temperature field change data and locking the triggering moment of the door opening event includes the following steps: The door magnetic sensor is used to capture the continuous change trajectory of the cold storage door displacement and generate a displacement time series curve. At the same time, the temperature change gradient data of each detection unit in the infrared temperature sensor array is extracted to generate three-dimensional temperature field change data aligned with the time axis of the displacement time series curve. When the displacement of the cold storage door increases and exceeds the preset displacement threshold within the specified time, it is marked as the potential door opening starting point. The multi-point temperature gradient change trend of the cold storage door area in the three-dimensional temperature field change data is analyzed and compared with the preset judgment rules. The temperature mutation node is marked according to the comparison result. Finally, the time overlap between the potential door opening starting point and the temperature mutation node of the displacement timing curve is cross-validated to confirm the triggering moment of the effective door opening event.

3. The artificial intelligence-based screw compressor operating state control system according to claim 2, characterized in that: The spatial zoning method of dividing the core influence area and the stable conduction area with the cold storage door as the center includes: Taking the geometric center of the cold storage door as the origin, a three-dimensional fan-shaped space area perpendicular to the door opening direction is constructed as the core influence area. Its radial extension distance is dynamically adjusted according to the volume of the cold storage, and a layered detection grid is set in the core influence area. A spherical space area is delineated with the evaporator unit as the center, which is used as the stable conduction zone. Its radius is adaptively adjusted according to the current operating power of the screw compressor. A transition buffer zone is set at the junction of the stable conduction zone and the core influence zone to suppress the boundary effect error of the temperature field analysis.

4. The artificial intelligence-based screw compressor operating state control system according to claim 3, characterized in that: The operation of calculating the average temperature change rate of the core influence area and the stable conduction area within the time window includes: Sampling points are selected in the layered detection grid of the core impact area. Abnormal temperature data are eliminated according to the sampling points to form valid sampling points. The first-order derivative of the temperature change curve of the valid sampling points is calculated to extract the instantaneous value of the temperature change rate of each valid sampling point. Finally, the weighted average algorithm is used to fuse the instantaneous values ​​of the temperature change rate of each valid sampling point to obtain the average temperature change rate of the core impact area. The full-space temperature field integral operation is performed in the stable conduction zone to generate the overall temperature change trend function. The overall temperature change trend function is smoothed, and its linear fitting slope in the time window is extracted. The linear fitting slope is used as the average temperature change rate in the stable conduction zone.

5. The artificial intelligence-based screw compressor operating state control system according to claim 4, characterized in that: The steps for analyzing the correlation between the average temperature change rate of the core impact area and the opening angle of the cold storage door and obtaining the correlation coefficient include: The real-time opening angle is calculated by integrating the cold storage door displacement time series curve, and an angle-time change relationship table is established. According to the angle-time change relationship table, the angle change curve is subjected to piecewise linear interpolation processing to generate an angle time series sequence; The time series of the average temperature change rate in the core influence area is extracted and input into the dynamic time warping algorithm together with the angle time series to eliminate the time series offset caused by the physical response delay between the two. The covariance matrix is ​​obtained, and the normalized cumulative value of the main diagonal elements in the covariance matrix is ​​extracted as the correlation coefficient.

6. The artificial intelligence-based screw compressor operating state control system according to claim 5, characterized in that: The operation of synchronously detecting the temperature fluctuation amplitude value of the stable conduction area at the time of triggering the door opening event specifically includes: During the preset time period before the door opening event is triggered, the three-dimensional temperature field historical data of the stable conduction area is collected, and the theoretical temperature field distribution is predicted by the autoregressive model. The measured temperature field data after the door opening event is triggered are compared point by point with the theoretical temperature field, and the absolute value of the temperature deviation at each spatial point is calculated. The absolute value of the temperature deviation in the entire area is statistically analyzed, and the absolute value of the temperature deviation at the preset quantile is taken as the temperature fluctuation amplitude value.

7. The artificial intelligence-based screw compressor operating state control system according to claim 6, characterized in that: The correlation coefficient and temperature fluctuation amplitude values ​​are analyzed and compared with their corresponding thresholds, including: According to the current operating mode of the cold storage, the preset threshold reference library is called to obtain the initial threshold parameters. The threshold parameters are corrected in real time in combination with the compressor load rate and the temperature difference between the inside and outside of the cold storage. The first threshold and the second threshold are set according to the corrected threshold parameters. The correlation coefficient and the temperature fluctuation amplitude value are judged according to the first threshold and the second threshold respectively, and the correlation coefficient judgment result and the temperature fluctuation amplitude judgment result are obtained.

8. The artificial intelligence-based screw compressor operating state control system according to claim 7, characterized in that: The output method of the comparison result in the disturbance identification unit (2) specifically includes: The correlation coefficient determination result and the temperature fluctuation amplitude determination result are combined into a four-digit status code. The first two digits represent the correlation strength, and the last two digits represent the temperature fluctuation amplitude level. The corresponding semantic description text is queried in the preset mapping table based on the four-digit status code. A matching search is performed in the credibility knowledge base based on the semantic description text to generate a confidence score. The confidence score is reversely calculated based on the misjudgment rate of similar four-digit status codes in historical data, and the confidence score is used as the output comparison result.

9. The artificial intelligence-based screw compressor operating state control system according to claim 8, characterized in that: The method for performing differentiated control on the screw compressor according to the door opening disturbance identification factor or the equipment abnormality identification factor is as follows: When the door opening disturbance identification factor is output, the triggering time of the compressor frequency reduction instruction is delayed. The delay time is dynamically adjusted according to the temperature fluctuation amplitude value, and the defrost program execution is frozen until the temperature field returns to the preset safety range of the stable conduction zone; When the output device abnormality identification factor is detected, the compressor frequency reduction protection is immediately activated and the oil system pressure is simultaneously increased to a safe threshold. An early warning signal containing an abnormality location code is generated and uploaded to the remote operation and maintenance platform through the Internet of Things module.

10. A method for implementing the artificial intelligence-based screw compressor operating state control system according to any one of claims 1 to 9, characterized in that: The following steps are involved: S1. The door magnetic sensor captures the cold storage door displacement timing signal in real time, and simultaneously drives the infrared temperature sensor array to collect dynamic three-dimensional temperature field data on the inner facade of the door. Based on the spatiotemporal coupling relationship between the sudden increase in displacement and the radial temperature drop pattern in the door area, the triggering moment of the effective door opening event is verified and locked, completing the initial perception of the cold storage disturbance event. S2. Based on the door opening trigger moment, sector modeling of the core impact area and spherical spatial division of the stable conduction area are performed. The anti-interference weighted average temperature change rate of the two areas within the asynchronous time window is calculated. The hysteresis correlation between the core area temperature change rate and the door opening angle is analyzed using the dynamic time warping algorithm. The temperature fluctuation amplitude value of the stable conduction area is simultaneously detected. Combined with the dynamic correction threshold, a multi-dimensional comparison decision is made, and a four-bit status code is generated with an additional confidence score. Finally, the door opening disturbance identification factor or equipment abnormality identification factor is output; S3. When the door opening disturbance identification factor is output, the compressor frequency reduction response time is delayed and the defrost program is frozen until the temperature in the stable conduction zone returns to the preset safety threshold. When the device abnormality identification factor is output, the compressor frequency reduction protection is immediately triggered and the oil circuit pressure is increased to the safety critical value. At the same time, an early warning signal with positioning code is generated and uploaded to the remote operation and maintenance platform to locate and intervene in the fault.

Images