Graded early warning method and system based on pressure trend analysis and null drift discrimination

Abstract

The invention discloses a graded early warning method and system based on pressure trend analysis and null drift discrimination, and relates to the technical field of intracranial pressure monitoring. The method comprises the steps that pressure data of the intracranial pressure monitoring equipment are collected in real time to obtain a pressure data sequence, and the running state of the intracranial pressure monitoring equipment is obtained; performing trend analysis on the pressure data sequence to obtain a change trend result; performing null drift correction processing on the pressure data sequence to obtain a corrected pressure data sequence; and performing early warning on the intracranial pressure state according to the change trend result and the corrected pressure data sequence. According to the method, the problem of early warning lag caused by lack of trend pre-judgment in traditional monitoring is solved, false alarm caused by signal drift is avoided through zero drift correction, and the accuracy of intracranial pressure early warning is improved.

Description

Technical Field

[0001] This invention belongs to the field of intracranial pressure monitoring technology, specifically relating to a graded early warning method and system based on pressure trend analysis and zero drift identification. Background Technology

[0002] In the field of intracranial pressure monitoring, real-time monitoring and early warning of abnormalities are core methods for assessing the patient's brain status and preventing serious complications such as brain herniation in medical settings such as neurosurgery. Currently, existing intracranial pressure monitoring systems mainly collect real-time pressure data through intracranial pressure sensors, compare the collected pressure values ​​with preset safety thresholds, and trigger audible and visual alarms when the real-time intracranial pressure exceeds the threshold to prompt medical staff to intervene. At the same time, some systems also record historical intracranial pressure data for subsequent retrospective analysis of the patient's condition.

[0003] In practical applications, existing intracranial pressure (ICP) monitoring systems have significant technical shortcomings. Generally, ICP monitoring relies solely on real-time ICP exceeding limits to trigger alarms, lacking the ability to dynamically analyze ICP trends. They cannot identify early warning signs such as abnormally rapid increases in ICP or intensified periodic fluctuations within the normal range. Alarms are often only issued after ICP has exceeded safety thresholds and the patient's condition has deteriorated to an emergency state, hindering early intervention. Furthermore, ICP sensors are susceptible to signal drift due to factors such as the human physiological environment and equipment contact conditions. Existing systems lack specific logic to detect ICP signal drift, often misinterpreting baseline fluctuations caused by drift as actual abnormal changes in ICP. This not only increases unnecessary verification work for medical staff but may also reduce their attention to alarm signals due to frequent false alarms. Summary of the Invention

[0004] The purpose of this invention is to solve the problems of existing intracranial pressure monitoring systems that rely solely on real-time intracranial pressure exceeding the limit to trigger alarms, lack the ability to analyze pressure change trends leading to delayed early warnings, and lack dedicated logic for identifying zero drift in intracranial pressure signals, which easily leads to misjudging zero drift as actual pressure abnormalities. Therefore, this invention proposes a graded early warning method and system based on pressure trend analysis and zero drift identification.

[0005] In a first aspect of this invention, a graded early warning method based on pressure trend analysis and zero drift identification is first proposed, the method comprising: Pressure data from the intracranial pressure monitoring device is collected in real time to obtain a pressure data sequence, and the operating status of the intracranial pressure monitoring device is obtained. Trend analysis was performed on the pressure data sequence to obtain the trend results. The pressure data sequence is subjected to zero-drift correction processing to obtain a corrected pressure data sequence; The intracranial pressure status is given early warning based on the changing trend results and the corrected pressure data sequence.

[0006] Optionally, trend analysis of the pressure data sequence to obtain trend analysis results includes: The first rate of change and the first fluctuation value are obtained by calculating the instantaneous rate of change and the intensity of high-frequency fluctuations of the pressure data sequence through a first preset sliding window. The second rate of change and the second fluctuation value are obtained by calculating the long-term change slope and low-frequency fluctuation intensity of the pressure data sequence through a first preset sliding window; the length of the first preset sliding window is less than the length of the second preset sliding window. The trend change value is obtained by weighting the first rate of change and the second rate of change; The fluctuation quantization value is obtained by calculating the ratio of the first fluctuation value to the second fluctuation value. The trend change result is determined based on the trend change value and the fluctuation quantification value.

[0007] Optionally, performing zero-drift correction processing on the pressure data sequence to obtain a corrected pressure data sequence includes: The average rate value is obtained by calculating the average rate of change of the pressure data sequence through a first preset sliding window. If the time during which the absolute value of the average rate is less than the zero drift determination rate threshold is greater than the duration of the second preset sliding window, then the cumulative deviation of the pressure data sequence relative to the preset pressure reference value is calculated; the length of the first preset sliding window is less than the length of the second preset sliding window. If the absolute value of the cumulative deviation is greater than the preset zero drift confirmation deviation threshold, the pressure data sequence is corrected according to the cumulative deviation to obtain a corrected pressure data sequence.

[0008] Optionally, calculating the cumulative deviation of the pressure data sequence relative to a preset pressure reference value includes: pass The cumulative deviation value is obtained; where N is the total number of data points in the second preset sliding window. For the i-th pressure data, Here, 'a' represents the preset pressure reference value, 'b' represents the time decay coefficient, and 'a' represents the period coefficient. This is the time step from the current data point back to the historical data point.

[0009] Optionally, providing early warning of intracranial pressure status based on the changing trend results and the dynamic alarm threshold includes: If the current pressure value in the pressure data sequence is less than the dynamic threshold, but the trend result meets the preset risk conditions, an early warning state is triggered and the first type of prompt operation is executed. When the current pressure value in the pressure data sequence is greater than the dynamic threshold, an alarm state is triggered, and the second type of prompt operation is executed.

[0010] In a second aspect of this invention, a graded early warning system based on pressure trend analysis and zero drift identification is proposed, comprising: The acquisition module is used to acquire pressure data and environmental parameters from the intracranial pressure monitoring device in real time, obtain pressure data sequences and environmental data sequences, and acquire the operating status of the intracranial pressure monitoring device. The trend analysis module is used to perform trend analysis on the pressure data sequence to obtain the changing trend results; A zero-drift correction module is used to perform zero-drift correction processing on the pressure data sequence to obtain a corrected pressure data sequence. The threshold calculation module is used to calculate the dynamic alarm threshold based on the corrected pressure data sequence; The early warning module is used to provide early warning of intracranial pressure status based on the changing trend results and the dynamic alarm threshold.

[0011] Optionally, the trend analysis module includes: The first calculation module is used to calculate the instantaneous change rate and high-frequency fluctuation intensity of the pressure data sequence through a first preset sliding window to obtain a first change rate and a first fluctuation value; The second calculation module is used to calculate the long-term change slope and low-frequency fluctuation intensity of the pressure data sequence through a first preset sliding window to obtain a second change rate and a second fluctuation value; the length of the first preset sliding window is less than the length of the second preset sliding window. The weighting module is used to weight the first rate of change and the second rate of change to obtain the trend change value; The volatility quantization value calculation module is used to calculate the ratio of the first volatility value and the second volatility value to obtain the volatility quantization value; The result determination module is used to determine the trend change result based on the trend change value and the fluctuation quantification value.

[0012] Optionally, the zero-drift correction module includes: The average rate calculation module is used to calculate the average rate of change of the pressure data sequence through a first preset sliding window to obtain the average rate value. The cumulative deviation calculation module is used to calculate the cumulative deviation value of the pressure data sequence relative to a preset pressure reference value if the time when the absolute value of the average rate value is less than the zero drift determination rate threshold is greater than the duration of the second preset sliding window; the length of the first preset sliding window is less than the length of the second preset sliding window. The correction module is used to correct the pressure data sequence according to the cumulative deviation value if the absolute value of the cumulative deviation value is greater than the preset zero drift confirmation deviation threshold, so as to obtain a corrected pressure data sequence.

[0013] Optionally, the cumulative deviation calculation module includes: pass The cumulative deviation value is obtained; where N is the total number of data points in the second preset sliding window. For the i-th pressure data, Here, 'a' represents the preset pressure reference value, 'b' represents the time decay coefficient, and 'a' represents the period coefficient. This is the time step from the current data point back to the historical data point.

[0014] Optionally, the early warning module includes: The first condition module is used to trigger an early warning state and execute a first type of prompt operation if the current pressure value in the pressure data sequence is less than the dynamic threshold, but the trend result meets the preset risk conditions. The second condition module is used to trigger an alarm state and perform a second type of prompt operation when the current pressure value in the pressure data sequence is greater than a dynamic threshold.

[0015] The beneficial effects of this invention are as follows: This invention proposes a method that, by real-time acquisition of pressure data and the operating status of an intracranial pressure monitoring device, first performs trend analysis on the pressure data sequence to understand its changing patterns, then obtains a precise corrected pressure data sequence through zero-drift correction processing, and finally combines the trend results and corrected data to provide early warning of intracranial pressure status. This method not only solves the problem of delayed early warning caused by the lack of trend prediction in traditional monitoring, but also avoids false alarms caused by signal drift through zero-drift correction, thus improving the accuracy of intracranial pressure early warning. Attached Figure Description

[0016] The present invention will now be further described with reference to the accompanying drawings.

[0017] Figure 1 A flowchart of a graded early warning method based on pressure trend analysis and zero drift identification provided in an embodiment of the present invention; Figure 2 This is a framework diagram of a graded early warning system based on pressure trend analysis and zero drift identification, provided for an embodiment of the present invention. Detailed Implementation

[0018] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0019] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] This invention provides a graded early warning method based on pressure trend analysis and zero drift identification. See also... Figure 1 , Figure 1 A flowchart illustrating a graded early warning method based on pressure trend analysis and zero drift identification, provided for an embodiment of the present invention. The method includes the following steps: S101, real-time acquisition of pressure data from intracranial pressure monitoring equipment to obtain pressure data sequence, and acquisition of the operating status of intracranial pressure monitoring equipment; S102, perform trend analysis on the pressure data sequence to obtain the trend results; S103, perform zero-drift correction on the pressure data sequence to obtain the corrected pressure data sequence; S104 provides early warning of intracranial pressure status based on trend results and corrected pressure data sequences.

[0021] This invention provides a graded early warning method based on pressure trend analysis and zero-drift discrimination. By simultaneously acquiring the pressure data sequence output by an intracranial pressure monitoring device and the device's own operating status, the method first performs trend analysis on the pressure data sequence to clarify its dynamic change trend. Then, a zero-drift correction operation is performed on the data sequence to obtain a precisely corrected pressure data sequence that eliminates signal drift interference. Finally, based on the aforementioned trend results and the corrected pressure data sequence, an early warning determination of the intracranial pressure status is achieved. This method, on the one hand, compensates for the early warning lag defect of traditional intracranial pressure monitoring due to the lack of trend prediction capability; on the other hand, it effectively avoids false alarms caused by signal drift through zero-drift correction, thereby improving the accuracy of intracranial pressure early warning.

[0022] In one embodiment, trend analysis of a pressure data series yields the following results: The first rate of change and the first fluctuation value are obtained by calculating the instantaneous rate of change and the intensity of high-frequency fluctuations of the pressure data sequence through the first preset sliding window. The second rate of change and the second fluctuation value are obtained by calculating the long-term change slope and the low-frequency fluctuation intensity of the pressure data sequence through the first preset sliding window; the length of the first preset sliding window is less than the length of the second preset sliding window. The trend change value is obtained by weighting the first rate of change and the second rate of change. The quantized value of the fluctuation is obtained by calculating the ratio of the first fluctuation value to the second fluctuation value. The trend change result is determined based on the trend change value and the fluctuation quantification value.

[0023] In one implementation, the length of the first preset sliding window is 25 minutes, and the length of the second preset sliding window is 3 hours.

[0024] In one implementation, in short-timescale analysis, the instantaneous rate of change is calculated using the central difference method. For the pressure value at the current moment, the average rate of change between the pressure values ​​at the previous and next adjacent time points is calculated. This is done by subtracting the pressure value at the previous sampling point from the pressure value at the next sampling point, and then dividing by twice the sampling interval. The high-frequency fluctuation intensity is obtained by calculating the standard deviation between the pressure data within the window and its linear fitting line. This involves performing linear regression on all pressure values ​​within the window to obtain a trend line, and then calculating the root mean square value of the difference between the pressure value at each sampling point and the corresponding value on the trend line as the first fluctuation value. In long-timescale analysis, the long-term slope is calculated using the least squares linear fitting method. A linear regression is performed on all pressure-time data points within the window, and the slope of the resulting line is the second rate of change, which reflects the overall trend of pressure change over a longer period. The low-frequency fluctuation intensity is obtained by calculating the envelope energy of the detrended sequence. First, a moving average method or a low-pass filter is used to extract the low-frequency component from the original pressure data, and then the root mean square value of this low-frequency component is calculated as the second fluctuation value.

[0025] In one implementation, the process of determining the trend result based on the trend change value and the fluctuation quantification value involves comparing the trend change value with a first preset threshold α and a second preset threshold β, while simultaneously comparing the fluctuation quantification value with a third preset threshold γ. When the trend change value is greater than α, regardless of the fluctuation quantification value, it is determined to be a rapid approaching state. When the trend change value is between β and α and the fluctuation quantification value is greater than γ, the system determines it to be an abnormal fluctuation and approaching state. When the trend change value is between β and α and the fluctuation quantification value is less than or equal to γ, the system determines it to be a slow approaching state. When the absolute value of the trend change value is less than or equal to β and the fluctuation quantification value is greater than γ, the system determines it to be an abnormal fluctuation state. When the absolute value of the trend change value is less than or equal to β and the fluctuation quantification value is less than or equal to γ, the system determines it to be a stable state. When the trend change value is less than β, regardless of the fluctuation quantification value, the system determines it to be a slow moving away state. This system, through the above-mentioned combination comparison logic based on the triple thresholds, maps continuous quantification parameters to discrete trend state classifications, providing structured input for subsequent early warning decisions.

[0026] In one embodiment, performing zero-drift correction processing on the pressure data sequence to obtain a corrected pressure data sequence includes: The average rate value is obtained by calculating the average rate of change of the pressure data sequence through the first preset sliding window. If the time when the absolute value of the average rate is less than the zero drift determination rate threshold is greater than the duration of the second preset sliding window, then the cumulative deviation of the pressure data sequence relative to the preset pressure reference value is calculated; the length of the first preset sliding window is less than the length of the second preset sliding window. If the absolute value of the cumulative deviation is greater than the preset zero drift confirmation deviation threshold, the pressure data sequence is corrected based on the cumulative deviation to obtain the corrected pressure data sequence.

[0027] In one implementation, the preset pressure reference value is determined by selecting a stable period of pressure data that has been confirmed to be free of abnormalities after equipment calibration or in the early stage of monitoring, and calculating its statistical median as a reference. The weighting coefficient usually decays exponentially over time, making the contribution of recent deviations to the cumulative value greater. This cumulative calculation can amplify small but continuous offsets to a detectable level, while improving the signal-to-noise ratio through the integration process and effectively distinguishing between zero drift and random fluctuations. The correction adopts a real-time compensation mechanism, that is, from the current moment, all newly acquired pressure data are subtracted from the cumulative deviation value.

[0028] In one embodiment, calculating the cumulative deviation of the pressure data sequence relative to a preset pressure reference value includes: pass The cumulative deviation value is obtained; where N is the total number of data points in the second preset sliding window. For the i-th pressure data, Here, 'a' represents the preset pressure reference value, 'b' represents the time decay coefficient, and 'a' represents the period coefficient. This is the time step from the current data point back to the historical data point.

[0029] In one implementation, the time decay coefficient ranges from 0.1 to 1, and the period coefficient ranges from 0.2 to 1.5, which are specifically set by technical personnel.

[0030] In one implementation, the process uses the product of an exponentially decaying base and a periodic modulation factor as a weighting function to achieve non-uniform weighted accumulation of historical pressure deviations, with the exponentially decaying term... To ensure the system is more sensitive to recent data, consistent with the causal characteristic that the near-term impact of a physical system is greater than its long-term impact; while the periodic modulation term By increasing the weights at specific backtracking time steps, the system can identify and focus on potentially recurring periodic drift patterns or historical similarity features. The advantage of this approach is that the system can not only respond quickly to sudden zero drifts but also capture slow, periodic systematic shifts. This allows for more accurate differentiation between real zero drifts and random noise under complex operating conditions, improving the accuracy and reliability of early warnings, while avoiding misjudgments and missed judgments caused by fixed thresholds or simple averaging methods.

[0031] In one embodiment, providing early warning of intracranial pressure status based on trend results and dynamic alarm thresholds includes: If the current pressure value in the pressure data sequence is less than the dynamic threshold, but the trend result meets the preset risk conditions, an early warning state is triggered and the first type of prompt operation is executed. When the current pressure value in the pressure data sequence is greater than the dynamic threshold, an alarm state is triggered, and the second type of prompt operation is executed.

[0032] In one implementation, the first type of prompt is a mild, non-urgent prompt, such as a change in page color, slight flashing, or a low-priority sound prompt, designed to remind medical staff to pay attention in advance and prevent the situation from worsening; the second type of prompt is an emergency alarm, including forced pop-ups, strong sound and light alarms, and multi-channel push notifications, requiring medical staff to take immediate action.

[0033] Based on the same inventive concept, this invention also provides a graded early warning system based on pressure trend analysis and zero drift identification. See also Figure 2 , Figure 2 A framework diagram of a graded early warning system based on pressure trend analysis and zero drift identification provided in this embodiment of the invention includes: The acquisition module is used to acquire pressure data and environmental parameters from the intracranial pressure monitoring device in real time, obtain pressure data sequences and environmental data sequences, and acquire the operating status of the intracranial pressure monitoring device. The trend analysis module is used to perform trend analysis on pressure data sequences to obtain the results of changing trends. The zero-drift correction module is used to perform zero-drift correction processing on the pressure data sequence to obtain a corrected pressure data sequence. The early warning module is used to provide early warnings about intracranial pressure status based on the changing trend results and the corrected pressure data sequence.

[0034] This invention provides a graded early warning system based on pressure trend analysis and zero-drift discrimination. By simultaneously acquiring the pressure data sequence output by an intracranial pressure monitoring device and the device's own operating status, the system first performs trend analysis on the pressure data sequence to clarify its dynamic change trend. Then, a zero-drift correction operation is performed on the data sequence to obtain a precisely corrected pressure data sequence that eliminates signal drift interference. Finally, based on the aforementioned trend results and the corrected pressure data sequence, an early warning determination of the intracranial pressure status is achieved. This method, on the one hand, compensates for the early warning lag defect of traditional intracranial pressure monitoring due to the lack of trend prediction capability; on the other hand, it effectively avoids false alarms caused by signal drift through zero-drift correction, thereby improving the accuracy of intracranial pressure early warning.

[0035] In one embodiment, the trend analysis module includes: The first calculation module is used to calculate the instantaneous rate of change and the intensity of high-frequency fluctuations of the pressure data sequence through a first preset sliding window to obtain a first rate of change and a first fluctuation value; The second calculation module is used to calculate the long-term change slope and low-frequency fluctuation intensity of the pressure data sequence through a first preset sliding window to obtain a second change rate and a second fluctuation value; the length of the first preset sliding window is less than the length of the second preset sliding window. The weighting module is used to weight the first rate of change and the second rate of change to obtain the trend change value; The volatility quantization value calculation module is used to calculate the ratio of the first volatility value and the second volatility value to obtain the volatility quantization value. The result determination module is used to determine the trend change result based on the trend change value and the fluctuation quantification value.

[0036] In one embodiment, the zero-drift correction module includes: The average rate calculation module is used to calculate the average rate of change of the pressure data sequence through a first preset sliding window to obtain the average rate value. The cumulative deviation calculation module is used to calculate the cumulative deviation of the pressure data sequence relative to the preset pressure reference value if the time when the absolute value of the average rate value is less than the zero drift determination rate threshold is greater than the duration of the second preset sliding window; the length of the first preset sliding window is less than the length of the second preset sliding window. The correction module is used to correct the pressure data sequence based on the cumulative deviation value if the absolute value of the cumulative deviation value is greater than the preset zero drift confirmation deviation threshold, so as to obtain the corrected pressure data sequence.

[0037] In one embodiment, the cumulative deviation calculation module includes: pass The cumulative deviation value is obtained; where N is the total number of data points in the second preset sliding window. For the i-th pressure data, Here, 'a' represents the preset pressure reference value, 'b' represents the time decay coefficient, and 'a' represents the period coefficient. This is the time step from the current data point back to the historical data point.

[0038] In one embodiment, the early warning module includes: The first condition module is used to trigger an early warning state and execute the first type of prompt operation when the current pressure value in the pressure data sequence is less than the dynamic threshold, but the trend result meets the preset risk conditions. The second condition module is used to trigger an alarm state and execute the second type of prompt operation when the current pressure value in the pressure data sequence is greater than the dynamic threshold.

[0039] The foregoing has described one embodiment of the present invention in detail, but this content is merely a preferred embodiment and should not be considered as limiting the scope of the present invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the scope of the claims of this invention.

Claims

1. A graded early warning method based on pressure trend analysis and zero drift identification, characterized in that, The method includes: Pressure data from the intracranial pressure monitoring device is collected in real time to obtain a pressure data sequence, and the operating status of the intracranial pressure monitoring device is obtained. Trend analysis was performed on the pressure data sequence to obtain the trend results. The pressure data sequence is subjected to zero-drift correction processing to obtain a corrected pressure data sequence; The intracranial pressure status is given early warning based on the changing trend results and the corrected pressure data sequence.

2. The graded early warning method based on pressure trend analysis and zero drift identification according to claim 1, characterized in that, Trend analysis of the pressure data series yielded the following results: The first rate of change and the first fluctuation value are obtained by calculating the instantaneous rate of change and the intensity of high-frequency fluctuations of the pressure data sequence through a first preset sliding window. The second rate of change and the second fluctuation value are obtained by calculating the long-term change slope and low-frequency fluctuation intensity of the pressure data sequence through a first preset sliding window; the length of the first preset sliding window is less than the length of the second preset sliding window. The trend change value is obtained by weighting the first rate of change and the second rate of change; The fluctuation quantization value is obtained by calculating the ratio of the first fluctuation value to the second fluctuation value. The trend change result is determined based on the trend change value and the fluctuation quantification value.

3. The graded early warning method based on pressure trend analysis and zero drift identification according to claim 1, characterized in that, The corrected pressure data sequence is obtained by performing zero-drift correction processing on the pressure data sequence, including: The average rate value is obtained by calculating the average rate of change of the pressure data sequence through a first preset sliding window. If the time during which the absolute value of the average rate is less than the zero drift determination rate threshold is greater than the duration of the second preset sliding window, then the cumulative deviation of the pressure data sequence relative to the preset pressure reference value is calculated; the length of the first preset sliding window is less than the length of the second preset sliding window. If the absolute value of the cumulative deviation is greater than the preset zero drift confirmation deviation threshold, the pressure data sequence is corrected according to the cumulative deviation to obtain a corrected pressure data sequence.

4. The graded early warning method based on pressure trend analysis and zero drift identification according to claim 3, characterized in that, Calculating the cumulative deviation of the pressure data sequence relative to a preset pressure reference value includes: pass The cumulative deviation value is calculated; where N is the total number of data points in the second preset sliding window. For the i-th pressure data, Here, 'a' represents the preset pressure reference value, 'b' represents the time decay coefficient, and 'a' represents the period coefficient. This is the time step from the current data point back to the historical data point.

5. A graded early warning method based on pressure trend analysis and zero drift identification according to claim 1, characterized in that, Early warning of intracranial pressure status based on the aforementioned trend results and the revised pressure data sequence includes: If the current pressure value in the corrected pressure data sequence is less than a preset threshold, but the trend result meets the preset risk conditions, an early warning state is triggered and the first type of prompt operation is executed. When the current pressure value in the corrected pressure data sequence is greater than a preset threshold, an alarm state is triggered, and the second type of prompt operation is executed.

6. A graded early warning system based on pressure trend analysis and zero drift identification, characterized in that, The system includes: The acquisition module is used to acquire pressure data and environmental parameters from the intracranial pressure monitoring device in real time, obtain pressure data sequences and environmental data sequences, and acquire the operating status of the intracranial pressure monitoring device. The trend analysis module is used to perform trend analysis on the pressure data sequence to obtain the changing trend results; A zero-drift correction module is used to perform zero-drift correction processing on the pressure data sequence to obtain a corrected pressure data sequence. The early warning module is used to provide early warning of intracranial pressure status based on the changing trend results and the corrected pressure data sequence.

7. A graded early warning system based on pressure trend analysis and zero drift identification according to claim 6, characterized in that, The trend analysis module includes: The first calculation module is used to calculate the instantaneous change rate and high-frequency fluctuation intensity of the pressure data sequence through a first preset sliding window to obtain a first change rate and a first fluctuation value; The second calculation module is used to calculate the long-term change slope and low-frequency fluctuation intensity of the pressure data sequence through a first preset sliding window to obtain a second change rate and a second fluctuation value; the length of the first preset sliding window is less than the length of the second preset sliding window. The weighting module is used to weight the first rate of change and the second rate of change to obtain the trend change value; The volatility quantization value calculation module is used to calculate the ratio of the first volatility value and the second volatility value to obtain the volatility quantization value; The result determination module is used to determine the trend change result based on the trend change value and the fluctuation quantification value.

8. A graded early warning system based on pressure trend analysis and zero drift identification according to claim 6, characterized in that, The zero-drift correction module includes: The average rate calculation module is used to calculate the average rate of change of the pressure data sequence through a first preset sliding window to obtain the average rate value. The cumulative deviation calculation module is used to calculate the cumulative deviation value of the pressure data sequence relative to a preset pressure reference value if the time when the absolute value of the average rate value is less than the zero drift determination rate threshold is greater than the duration of the second preset sliding window; the length of the first preset sliding window is less than the length of the second preset sliding window. The correction module is used to correct the pressure data sequence according to the cumulative deviation value if the absolute value of the cumulative deviation value is greater than the preset zero drift confirmation deviation threshold, so as to obtain a corrected pressure data sequence.

9. A graded early warning system based on pressure trend analysis and zero drift identification according to claim 6, characterized in that, The cumulative deviation calculation module includes: pass The cumulative deviation value is obtained; where N is the total number of data points in the second preset sliding window. For the i-th pressure data, Here, 'a' represents the preset pressure reference value, 'b' represents the time decay coefficient, and 'a' represents the period coefficient. This is the time step from the current data point back to the historical data point.

10. A graded early warning system based on pressure trend analysis and zero drift identification according to claim 6, characterized in that, The early warning module includes: The first condition module is used to trigger an early warning state and execute a first type of prompt operation if the current pressure value in the pressure data sequence is less than the dynamic threshold, but the trend result meets the preset risk conditions. The second condition module is used to trigger an alarm state and perform a second type of prompt operation when the current pressure value in the pressure data sequence is greater than a dynamic threshold.

Images