Intelligent pressure relief control system for overpressure of closed space

Abstract

The application discloses a kind of intelligent pressure relief control systems for closed space overpressure, including historical pressure signal acquisition module, historical pressure fluctuation event identification module, historical time domain feature extraction module, judgment reference set establishment module, it is related to pressure automatic control technical field, the judgment reference set establishment module is constructed judgment reference set by the present application, judgment reference set can self-adaptively determine the rate boundary, amplitude boundary and time length boundary of distinguishing water hammer type pseudo overshoot event and real overpressure event, and then in pressure relief decision module, whether the current pressure fluctuation event is determined as water hammer type pseudo overshoot event is judged, and water hammer type pseudo overshoot event does not trigger any pressure relief control action, relieve the water hammer type pressure peak generated in the moment that patient breathing mask exhaust valve is quickly closed in the multiple person oxygen cabin is misjudged as real cabin pressure rising trend, and then cause pressure relief valve invalid action, aggravate valve core wear and produce unnecessary cabin noise problem.

Description

Technical Field

[0001] This invention relates to the field of automatic pressure control technology, and in particular to an intelligent pressure relief control system for overpressure in confined spaces. Background Technology

[0002] Existing intelligent pressure relief control systems for confined space overpressure typically acquire pressure signals within the confined space and determine whether a real overpressure has occurred based on a preset pressure threshold or a single pressure change rate threshold. When the pressure signal exceeds the preset threshold, a pressure relief control operation corresponding to the confined space overpressure protection is triggered. In confined space applications such as the treatment chamber of a multi-person medical hyperbaric oxygen chamber, patients wear breathing masks and their exhaled air is directly discharged outside the chamber through the oxygen exhaust tubing. During this process, the pressure signal inside the chamber fluctuates with the patient's breathing movements.

[0003] Currently, in the actual operation of multi-person hyperbaric oxygen therapy chambers, a short-lived water hammer pressure spike is generated within the chamber when the exhalation valve of the breathing mask worn by the patient closes rapidly. After being collected by a pressure sensor and smoothed by conventional filtering, this water hammer pressure spike forms a pseudo-overshoot signal with an extremely short duration, a significant pressure peak amplitude, and extremely high rates of pressure rise and fall. Currently, intelligent pressure relief control systems easily misinterpret this water hammer pressure spike as a genuine increase in chamber pressure, thus executing an unnecessary micro-pressure relief action. This unnecessary pressure relief causes the pressure to momentarily drop slightly below the set value, prompting the system to replenish air again. This repeated cycle causes the pressure relief valve to frequently and ineffectively operate even without genuine overpressure, significantly accelerating valve core wear and generating unnecessary chamber noise. Summary of the Invention

[0004] To address the technical problems existing in the background art, the present invention proposes an intelligent pressure relief control system for overpressure in confined spaces.

[0005] This invention proposes an intelligent pressure relief control system for overpressure in confined spaces, comprising a historical pressure signal acquisition module, a historical pressure fluctuation event identification module, a historical time-domain feature extraction module, a judgment benchmark set establishment module, a current pressure signal acquisition module, a current pressure fluctuation event identification module, a current time-domain feature extraction module, and a pressure relief decision module; wherein: Historical pressure signal acquisition module: acquires historical pressure signals at multiple sampling times within the confined space, arranges the historical pressure signals at multiple sampling times in chronological order to form a historical pressure signal sequence, and each historical pressure signal includes the historical pressure value of the confined space. Historical pressure fluctuation event identification module: Identifies multiple historical pressure fluctuation events from historical pressure signal sequences; Historical temporal feature extraction module: Extracts temporal feature parameters for each historical pressure fluctuation event. The temporal feature parameters include: pressure rise rate, pressure fall rate, pressure peak amplitude, and pressure fluctuation duration. Judgment benchmark set establishment module: Based on the time domain characteristic parameters of each historical pressure fluctuation event, a judgment benchmark set is established to identify water hammer pseudo overshoot events. The judgment benchmark set includes: upper limit benchmark for rise rate, upper limit benchmark for fall rate, lower limit benchmark for amplitude, and upper limit benchmark for duration. Current pressure signal acquisition module: acquires pressure signals at multiple sampling times within the confined space, arranges the pressure signals at multiple sampling times in chronological order to form the current pressure signal sequence, and each pressure signal includes the pressure value of the confined space; Current pressure fluctuation event identification module: Identifies current pressure fluctuation events from the current pressure signal sequence; Current time-domain feature extraction module: Extracts the time-domain feature parameters of the current pressure fluctuation event; Pressure relief decision module: Based on the judgment benchmark set, determine whether the current pressure fluctuation event is a water hammer pseudo-overshoot event, and do not trigger any pressure relief control action for water hammer pseudo-overshoot events.

[0006] Preferably, the historical pressure fluctuation event identification module identifies multiple historical pressure fluctuation events from the historical pressure signal sequence, as follows: Extract the historical pressure value of the confined space corresponding to each historical pressure signal from the historical pressure signal sequence to form a historical pressure value sequence; To clarify, forming a historical pressure value sequence means that after obtaining all historical closed space pressure values, the historical pressure values ​​are sorted according to the order of the historical pressure signals corresponding to the historical closed space pressure values ​​in the historical pressure signal sequence to obtain the historical pressure value sequence.

[0007] The first-order difference is performed on the historical pressure values ​​of the confined space corresponding to two adjacent sampling times in the historical pressure value sequence to obtain the difference result. The difference result is used as the pressure change rate of the later sampling time in the two adjacent sampling times. Obtain all pressure change rates, and arrange all pressure change rates in chronological order according to the sampling time corresponding to each pressure change rate to form a pressure change rate sequence. Obtain the absolute value of the pressure change rate for each pressure change rate in the pressure change rate sequence to form a sequence of absolute pressure change rate values. Set an absolute value threshold for the rate of change and a rate of change threshold, with the rate of change threshold being greater than the absolute value threshold for the rate of change; In the absolute value sequence of pressure change rate, if there are N1 consecutive sampling times where the absolute value of pressure change rate is less than or equal to the absolute value threshold of the change rate, the last sampling time corresponding to the N1 consecutive sampling times is marked as the candidate event start point. In the pressure change rate sequence, if the pressure change rate is greater than the change rate threshold for N2 consecutive sampling times after the candidate event start point, the candidate event start point is marked as the event start point, where N1 and N2 are both positive integers greater than 1. In the absolute value sequence of pressure change rate, after a certain event start point, when it is detected that there are N3 consecutive sampling times where the absolute value of pressure change rate is less than or equal to the absolute value threshold of the change rate, the starting sampling time corresponding to the N3 consecutive sampling times is marked as the event end point corresponding to the event start point, where N3 is a positive integer greater than 1; Based on the start point of an event and the end point of the event corresponding to the start point, extract the historical pressure values ​​of the enclosed space from the sampling time corresponding to the start point of the event to the sampling time corresponding to the end point of the event in the historical pressure value sequence to form an event pressure value subsequence. The event pressure value subsequence is regarded as a historical pressure fluctuation event. The sampling time corresponding to the start point of the historical pressure fluctuation event is taken as the start time of the historical pressure fluctuation event, and the sampling time corresponding to the end point of the historical pressure fluctuation event is taken as the end time of the historical pressure fluctuation event. The time interval consisting of the start time and the end time is taken as the time interval of the historical pressure fluctuation event. Obtain all historical pressure fluctuation events to form a set of historical pressure fluctuation events.

[0008] Preferably, in the historical time-domain feature extraction module, the time-domain feature parameters of each historical pressure fluctuation event are extracted as follows: For any historical pressure fluctuation event: Extract the maximum value of the historical pressure in the confined space during the historical pressure fluctuation event as the highest pressure value of the event; obtain the sampling time corresponding to the highest pressure value of the event as the sampling time of the highest pressure value of the event; if there are multiple corresponding sampling times for the highest pressure value of the event, take the earliest sampling time as the sampling time of the highest pressure value of the event. Extract the historical pressure value of the enclosed space corresponding to the start point of the historical pressure fluctuation event, and use it as the stable pressure value before the start of the event. Subtracting the stable pressure value before the event begins from the highest pressure value of the event yields the peak pressure amplitude of the historical pressure fluctuation event. Starting from the sampling time corresponding to the beginning point of the historical pressure fluctuation event, and ending from the sampling time of the highest pressure value of the event, a pressure rise phase is formed. During this pressure rise phase, each sampling moment after the starting point is taken as the rise sampling moment; For clarification, each sampling time after the starting point does not include the sampling time corresponding to the starting point; Extract the pressure change rate corresponding to each rising sampling moment from the pressure change rate sequence as the rising stage pressure change rate; obtain all rising stage pressure change rates to form a set of rising stage pressure change rates. From the set of pressure change rates during the rising phase, obtain the maximum value of the pressure change rate during the rising phase, and use it as the pressure rise rate of this historical pressure fluctuation event. Starting from the sampling time of the highest pressure value of the event, and ending with the sampling time corresponding to the end point of the event in the historical pressure fluctuation event, a pressure decline phase is formed. During this pressure fallback phase, each sampling moment after the starting point will be taken as the fallback sampling moment; Extract the absolute value of the pressure change rate corresponding to each sampling time of the fall back from the absolute value sequence of pressure change rate, and use it as the absolute value of the pressure change rate during the fall back phase; obtain all the absolute values ​​of the pressure change rate during the fall back phase to form a set of absolute values ​​of the pressure change rate during the fall back phase. From the set of absolute values ​​of pressure change rate during the decline phase, obtain the maximum value of the absolute value of pressure change rate during the decline phase, and take it as the pressure decline rate of this historical pressure fluctuation event. The time difference between the end point and the start point of the historical pressure fluctuation event is taken as the duration of the pressure fluctuation of the historical pressure fluctuation event. For clarification, the time difference between the end point and the start point of the historical pressure fluctuation event refers to the sampling time corresponding to the end point of the historical pressure fluctuation event minus the sampling time corresponding to the start point of the historical pressure fluctuation event.

[0009] Preferably, in the judgment benchmark set establishment module, a judgment benchmark set for identifying water hammer pseudo-overshoot events is established based on the time-domain characteristic parameters of each historical pressure fluctuation event, as follows: Set a first statistical time window, a second statistical time window, and a third statistical time window. The duration of the first statistical time window, the second statistical time window, and the third statistical time window is a fixed duration, and the duration of the second statistical time window is longer than the duration of the first statistical time window. Obtain the sampling time corresponding to the last historical pressure signal in the historical pressure signal sequence as the reference time; The duration of the first statistical time window is taken as the first fixed duration; The baseline time is taken as the end point of the first statistical time window, and a first fixed duration is extracted before it to serve as the current time range of the first statistical time window; Extract historical pressure fluctuation events from the set of historical pressure fluctuation events whose time intervals fall within the first statistical time window, and use them as the first statistical sample. Obtain all the first statistical samples to form the first statistical sample set; In the first statistical sample set, for any first statistical sample, the pressure rise rate corresponding to the first statistical sample is obtained as the sample pressure rise rate; Obtain the pressure rise rate of all samples and form a set of sample pressure rise rates; For the set of sample pressure rise rates, the empirical cumulative distribution function method is used to obtain the empirical cumulative distribution curve of the sample pressure rise rate; The second derivative of the empirical cumulative distribution curve of the sample pressure rise rate is obtained, and the sample pressure rise rate corresponding to the point where the absolute value of the second derivative on the empirical cumulative distribution curve of the sample pressure rise rate is the upper limit benchmark of the rise rate. In the first statistical sample set, for any first statistical sample, the pressure drop rate corresponding to that first statistical sample is obtained as the sample pressure drop rate; Obtain the pressure drop rate of all samples and form a set of sample pressure drop rates; For the set of sample pressure decrease rates, the empirical cumulative distribution function method is used to obtain the empirical cumulative distribution curve of the sample pressure decrease rate; The second derivative of the empirical cumulative distribution curve of the sample pressure decrease rate is obtained, and the sample pressure decrease rate corresponding to the point where the absolute value of the second derivative on the empirical cumulative distribution curve of the sample pressure decrease rate is the upper limit benchmark of the decrease rate. The duration of the second statistical time window is used as the second fixed duration; The baseline time is taken as the end point of the second statistical time window, and a second fixed duration is extracted before it to serve as the current time range of the second statistical time window. Extract historical pressure fluctuation events from the set of historical pressure fluctuation events whose time intervals fall within the current time range of the second statistical time window, and use them as the second statistical sample; Obtain all second statistical samples to form a second statistical sample set; In the second statistical sample set, for any second statistical sample, the pressure peak amplitude corresponding to that second statistical sample is obtained as the sample pressure peak amplitude; Obtain the peak stress amplitude of all samples to form a sample peak stress amplitude set; Obtain the mean of the peak pressure amplitudes of all samples in the sample peak pressure amplitude set, and use it as the mean of peak pressure amplitudes; obtain the standard deviation of the peak pressure amplitudes of all samples in the sample peak pressure amplitude set, and use it as the standard deviation of peak pressure amplitudes; Add 3 to the mean of the peak pressure amplitude and multiply by the standard deviation of the peak pressure amplitude to obtain the lower limit of the amplitude. As an explanation, the mean of the peak pressure amplitude is added to 3 and multiplied by the standard deviation of the peak pressure amplitude, which adopts the existing three sigma criterion.

[0010] The duration of the third statistical time window is taken as the third fixed duration; The baseline time is taken as the end point of the third statistical time window, and a third fixed duration is extracted before it to serve as the current time range of the third statistical time window. Extract historical pressure fluctuation events from the set of historical pressure fluctuation events whose time intervals fall within the third statistical time window, and use them as the third statistical sample. Obtain all third statistical samples to form a third statistical sample set; In the third statistical sample set, for any third statistical sample, the duration of pressure fluctuation corresponding to the third statistical sample is obtained as the duration of sample pressure fluctuation. Obtain the duration of pressure fluctuations for all samples to form a set of sample pressure fluctuation durations. Obtain the mean of the duration of pressure fluctuations for all samples in the set of sample pressure fluctuation durations, and use it as the mean of the duration of pressure fluctuations; obtain the standard deviation of the duration of pressure fluctuations for all samples in the set of sample pressure fluctuation durations, and use it as the standard deviation of the duration of pressure fluctuations. Add 3 to the mean duration of pressure fluctuations and multiply by the standard deviation of the duration of pressure fluctuations to get the upper limit benchmark of the duration. As an explanation, the average duration of pressure fluctuations is multiplied by 3 by the standard deviation of the duration of pressure fluctuations, which is the existing three-sigma criterion.

[0011] Preferably, in the current pressure fluctuation event identification module, the current pressure fluctuation event is identified from the current pressure signal sequence as follows: Extract the pressure value of the confined space corresponding to the current pressure signal sequence to form a pressure value sequence; To illustrate, forming a pressure value sequence means that after obtaining all the pressure values ​​of the confined space, the pressure values ​​of the multiple confined spaces are sorted according to the arrangement order of the corresponding pressure signals in the current pressure signal sequence to obtain the pressure value sequence.

[0012] The pressure values ​​of the confined space corresponding to two adjacent sampling times in the pressure value sequence are subjected to first-order difference to obtain the difference result. The difference result is used as the current pressure change rate of the later sampling time in the two adjacent sampling times. Obtain all current pressure change rates, and arrange all current pressure change rates in chronological order according to the sampling time corresponding to each current pressure change rate to form a current pressure change rate sequence. Obtain the absolute value of the current pressure change rate for each current pressure change rate in the current pressure change rate sequence, forming a sequence of absolute values ​​of the current pressure change rate; In the current pressure change rate absolute value sequence, if there are N1 consecutive sampling times where the current pressure change rate absolute value is less than or equal to the change rate absolute value threshold, the last sampling time corresponding to the N1 consecutive sampling times is marked as the current candidate event start point. In the current pressure change rate sequence, if the current pressure change rate is greater than the change rate threshold for N2 consecutive sampling times after the current candidate event start point, the current candidate event start point is marked as the current event start point; In the current pressure change rate absolute value sequence, after a certain current event start point, when it is detected that there are N3 consecutive sampling times where the current pressure change rate absolute value is less than or equal to the change rate absolute value threshold, the starting sampling time corresponding to the N3 consecutive sampling times is marked as the current event end point corresponding to the current event start point. Based on a certain current event start point and the current event end point corresponding to the current event start point, extract the closed space pressure values ​​from the pressure value sequence at all sampling times between the sampling time corresponding to the current event start point and the sampling time corresponding to the current event end point to form a current event pressure value subsequence, and take the current event pressure value subsequence as the current pressure fluctuation event; As an explanation, if the current pressure fluctuation event is not present in the current pressure signal sequence, the process returns to the current pressure signal acquisition module, continues to acquire pressure signals for a future period of time, and then executes the current pressure fluctuation event identification module.

[0013] Preferably, in the current time-domain feature extraction module, the time-domain feature parameters of the current pressure fluctuation event are extracted as follows: Extract the maximum value of the pressure in the confined space during the current pressure fluctuation event, and take it as the highest pressure value of the current event; obtain the sampling time corresponding to the highest pressure value of the current event, and take it as the sampling time of the highest pressure value of the current event; if there are multiple corresponding sampling times for the highest pressure value of the current event, take the earliest sampling time as the sampling time of the highest pressure value of the current event. Extract the pressure value of the enclosed space corresponding to the start point of the current pressure fluctuation event, and use it as the stable pressure value before the start of the current event. Subtracting the stable pressure value before the start of the current event from the highest pressure value of the current event yields the peak pressure amplitude of the current pressure fluctuation event. The current pressure rise phase is formed by taking the sampling time corresponding to the start point of the current event in the current pressure fluctuation event as the starting point and the sampling time of the highest pressure value of the current event as the ending point. During this current pressure rise phase, each sampling moment after the starting point is taken as the current rise sampling moment; For clarification, each sampling time after the starting point does not include the sampling time corresponding to the starting point; Extract the current pressure change rate corresponding to each current rising sampling moment from the current pressure change rate sequence, and use it as the current rising stage pressure change rate; obtain all current rising stage pressure change rates to form a set of current rising stage pressure change rates. From the set of pressure change rates during the current upward phase, obtain the maximum value of the pressure change rate during the current upward phase, and use it as the pressure rise rate of the current pressure fluctuation event. Starting from the sampling time of the highest pressure value of the current event, and ending with the sampling time corresponding to the end point of the current event in the current pressure fluctuation event, the current pressure decline phase is formed. During this current pressure decline phase, each sampling moment after the starting point will be taken as the current decline sampling moment; Extract the absolute value of the current pressure change rate corresponding to each current pullback sampling moment from the absolute value sequence of pressure change rates, and use it as the absolute value of the pressure change rate of the current pullback stage; obtain all the absolute values ​​of the pressure change rate of the current pullback stage to form a set of absolute values ​​of the pressure change rate of the current pullback stage. From the set of absolute values ​​of the pressure change rate during the current pullback phase, obtain the maximum value of the absolute value of the pressure change rate during the current pullback phase, and use it as the pressure decrease rate of the current pressure fluctuation event. The time difference between the end point and the start point of the current pressure fluctuation event is taken as the duration of the pressure fluctuation in the current pressure fluctuation event. For clarification, the time difference between the end point and the start point of the current pressure fluctuation event refers to the sampling time corresponding to the end point of the current pressure fluctuation event minus the sampling time corresponding to the start point of the current pressure fluctuation event.

[0014] Preferably, in the pressure relief decision module, based on the judgment benchmark set, it is determined whether the current pressure fluctuation event is a water hammer-type pseudo-overshoot event, as follows: If the pressure rise rate of the current pressure fluctuation event is greater than the upper limit benchmark of the rise rate, the pressure fall rate of the current pressure fluctuation event is greater than the upper limit benchmark of the fall rate, the pressure peak amplitude of the current pressure fluctuation event is greater than the lower limit benchmark of the amplitude, and the pressure fluctuation duration of the current pressure fluctuation event is less than the upper limit benchmark of the duration, the current pressure fluctuation event will be judged as a water hammer pseudo-overshoot event. As an explanation, the fact that the pressure relief decision module does not trigger any pressure relief control action for water hammer pseudo-overflow events means that the current state of the pressure relief valve remains unchanged, and the response to water hammer pseudo-overflow events is directly skipped in the pressure relief control logic. When not all conditions in the pressure relief decision module are met, the pressure relief control operation corresponding to the overpressure protection of the confined space is executed normally.

[0015] The intelligent pressure relief control system for overpressure in confined spaces proposed in this invention has the following beneficial technical effects: This application establishes a benchmark set based on the temporal characteristic parameters of each historical pressure fluctuation event through a benchmark set establishment module. It calculates the second derivative of the empirical cumulative distribution curve of the sample pressure rise rate, and uses the sample pressure rise rate corresponding to the point with the largest absolute value of the second derivative on the empirical cumulative distribution curve as the upper limit benchmark for the rise rate. Similarly, it calculates the second derivative of the empirical cumulative distribution curve of the sample pressure fall rate, and uses the sample pressure fall rate corresponding to the point with the largest absolute value of the second derivative on the empirical cumulative distribution curve of the sample pressure fall rate as the upper limit benchmark for the fall rate. Furthermore, it employs a design that adds 3 to the average pressure peak amplitude and multiplies it by the standard deviation of the pressure peak amplitude as the lower limit benchmark for amplitude, and adds 3 to the average pressure fluctuation duration and multiplies it by the standard deviation of the pressure fluctuation duration as the upper limit benchmark for duration. This design allows the benchmark set to be derived from historical data. The statistical distribution characteristics of the pressure fluctuation event set adaptively determine the rate boundary, amplitude boundary, and duration boundary to distinguish between water hammer pseudo-overshoot events and real overpressure events. Then, in the pressure relief decision module, if the pressure rise rate of the current pressure fluctuation event is greater than the upper limit benchmark of the rise rate, the pressure fall rate of the current pressure fluctuation event is greater than the upper limit benchmark of the fall rate, the pressure peak amplitude of the current pressure fluctuation event is greater than the lower limit benchmark of the amplitude, and the pressure fluctuation duration of the current pressure fluctuation event is less than the upper limit benchmark of the duration, then the current pressure fluctuation event is judged as a water hammer pseudo-overshoot event, and no pressure relief control action is triggered for water hammer pseudo-overshoot events. This effectively alleviates the problem that water hammer pressure spikes generated when the patient's breathing mask exhalation valve closes rapidly in a multi-person oxygen chamber are misjudged as a real chamber pressure rise trend, which leads to ineffective action of the pressure relief valve, aggravated valve core wear, and unnecessary chamber noise. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the system of the present invention. Detailed Implementation

[0017] Embodiments of the present invention are described in detail below. Examples of these embodiments are illustrated in the accompanying drawings, wherein the same or similar symbols denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0018] like Figure 1 The diagram illustrates an intelligent pressure relief control system for overpressure in confined spaces, suitable for pressure safety protection in oxygen chambers. It includes a historical pressure signal acquisition module, a historical pressure fluctuation event identification module, a historical time-domain feature extraction module, a judgment benchmark set establishment module, a current pressure signal acquisition module, a current pressure fluctuation event identification module, a current time-domain feature extraction module, and a pressure relief decision module. Wherein: Historical pressure signal acquisition module: acquires historical pressure signals at multiple sampling times within the confined space, arranges the historical pressure signals at multiple sampling times in chronological order to form a historical pressure signal sequence, and each historical pressure signal includes the historical pressure value of the confined space. Historical pressure fluctuation event identification module: Identifies multiple historical pressure fluctuation events from historical pressure signal sequences; In an optional embodiment, the historical pressure fluctuation event identification module identifies multiple historical pressure fluctuation events from the historical pressure signal sequence, as follows: Extract the historical pressure value of the confined space corresponding to each historical pressure signal from the historical pressure signal sequence to form a historical pressure value sequence; To clarify, forming a historical pressure value sequence means that after obtaining all historical closed space pressure values, the historical pressure values ​​are sorted according to the order of the historical pressure signals corresponding to the historical closed space pressure values ​​in the historical pressure signal sequence to obtain the historical pressure value sequence.

[0019] The first-order difference is performed on the historical pressure values ​​of the confined space corresponding to two adjacent sampling times in the historical pressure value sequence to obtain the difference result. The difference result is used as the pressure change rate of the later sampling time in the two adjacent sampling times. Obtain all pressure change rates, and arrange all pressure change rates in chronological order according to the sampling time corresponding to each pressure change rate to form a pressure change rate sequence. Obtain the absolute value of the pressure change rate for each pressure change rate in the pressure change rate sequence to form a sequence of absolute pressure change rate values. Set an absolute value threshold for the rate of change and a rate of change threshold, with the rate of change threshold being greater than the absolute value threshold for the rate of change; In the absolute value sequence of pressure change rate, if there are N1 consecutive sampling times where the absolute value of pressure change rate is less than or equal to the absolute value threshold of the change rate, the last sampling time corresponding to the N1 consecutive sampling times is marked as the candidate event start point. In the pressure change rate sequence, if the pressure change rate is greater than the change rate threshold for N2 consecutive sampling times after the candidate event start point, the candidate event start point is marked as the event start point, where N1 and N2 are both positive integers greater than 1. In the absolute value sequence of pressure change rate, after a certain event start point, when it is detected that there are N3 consecutive sampling times where the absolute value of pressure change rate is less than or equal to the absolute value threshold of the change rate, the starting sampling time corresponding to the N3 consecutive sampling times is marked as the event end point corresponding to the event start point, where N3 is a positive integer greater than 1; Based on the start point of an event and the end point of the event corresponding to the start point, extract the historical pressure values ​​of the enclosed space from the sampling time corresponding to the start point of the event to the sampling time corresponding to the end point of the event in the historical pressure value sequence to form an event pressure value subsequence. The event pressure value subsequence is regarded as a historical pressure fluctuation event. The sampling time corresponding to the start point of the historical pressure fluctuation event is taken as the start time of the historical pressure fluctuation event, and the sampling time corresponding to the end point of the historical pressure fluctuation event is taken as the end time of the historical pressure fluctuation event. The time interval consisting of the start time and the end time is taken as the time interval of the historical pressure fluctuation event. Obtain all historical pressure fluctuation events and form a set of historical pressure fluctuation events; In an optional embodiment, the absolute value threshold of the rate of change and the rate of change threshold can be generated as follows: During the initial sampling phase of the confined space, historical pressure values ​​of the confined space at multiple sampling times are continuously acquired. The first-order difference is performed on the historical pressure values ​​of the confined space corresponding to two adjacent sampling times to obtain the difference result. The difference result is used as the initial pressure change rate of the later sampling time in the two adjacent sampling times. Obtain all initial pressure change rates, and arrange all initial pressure change rates in chronological order according to the sampling time corresponding to each initial pressure change rate to form an initial pressure change rate sequence. Statistical analysis was performed on the absolute values ​​of each initial pressure change rate in the initial pressure change rate sequence to obtain the average value and the degree of fluctuation dispersion of the absolute values ​​of the pressure change rate. Let the average value be μ, the degree of fluctuation dispersion be σ, and the threshold value of the absolute value of the rate of change be T1, then: Where k1 is an adjustment coefficient greater than 0; Let the threshold of the rate of change be T2, then: Where k2 is greater than 1; The values ​​of k1 and k2 can be obtained by grid search or Bayesian optimization. Historical temporal feature extraction module: Extracts temporal feature parameters for each historical pressure fluctuation event. The temporal feature parameters include: pressure rise rate, pressure fall rate, pressure peak amplitude, and pressure fluctuation duration. In an optional embodiment, the historical time-domain feature extraction module extracts the time-domain feature parameters for each historical pressure fluctuation event, as follows: For any historical pressure fluctuation event: Extract the maximum value of the historical pressure in the confined space during the historical pressure fluctuation event as the highest pressure value of the event; obtain the sampling time corresponding to the highest pressure value of the event as the sampling time of the highest pressure value of the event; if there are multiple corresponding sampling times for the highest pressure value of the event, take the earliest sampling time as the sampling time of the highest pressure value of the event. Extract the historical pressure value of the enclosed space corresponding to the start point of the historical pressure fluctuation event, and use it as the stable pressure value before the start of the event. Subtracting the stable pressure value before the event begins from the highest pressure value of the event yields the peak pressure amplitude of the historical pressure fluctuation event. Starting from the sampling time corresponding to the beginning point of the historical pressure fluctuation event, and ending from the sampling time of the highest pressure value of the event, a pressure rise phase is formed. During this pressure rise phase, each sampling moment after the starting point is taken as the rise sampling moment; For clarification, each sampling time after the starting point does not include the sampling time corresponding to the starting point; Extract the pressure change rate corresponding to each rising sampling moment from the pressure change rate sequence as the rising stage pressure change rate; obtain all rising stage pressure change rates to form a set of rising stage pressure change rates. From the set of pressure change rates during the rising phase, obtain the maximum value of the pressure change rate during the rising phase, and use it as the pressure rise rate of this historical pressure fluctuation event. Starting from the sampling time of the highest pressure value of the event, and ending with the sampling time corresponding to the end point of the event in the historical pressure fluctuation event, a pressure decline phase is formed. During this pressure fallback phase, each sampling moment after the starting point will be taken as the fallback sampling moment; Extract the absolute value of the pressure change rate corresponding to each sampling time of the fall back from the absolute value sequence of pressure change rate, and use it as the absolute value of the pressure change rate during the fall back phase; obtain all the absolute values ​​of the pressure change rate during the fall back phase to form a set of absolute values ​​of the pressure change rate during the fall back phase. From the set of absolute values ​​of pressure change rate during the decline phase, obtain the maximum value of the absolute value of pressure change rate during the decline phase, and take it as the pressure decline rate of this historical pressure fluctuation event. The time difference between the end point and the start point of the historical pressure fluctuation event is taken as the duration of the pressure fluctuation of the historical pressure fluctuation event. For clarification, the time difference between the end point and the start point of the historical pressure fluctuation event refers to the sampling time corresponding to the end point of the historical pressure fluctuation event minus the sampling time corresponding to the start point of the historical pressure fluctuation event.

[0020] Judgment benchmark set establishment module: Based on the time domain characteristic parameters of each historical pressure fluctuation event, a judgment benchmark set is established to identify water hammer pseudo overshoot events. The judgment benchmark set includes: upper limit benchmark for rise rate, upper limit benchmark for fall rate, lower limit benchmark for amplitude, and upper limit benchmark for duration. In an optional embodiment, the judgment benchmark set establishment module establishes a judgment benchmark set for identifying water hammer pseudo-overshoot events based on the time-domain characteristic parameters of each historical pressure fluctuation event, as follows: Set a first statistical time window, a second statistical time window, and a third statistical time window. The duration of the first statistical time window, the second statistical time window, and the third statistical time window is a fixed duration, and the duration of the second statistical time window is longer than the duration of the first statistical time window. Obtain the sampling time corresponding to the last historical pressure signal in the historical pressure signal sequence as the reference time; The upper limit benchmarks for the rate of ascent and the upper limit benchmarks for the rate of descent are generated in the following manner: The duration of the first statistical time window is taken as the first fixed duration; The baseline time is taken as the end point of the first statistical time window, and a first fixed duration is extracted before it to serve as the current time range of the first statistical time window; Extract historical pressure fluctuation events from the set of historical pressure fluctuation events whose time intervals fall within the first statistical time window, and use them as the first statistical sample. Obtain all the first statistical samples to form the first statistical sample set; In the first statistical sample set, for any first statistical sample, the pressure rise rate corresponding to the first statistical sample is obtained as the sample pressure rise rate; Obtain the pressure rise rate of all samples and form a set of sample pressure rise rates; For the set of sample pressure rise rates, the empirical cumulative distribution function method is used to obtain the empirical cumulative distribution curve of the sample pressure rise rate; The second derivative of the empirical cumulative distribution curve of the sample pressure rise rate is obtained, and the sample pressure rise rate corresponding to the point where the absolute value of the second derivative on the empirical cumulative distribution curve of the sample pressure rise rate is the upper limit benchmark of the rise rate. In the first statistical sample set, for any first statistical sample, the pressure drop rate corresponding to that first statistical sample is obtained as the sample pressure drop rate; Obtain the pressure drop rate of all samples and form a set of sample pressure drop rates; For the set of sample pressure decrease rates, the empirical cumulative distribution function method is used to obtain the empirical cumulative distribution curve of the sample pressure decrease rate; The second derivative of the empirical cumulative distribution curve of the sample pressure decrease rate is obtained, and the sample pressure decrease rate corresponding to the point where the absolute value of the second derivative on the empirical cumulative distribution curve of the sample pressure decrease rate is the upper limit benchmark of the decrease rate. The duration of the second statistical time window is used as the second fixed duration; The baseline time is taken as the end point of the second statistical time window, and a second fixed duration is extracted before it to serve as the current time range of the second statistical time window. Extract historical pressure fluctuation events from the set of historical pressure fluctuation events whose time intervals fall within the current time range of the second statistical time window, and use them as the second statistical sample; Obtain all second statistical samples to form a second statistical sample set; In the second statistical sample set, for any second statistical sample, the pressure peak amplitude corresponding to that second statistical sample is obtained as the sample pressure peak amplitude; Obtain the peak stress amplitude of all samples to form a sample peak stress amplitude set; Obtain the mean of the peak pressure amplitudes of all samples in the sample peak pressure amplitude set, and use it as the mean of peak pressure amplitudes; obtain the standard deviation of the peak pressure amplitudes of all samples in the sample peak pressure amplitude set, and use it as the standard deviation of peak pressure amplitudes; Add 3 to the mean of the peak pressure amplitude and multiply by the standard deviation of the peak pressure amplitude to obtain the lower limit of the amplitude. As an explanation, the mean of the peak pressure amplitude is added to 3 and multiplied by the standard deviation of the peak pressure amplitude, which adopts the existing three sigma criterion.

[0021] The duration of the third statistical time window is taken as the third fixed duration; The baseline time is taken as the end point of the third statistical time window, and a third fixed duration is extracted before it to serve as the current time range of the third statistical time window. Extract historical pressure fluctuation events from the set of historical pressure fluctuation events whose time intervals fall within the third statistical time window, and use them as the third statistical sample. Obtain all third statistical samples to form a third statistical sample set; In the third statistical sample set, for any third statistical sample, the duration of pressure fluctuation corresponding to the third statistical sample is obtained as the duration of sample pressure fluctuation. Obtain the duration of pressure fluctuations for all samples to form a set of sample pressure fluctuation durations. Obtain the mean of the duration of pressure fluctuations for all samples in the set of sample pressure fluctuation durations, and use it as the mean of the duration of pressure fluctuations; obtain the standard deviation of the duration of pressure fluctuations for all samples in the set of sample pressure fluctuation durations, and use it as the standard deviation of the duration of pressure fluctuations. Add 3 to the mean duration of pressure fluctuations and multiply by the standard deviation of the duration of pressure fluctuations to get the upper limit benchmark of the duration. As an explanation, the average duration of pressure fluctuations is multiplied by 3 by the standard deviation of the duration of pressure fluctuations, which is the existing three-sigma criterion.

[0022] In an optional embodiment, the duration of the third statistical time window is greater than the duration of the first statistical time window; Current pressure signal acquisition module: acquires pressure signals at multiple sampling times within the confined space, arranges the pressure signals at multiple sampling times in chronological order to form the current pressure signal sequence, and each pressure signal includes the pressure value of the confined space; As an explanation, in the current pressure signal acquisition module, the current time is first acquired, and the sampling time corresponding to the pressure signal at the end of the current pressure signal sequence is the sampling time corresponding to the current time. Current pressure fluctuation event identification module: Identifies current pressure fluctuation events from the current pressure signal sequence; In an optional embodiment, the current pressure fluctuation event identification module identifies the current pressure fluctuation event from the current pressure signal sequence as follows: Extract the pressure value of the confined space corresponding to the current pressure signal sequence to form a pressure value sequence; To illustrate, forming a pressure value sequence means that after obtaining all the pressure values ​​of the confined space, the pressure values ​​of the multiple confined spaces are sorted according to the arrangement order of the corresponding pressure signals in the current pressure signal sequence to obtain the pressure value sequence.

[0023] The pressure values ​​of the confined space corresponding to two adjacent sampling times in the pressure value sequence are subjected to first-order difference to obtain the difference result. The difference result is used as the current pressure change rate of the later sampling time in the two adjacent sampling times. Obtain all current pressure change rates, and arrange all current pressure change rates in chronological order according to the sampling time corresponding to each current pressure change rate to form a current pressure change rate sequence. Obtain the absolute value of the current pressure change rate for each current pressure change rate in the current pressure change rate sequence, forming a sequence of absolute values ​​of the current pressure change rate; In the current pressure change rate absolute value sequence, if there are N1 consecutive sampling times where the current pressure change rate absolute value is less than or equal to the change rate absolute value threshold, the last sampling time corresponding to the N1 consecutive sampling times is marked as the current candidate event start point. In the current pressure change rate sequence, if the current pressure change rate is greater than the change rate threshold for N2 consecutive sampling times after the current candidate event start point, the current candidate event start point is marked as the current event start point; In the current pressure change rate absolute value sequence, after a certain current event start point, when it is detected that there are N3 consecutive sampling times where the current pressure change rate absolute value is less than or equal to the change rate absolute value threshold, the starting sampling time corresponding to the N3 consecutive sampling times is marked as the current event end point corresponding to the current event start point. Based on a certain current event start point and the current event end point corresponding to the current event start point, extract the closed space pressure values ​​from the pressure value sequence at all sampling times between the sampling time corresponding to the current event start point and the sampling time corresponding to the current event end point to form a current event pressure value subsequence, and take the current event pressure value subsequence as the current pressure fluctuation event; As an explanation, if the current pressure fluctuation event is not present in the current pressure signal sequence, the process returns to the current pressure signal acquisition module, continues to acquire pressure signals for a future period of time, and then executes the current pressure fluctuation event identification module.

[0024] Current time-domain feature extraction module: Extracts the time-domain feature parameters of the current pressure fluctuation event; In an optional embodiment, the current time-domain feature extraction module extracts the time-domain feature parameters of the current pressure fluctuation event as follows: Extract the maximum value of the pressure in the confined space during the current pressure fluctuation event, and take it as the highest pressure value of the current event; obtain the sampling time corresponding to the highest pressure value of the current event, and take it as the sampling time of the highest pressure value of the current event; if there are multiple corresponding sampling times for the highest pressure value of the current event, take the earliest sampling time as the sampling time of the highest pressure value of the current event. Extract the pressure value of the enclosed space corresponding to the start point of the current pressure fluctuation event, and use it as the stable pressure value before the start of the current event. Subtracting the stable pressure value before the start of the current event from the highest pressure value of the current event yields the peak pressure amplitude of the current pressure fluctuation event. The current pressure rise phase is formed by taking the sampling time corresponding to the start point of the current event in the current pressure fluctuation event as the starting point and the sampling time of the highest pressure value of the current event as the ending point. During this current pressure rise phase, each sampling moment after the starting point is taken as the current rise sampling moment; For clarification, each sampling time after the starting point does not include the sampling time corresponding to the starting point; Extract the current pressure change rate corresponding to each current rising sampling moment from the current pressure change rate sequence, and use it as the current rising stage pressure change rate; obtain all current rising stage pressure change rates to form a set of current rising stage pressure change rates. From the set of pressure change rates during the current upward phase, obtain the maximum value of the pressure change rate during the current upward phase, and use it as the pressure rise rate of the current pressure fluctuation event. Starting from the sampling time of the highest pressure value of the current event, and ending with the sampling time corresponding to the end point of the current event in the current pressure fluctuation event, the current pressure decline phase is formed. During this current pressure decline phase, each sampling moment after the starting point will be taken as the current decline sampling moment; Extract the absolute value of the current pressure change rate corresponding to each current pullback sampling moment from the absolute value sequence of pressure change rates, and use it as the absolute value of the pressure change rate of the current pullback stage; obtain all the absolute values ​​of the pressure change rate of the current pullback stage to form a set of absolute values ​​of the pressure change rate of the current pullback stage. From the set of absolute values ​​of the pressure change rate during the current pullback phase, obtain the maximum value of the absolute value of the pressure change rate during the current pullback phase, and use it as the pressure decrease rate of the current pressure fluctuation event. The time difference between the end point and the start point of the current pressure fluctuation event is taken as the duration of the pressure fluctuation in the current pressure fluctuation event. For clarification, the time difference between the end point and the start point of the current pressure fluctuation event refers to the sampling time corresponding to the end point of the current pressure fluctuation event minus the sampling time corresponding to the start point of the current pressure fluctuation event.

[0025] Pressure relief decision module: Based on the judgment benchmark set, determine whether the current pressure fluctuation event is a water hammer pseudo-overshoot event, and do not trigger any pressure relief control action for water hammer pseudo-overshoot events.

[0026] In an optional embodiment, the pressure relief decision module determines whether the current pressure fluctuation event is a water hammer pseudo-overshoot event based on the judgment benchmark set, as follows: If the pressure rise rate of the current pressure fluctuation event is greater than the upper limit benchmark of the rise rate, the pressure fall rate of the current pressure fluctuation event is greater than the upper limit benchmark of the fall rate, the pressure peak amplitude of the current pressure fluctuation event is greater than the lower limit benchmark of the amplitude, and the pressure fluctuation duration of the current pressure fluctuation event is less than the upper limit benchmark of the duration, the current pressure fluctuation event will be judged as a water hammer pseudo-overshoot event. As an explanation, the fact that the pressure relief decision module does not trigger any pressure relief control action for water hammer pseudo-overflow events means that the current state of the pressure relief valve remains unchanged, and the response to water hammer pseudo-overflow events is directly skipped in the pressure relief control logic. When the conditions in the pressure relief decision module are not all met, the pressure relief control operation corresponding to the overpressure protection of the confined space is executed normally. As an explanation, the enclosed space can be the treatment chamber of a multi-person medical hyperbaric oxygen chamber, and the water hammer pseudo-overflow event can be caused by the pressure surge caused by the rapid closure of the exhalation valve of the breathing mask worn by the patient inside the chamber.

[0027] This application establishes a benchmark set based on the temporal characteristic parameters of each historical pressure fluctuation event through a benchmark set establishment module. It calculates the second derivative of the empirical cumulative distribution curve of the sample pressure rise rate, and uses the sample pressure rise rate corresponding to the point with the largest absolute value of the second derivative on the empirical cumulative distribution curve as the upper limit benchmark for the rise rate. Similarly, it calculates the second derivative of the empirical cumulative distribution curve of the sample pressure fall rate, and uses the sample pressure fall rate corresponding to the point with the largest absolute value of the second derivative on the empirical cumulative distribution curve of the sample pressure fall rate as the upper limit benchmark for the fall rate. Furthermore, it employs a design that adds 3 to the average pressure peak amplitude and multiplies it by the standard deviation of the pressure peak amplitude as the lower limit benchmark for amplitude, and adds 3 to the average pressure fluctuation duration and multiplies it by the standard deviation of the pressure fluctuation duration as the upper limit benchmark for duration. This design allows the benchmark set to be derived from historical data. The statistical distribution characteristics of the pressure fluctuation event set adaptively determine the rate boundary, amplitude boundary, and duration boundary to distinguish between water hammer pseudo-overshoot events and real overpressure events. Then, in the pressure relief decision module, if the pressure rise rate of the current pressure fluctuation event is greater than the upper limit benchmark of the rise rate, the pressure fall rate of the current pressure fluctuation event is greater than the upper limit benchmark of the fall rate, the pressure peak amplitude of the current pressure fluctuation event is greater than the lower limit benchmark of the amplitude, and the pressure fluctuation duration of the current pressure fluctuation event is less than the upper limit benchmark of the duration, then the current pressure fluctuation event is judged as a water hammer pseudo-overshoot event, and no pressure relief control action is triggered for water hammer pseudo-overshoot events. This effectively alleviates the problem that water hammer pressure spikes generated when the patient's breathing mask exhalation valve closes rapidly in a multi-person oxygen chamber are misjudged as a real chamber pressure rise trend, which leads to ineffective action of the pressure relief valve, aggravated valve core wear, and unnecessary chamber noise.

[0028] For clarification, "acquisition" in this application refers to obtaining the required content or data using existing technical means.

[0029] Furthermore, any content not described in detail in this specification is existing technology known to those skilled in the art. In the embodiments provided by this invention, it should be understood that the disclosed system or method can be implemented in other ways. For example, the embodiments of the invention described above are merely illustrative; for instance, the division of modules is only a logical functional division, and there may be other division methods in actual implementation.

[0030] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0031] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated module can be implemented in hardware or in the form of hardware plus software functional modules.

[0032] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the basic characteristics of the present invention.

[0033] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An intelligent pressure relief control system for overpressure in a confined space, characterized in that, include: Historical pressure signal acquisition module: acquires historical pressure signals at multiple sampling times within the confined space, arranges the historical pressure signals at multiple sampling times in chronological order to form a historical pressure signal sequence, and each historical pressure signal includes the historical pressure value of the confined space. Historical pressure fluctuation event identification module: Identifies multiple historical pressure fluctuation events from historical pressure signal sequences; Historical temporal feature extraction module: Extracts temporal feature parameters for each historical pressure fluctuation event. The temporal feature parameters include: pressure rise rate, pressure fall rate, pressure peak amplitude, and pressure fluctuation duration. Judgment benchmark set establishment module: Based on the time domain characteristic parameters of each historical pressure fluctuation event, a judgment benchmark set is established to identify water hammer pseudo overshoot events. The judgment benchmark set includes: upper limit benchmark for rise rate, upper limit benchmark for fall rate, lower limit benchmark for amplitude, and upper limit benchmark for duration. Current pressure signal acquisition module: acquires pressure signals at multiple sampling times within the confined space, arranges the pressure signals at multiple sampling times in chronological order to form the current pressure signal sequence, and each pressure signal includes the pressure value of the confined space; Current pressure fluctuation event identification module: Identifies current pressure fluctuation events from the current pressure signal sequence; Current time-domain feature extraction module: Extracts the time-domain feature parameters of the current pressure fluctuation event; Pressure relief decision module: Based on the judgment benchmark set, determine whether the current pressure fluctuation event is a water hammer pseudo-overshoot event, and do not trigger any pressure relief control action for water hammer pseudo-overshoot events.

2. The intelligent pressure relief control system for overpressure in a confined space according to claim 1, characterized in that, The historical pressure fluctuation event identification module identifies multiple historical pressure fluctuation events from the historical pressure signal sequence, as follows: Extract the historical pressure value of the confined space corresponding to each historical pressure signal from the historical pressure signal sequence to form a historical pressure value sequence; The first-order difference is performed on the historical pressure values ​​of the confined space corresponding to two adjacent sampling times in the historical pressure value sequence to obtain the difference result. The difference result is used as the pressure change rate of the later sampling time in the two adjacent sampling times. Obtain all pressure change rates, and arrange all pressure change rates in chronological order according to the sampling time corresponding to each pressure change rate to form a pressure change rate sequence. Obtain the absolute value of the pressure change rate for each pressure change rate in the pressure change rate sequence to form a sequence of absolute pressure change rate values. Set an absolute value threshold for the rate of change and a rate of change threshold, with the rate of change threshold being greater than the absolute value threshold for the rate of change; In the absolute value sequence of pressure change rate, if there are N1 consecutive sampling times where the absolute value of pressure change rate is less than or equal to the absolute value threshold of the change rate, the last sampling time corresponding to the N1 consecutive sampling times is marked as the candidate event start point. In the pressure change rate sequence, if the pressure change rate is greater than the change rate threshold for N2 consecutive sampling times after the candidate event start point, the candidate event start point is marked as the event start point, where N1 and N2 are both positive integers greater than 1. In the absolute value sequence of pressure change rate, after a certain event start point, when it is detected that there are N3 consecutive sampling times where the absolute value of pressure change rate is less than or equal to the absolute value threshold of the change rate, the starting sampling time corresponding to the N3 consecutive sampling times is marked as the event end point corresponding to the event start point, where N3 is a positive integer greater than 1; Based on the start point of an event and the end point of the event corresponding to the start point, extract the historical pressure values ​​of the enclosed space from the sampling time corresponding to the start point of the event to the sampling time corresponding to the end point of the event in the historical pressure value sequence to form an event pressure value subsequence. The event pressure value subsequence is regarded as a historical pressure fluctuation event. The sampling time corresponding to the start point of the historical pressure fluctuation event is taken as the start time of the historical pressure fluctuation event, and the sampling time corresponding to the end point of the historical pressure fluctuation event is taken as the end time of the historical pressure fluctuation event. The time interval consisting of the start time and the end time is taken as the time interval of the historical pressure fluctuation event. Obtain all historical pressure fluctuation events to form a set of historical pressure fluctuation events.

3. The intelligent pressure relief control system for overpressure in a confined space according to claim 2, characterized in that, In the historical time-domain feature extraction module, the time-domain feature parameters of each historical pressure fluctuation event are extracted as follows: For any historical pressure fluctuation event: Extract the maximum value of the historical pressure in the confined space during the historical pressure fluctuation event as the highest pressure value of the event; obtain the sampling time corresponding to the highest pressure value of the event as the sampling time of the highest pressure value of the event. If there are multiple corresponding sampling times for the highest pressure value of an event, the earliest sampling time shall be taken as the sampling time for the highest pressure value of the event. Extract the historical pressure value of the enclosed space corresponding to the start point of the historical pressure fluctuation event, and use it as the stable pressure value before the start of the event. Subtracting the stable pressure value before the event begins from the highest pressure value of the event yields the peak pressure amplitude of the historical pressure fluctuation event. Starting from the sampling time corresponding to the beginning point of the historical pressure fluctuation event, and ending from the sampling time of the highest pressure value of the event, a pressure rise phase is formed. During this pressure rise phase, each sampling moment after the starting point is taken as the rise sampling moment; Extract the pressure change rate corresponding to each rising sampling moment from the pressure change rate sequence as the rising stage pressure change rate; obtain all rising stage pressure change rates to form a set of rising stage pressure change rates. From the set of pressure change rates during the rising phase, obtain the maximum value of the pressure change rate during the rising phase, and use it as the pressure rise rate of this historical pressure fluctuation event. Starting from the sampling time of the highest pressure value of the event, and ending with the sampling time corresponding to the end point of the event in the historical pressure fluctuation event, a pressure decline phase is formed. During this pressure drop phase, each sampling moment after the starting point will be taken as the drop sampling moment; Extract the absolute value of the pressure change rate corresponding to each sampling time of the fall back from the absolute value sequence of pressure change rate, and use it as the absolute value of the pressure change rate during the fall back phase; obtain all the absolute values ​​of the pressure change rate during the fall back phase to form a set of absolute values ​​of the pressure change rate during the fall back phase. From the set of absolute values ​​of pressure change rate during the decline phase, obtain the maximum value of the absolute value of pressure change rate during the decline phase, and take it as the pressure decline rate of this historical pressure fluctuation event. The time difference between the end point and the start point of the historical pressure fluctuation event is taken as the duration of the pressure fluctuation of that historical pressure fluctuation event.

4. The intelligent pressure relief control system for overpressure in a confined space according to claim 3, characterized in that, In the benchmark set establishment module, a benchmark set for identifying water hammer pseudo-overshoot events is established based on the time-domain characteristic parameters of each historical pressure fluctuation event, as follows: Set a first statistical time window, a second statistical time window, and a third statistical time window. The duration of the first statistical time window, the second statistical time window, and the third statistical time window is a fixed duration, and the duration of the second statistical time window is longer than the duration of the first statistical time window. Obtain the sampling time corresponding to the last historical pressure signal in the historical pressure signal sequence as the reference time; The duration of the first statistical time window is taken as the first fixed duration; The baseline time is taken as the end point of the first statistical time window, and a first fixed duration is extracted before it to serve as the current time range of the first statistical time window; Extract historical pressure fluctuation events from the set of historical pressure fluctuation events whose time intervals fall within the first statistical time window, and use them as the first statistical sample. Obtain all the first statistical samples to form the first statistical sample set; In the first statistical sample set, for any first statistical sample, the pressure rise rate corresponding to the first statistical sample is obtained as the sample pressure rise rate; Obtain the pressure rise rate of all samples and form a set of sample pressure rise rates; For the set of sample pressure rise rates, the empirical cumulative distribution function method is used to obtain the empirical cumulative distribution curve of the sample pressure rise rate; The second derivative of the empirical cumulative distribution curve of the sample pressure rise rate is obtained, and the sample pressure rise rate corresponding to the point where the absolute value of the second derivative on the empirical cumulative distribution curve of the sample pressure rise rate is the upper limit benchmark of the rise rate. In the first statistical sample set, for any first statistical sample, the pressure drop rate corresponding to that first statistical sample is obtained as the sample pressure drop rate; Obtain the pressure drop rate of all samples and form a set of sample pressure drop rates; For the set of sample pressure decrease rates, the empirical cumulative distribution function method is used to obtain the empirical cumulative distribution curve of the sample pressure decrease rate; The second derivative of the empirical cumulative distribution curve of the sample pressure decrease rate is obtained, and the sample pressure decrease rate corresponding to the point where the absolute value of the second derivative on the empirical cumulative distribution curve of the sample pressure decrease rate is the upper limit benchmark of the decrease rate. The duration of the second statistical time window is used as the second fixed duration; The baseline time is taken as the end point of the second statistical time window, and a second fixed duration is extracted before it to serve as the current time range of the second statistical time window. Extract historical pressure fluctuation events from the set of historical pressure fluctuation events whose time intervals fall within the current time range of the second statistical time window, and use them as the second statistical sample; Obtain all second statistical samples to form a second statistical sample set; In the second statistical sample set, for any second statistical sample, the pressure peak amplitude corresponding to that second statistical sample is obtained as the sample pressure peak amplitude; Obtain the peak stress amplitude of all samples to form a sample peak stress amplitude set; Obtain the mean of the peak pressure amplitudes of all samples in the sample peak pressure amplitude set, and use it as the mean of peak pressure amplitudes; obtain the standard deviation of the peak pressure amplitudes of all samples in the sample peak pressure amplitude set, and use it as the standard deviation of peak pressure amplitudes; Add 3 to the mean of the peak pressure amplitude and multiply by the standard deviation of the peak pressure amplitude to obtain the lower limit of the amplitude. The duration of the third statistical time window is taken as the third fixed duration; The baseline time is taken as the end point of the third statistical time window, and a third fixed duration is extracted before it to serve as the current time range of the third statistical time window. Extract historical pressure fluctuation events from the set of historical pressure fluctuation events whose time intervals fall within the third statistical time window, and use them as the third statistical sample. Obtain all third statistical samples to form a third statistical sample set; In the third statistical sample set, for any third statistical sample, the duration of pressure fluctuation corresponding to the third statistical sample is obtained as the duration of sample pressure fluctuation. Obtain the duration of pressure fluctuations for all samples to form a set of sample pressure fluctuation durations. Obtain the mean of the duration of pressure fluctuations for all samples in the sample set of pressure fluctuation durations, and use it as the mean of the duration of pressure fluctuations; obtain the standard deviation of the duration of pressure fluctuations for all samples in the sample set of pressure fluctuation durations, and use it as the standard deviation of the duration of pressure fluctuations. Add 3 to the mean duration of pressure fluctuations and multiply by the standard deviation of the duration of pressure fluctuations to get the upper limit benchmark for the duration.

5. The intelligent pressure relief control system for overpressure in a confined space according to claim 1, characterized in that, The current pressure fluctuation event identification module identifies the current pressure fluctuation event from the current pressure signal sequence, as follows: Extract the pressure value of the confined space corresponding to the current pressure signal sequence to form a pressure value sequence; The pressure values ​​of the confined space corresponding to two adjacent sampling times in the pressure value sequence are subjected to first-order difference to obtain the difference result. The difference result is used as the current pressure change rate of the later sampling time in the two adjacent sampling times. Obtain all current pressure change rates, and arrange all current pressure change rates in chronological order according to the sampling time corresponding to each current pressure change rate to form a current pressure change rate sequence. Obtain the absolute value of the current pressure change rate for each current pressure change rate in the current pressure change rate sequence, forming a sequence of absolute values ​​of the current pressure change rate; In the current pressure change rate absolute value sequence, if there are N1 consecutive sampling times where the current pressure change rate absolute value is less than or equal to the change rate absolute value threshold, the last sampling time corresponding to the N1 consecutive sampling times is marked as the current candidate event start point. In the current pressure change rate sequence, if the current pressure change rate is greater than the change rate threshold for N2 consecutive sampling times after the current candidate event start point, the current candidate event start point is marked as the current event start point; In the current pressure change rate absolute value sequence, after a certain current event start point, when it is detected that there are N3 consecutive sampling times where the current pressure change rate absolute value is less than or equal to the change rate absolute value threshold, the starting sampling time corresponding to the N3 consecutive sampling times is marked as the current event end point corresponding to the current event start point. Based on a certain current event start point and the corresponding current event end point, extract the closed space pressure values ​​from all sampling times between the sampling time corresponding to the current event start point and the sampling time corresponding to the current event end point in the pressure value sequence to form a current event pressure value subsequence. The current event pressure value subsequence is used as the current pressure fluctuation event.

6. The intelligent pressure relief control system for overpressure in a confined space according to claim 5, characterized in that, In the current time-domain feature extraction module, the time-domain feature parameters of the current pressure fluctuation event are extracted as follows: Extract the maximum pressure value of the confined space in the current pressure fluctuation event as the highest pressure value of the current event; obtain the sampling time corresponding to the highest pressure value of the current event as the sampling time of the highest pressure value of the current event; If there are multiple corresponding sampling times for the current event's highest pressure value, then the earliest sampling time shall be taken as the sampling time for the current event's highest pressure value. Extract the pressure value of the enclosed space corresponding to the start point of the current pressure fluctuation event, and use it as the stable pressure value before the start of the current event. Subtracting the stable pressure value before the start of the current event from the highest pressure value of the current event yields the peak pressure amplitude of the current pressure fluctuation event. The current pressure rise phase is formed by taking the sampling time corresponding to the start point of the current event in the current pressure fluctuation event as the starting point and the sampling time of the highest pressure value of the current event as the ending point. During this current pressure rise phase, each sampling moment after the starting point is taken as the current rise sampling moment; Extract the current pressure change rate corresponding to each current rising sampling moment from the current pressure change rate sequence, and use it as the current rising stage pressure change rate; obtain all current rising stage pressure change rates to form a set of current rising stage pressure change rates. From the set of pressure change rates during the current upward phase, obtain the maximum value of the pressure change rate during the current upward phase, and use it as the pressure rise rate of the current pressure fluctuation event. Starting from the sampling time of the highest pressure value of the current event, and ending with the sampling time corresponding to the end point of the current event in the current pressure fluctuation event, the current pressure decline phase is formed. During this current pressure decline phase, each sampling moment after the starting point will be taken as the current decline sampling moment; Extract the absolute value of the current pressure change rate corresponding to each current pullback sampling moment from the absolute value sequence of pressure change rates, and use it as the absolute value of the pressure change rate of the current pullback stage; obtain all the absolute values ​​of the pressure change rate of the current pullback stage to form a set of absolute values ​​of the pressure change rate of the current pullback stage. From the set of absolute values ​​of the pressure change rate during the current pullback phase, obtain the maximum value of the absolute value of the pressure change rate during the current pullback phase, and use it as the pressure decrease rate of the current pressure fluctuation event. The time difference between the end point and the start point of the current pressure fluctuation event is taken as the duration of the pressure fluctuation in the current pressure fluctuation event.

7. The intelligent pressure relief control system for overpressure in a confined space according to claim 6, characterized in that, In the pressure relief decision module, based on the judgment benchmark set, it determines whether the current pressure fluctuation event is a water hammer pseudo-overshoot event, as follows: If the pressure rise rate of the current pressure fluctuation event is greater than the upper limit benchmark of the rise rate, the pressure fall rate of the current pressure fluctuation event is greater than the upper limit benchmark of the fall rate, the pressure peak amplitude of the current pressure fluctuation event is greater than the lower limit benchmark of the amplitude, and the pressure fluctuation duration of the current pressure fluctuation event is less than the upper limit benchmark of the duration, the current pressure fluctuation event is determined to be a water hammer pseudo-overshoot event.

Images