Multi-pressure-regulating-station management method, system and equipment based on airport operation data

Abstract

The invention belongs to the technical field of information technology, and particularly discloses a multi-pressure-regulating-station management method, system and equipment based on airport operation data, and the method comprises the steps: collecting air temperature data and water consumption demand data from an external environment and the interior of an airport building through deploying a sensor network, and obtaining a real-time environment change index and a load demand index; according to the real-time environment change index and the load demand index, a support vector machine algorithm is adopted to classify the high temperature influence, the extreme weather level is judged, and a weather influence coefficient is obtained; after a weather influence coefficient is obtained, the coefficient is transmitted to a pressure regulating station control system through a data interaction interface, and the expected amplification of the water supply load is determined; if the expected amplification exceeds a preset threshold value, a neural network model is adopted to predict a water pipe pressure adjustment demand, and an optimized pressure value and a water reserve distribution scheme are obtained; the invention aims to solve the problem of overlarge water supply load demand under extreme weather conditions such as high temperature and the like in the prior art.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of information technology, and in particular to a multi-regulating station management method, system and device based on airport operation data. BACKGROUND

[0002] In the field of modern airport operation management, the scientific management of regulating stations is directly related to the stable operation of the terminal building water supply system and the energy utilization efficiency, and its importance is self-evident. As a complex comprehensive facility, an airport involves the coordinated operation of multiple systems, especially the high-temperature water supply system. The regulating station not only needs to ensure the safety of water supply, but also needs to meet the dynamic demand of different regions. However, the current management method often cannot adapt to the requirements of complex environmental changes and multi-party coordination, especially in extreme weather conditions, the response capability and resource allocation efficiency of the system are inadequate, which poses a potential threat to the normal operation of the airport.

[0003] The existing solutions often ignore the dynamic correlation between environmental factors and actual demand when facing the management of airport regulating stations, resulting in a lack of foresight and overallness in resource allocation and emergency response. Especially in summer high temperature and other special periods, the system cannot adjust the operation strategy in time according to the changes of external conditions, and it is also difficult to realize the real-time linkage with the water demand of the terminal building. This limitation makes the regulating station often fall into the dilemma of passive adjustment when dealing with unexpected situations, affecting the stability of water demand and the rationality of energy use.

[0004] Focusing on the technical difficulties, a core problem in the management of airport regulating stations is how to realize the seamless connection between external environment monitoring and internal water demand. When high temperature comes, the increase of air temperature will directly lead to the surge of water supply load, and if the water pipe pressure and water reserve cannot be adjusted in advance, the water supply system of the boiler may not have enough water, resulting in insufficient water demand or waste of resources. The deeper problem is that there is a lack of effective data interaction mechanism between the regulating station and the terminal building water supply system, which makes them unable to form a synergistic effect. For example, in the critical time window before the arrival of high temperature, if the regulating station cannot timely know the actual load demand of the terminal building, it may miss the best opportunity for pressure adjustment, thereby affecting the comfort and safety of operation in the passenger area.

[0005] Therefore, how to break through the information barrier between the regulating station and the terminal building water supply system in extreme weather conditions such as high temperature, and realize the dynamic matching of water pipe pressure adjustment and water supply load demand, has become a key problem to be solved in the management of airport operation. SUMMARY

[0006] The application provides a multi-pressure regulating station management method, system and equipment based on airport operation data, aiming to solve the problem of excessive water supply load demand in extreme weather conditions such as high temperature in the prior art.

[0007] To solve the above technical problems, the technical solution adopted by the application is: The multi-pressure regulating station management method based on airport operation data comprises the following steps: collecting air temperature data and water consumption demand data from the external environment and the terminal building interior through a sensor network to obtain real-time environmental change indicators and load demand indicators; classifying high-temperature influence by using a support vector machine algorithm according to the real-time environmental change indicators and load demand indicators, judging the extreme weather level and obtaining a weather influence coefficient; after obtaining the weather influence coefficient, transmitting the coefficient to a pressure regulating station control system through a data interaction interface to determine the expected increase of the water supply load; if the expected increase exceeds a preset threshold, predicting the water pipe pressure adjustment demand by using a neural network model to obtain an optimized pressure value and a water reserve allocation scheme; extracting key parameters from the optimized pressure value and water reserve allocation scheme, sending instructions to the pressure regulating station equipment through a control module to obtain a dynamically adjusted operating state; according to the dynamically adjusted operating state, obtaining feedback data and comparing the feedback data with the initial load demand indicators to judge the matching degree and determine further fine-tuning demand; if the further fine-tuning demand exists, updating the water pipe pressure and the water reserve through an iterative optimization cycle to obtain a final water supply load matching result.

[0008] In one aspect of the application, the collecting of air temperature data and water consumption demand data from the external environment and the terminal building interior through a sensor network to obtain real-time environmental change indicators and load demand indicators comprises: obtaining air temperature data and water supply load data from the external environment and the terminal building interior through a sensor network to construct real-time environmental change indicators and load demand indicators; the obtained air temperature data and water supply load data are cleaned and standardized by a preprocessing module to obtain a structured environmental change data set and a load demand data set; for the structured environmental change data set and the load demand data set, a time series analysis method is used to determine the periodic pattern of environmental change and the fluctuation trend of load demand; if the periodic pattern of environmental change exceeds a preset threshold range, the abnormal points are removed by a data filtering module to obtain a corrected environmental change data set; according to the corrected environmental change data set, in combination with the load demand data set, a regression analysis model is used to judge the correlation strength between the water supply load and the air temperature change; based on the correlation strength result, the monitoring range and the data collection frequency of the sensor network are dynamically adjusted to obtain an optimized real-time collection strategy; After obtaining the optimized real-time collection strategy, update the sensor configuration of the deployment location, determine more accurate environmental change indicators and load demand indicators.

[0009] In an aspect of the present application, according to the real-time environmental change indicators and load demand indicators, the support vector machine algorithm is used to classify the high temperature influence, judge the extreme weather level and obtain the weather influence coefficient, which includes: From the environmental change and load demand data set, obtain the fluctuation of real-time indicators, use the support vector machine algorithm to classify and process the high temperature influence, and obtain the preliminary extreme weather classification result; According to the preliminary extreme weather classification result, obtain the judgment basis of weather level, if the classification result exceeds the preset threshold range, filter the abnormal data points, and determine the corrected weather level classification; According to the corrected weather level classification, combined with the calculation logic of the influence coefficient, if the influence coefficient deviates greatly from the historical data, adjust the coefficient through data smoothing technology to obtain stable weather influence coefficient; Through the stable weather influence coefficient, combined with the fluctuation of real-time indicators, use data comparison method to judge the correlation strength between high temperature influence and load demand; According to the correlation strength between high temperature influence and load demand, according to the coverage of the monitoring range, if the correlation strength is higher than the preset standard, adjust the density of sensor data collection to determine the optimized monitoring range configuration; Through the optimized monitoring range configuration, combined with the dynamic adjustment strategy logic, obtain the real-time corresponding relationship between environmental change and extreme weather, and judge the priority direction of subsequent data collection; According to the priority direction of subsequent data collection, according to the update demand of index correlation, use data integration technology to obtain more comprehensive comprehensive analysis result of high temperature influence and load demand.

[0010] In an aspect of the present application, after obtaining the weather influence coefficient, the coefficient is transmitted to the pressure regulating station control system through the data interaction interface to determine the expected increase of water supply load, which includes: Receive the weather influence coefficient through the data interaction interface, use linear regression algorithm to train the corresponding model between the coefficient and the historical water supply load record, and obtain the preliminary prediction value of the expected increase of water supply load; According to the preliminary prediction value of the expected increase of water supply load, according to the real-time running state data of the pressure regulating station control system, obtain the deviation amount of the current pressure regulation parameter and the prediction value, determine the initial correction direction of pressure regulation; According to the initial correction direction of pressure regulation, use threshold comparison method, if the deviation amount exceeds the preset range, trigger the automatic correction of pressure regulation parameter, and obtain the corrected pressure set value; According to the modified pressure set value, combined with the pipeline flow monitoring data of the pressure regulating station control system, the flow distribution adjustment demand of the water supply pipe network is judged through flow balance calculation, and the flow distribution scheme is determined; According to the flow distribution scheme, the matching degree index is obtained according to the matching condition of the expected increase of water supply load and the actual flow data, the heat supply stability of the water supply pipe network is judged, and the stability evaluation result is obtained; According to the stability evaluation result, the random forest algorithm is used for classification processing on the evaluation result and the historical water supply load record, and the final confirmation value of the water supply load increase is obtained; According to the final confirmation value of the water supply load increase, the data interaction interface is fed back to the pressure regulating station control system, and the pressure and flow combined regulation instruction of the next period is determined.

[0011] In one aspect of the present application, if the expected increase exceeds the preset threshold, a neural network model is used to predict the water pipe pressure adjustment demand, and an optimized pressure value and water quantity reserve distribution scheme are obtained, including: The expected increase data of the water supply system is obtained through the data acquisition module, and preliminary comparison is carried out combined with the preset threshold to judge whether the subsequent adjustment process is triggered; If the expected increase exceeds the preset threshold, a neural network model is used to deeply analyze the water pipe pressure data, and an optimized pressure value domain is obtained; According to the optimized pressure value domain, combined with the current capacity data of the water quantity reserve, the availability of the reserve resources is calculated, and a preliminary distribution scheme is determined; For the preliminary distribution scheme, the real-time running state data of the water supply system is obtained, and the feasibility of the distribution scheme is judged through comparison of the scheme matching degree; If the scheme matching degree does not reach the preset standard, the distribution scheme is recalculated through the data adjustment module, and a modified distribution scheme is obtained; Through the data transmission interface, the modified distribution scheme and the optimized pressure value domain are synchronized to the water consumption control system to determine the final execution instruction; According to the final execution instruction, the running feedback data of the water supply system is obtained, the stability of the instruction execution is judged, and the process closed loop is completed.

[0012] In one aspect of the present application, the key parameters are extracted from the optimized pressure value and water quantity reserve distribution scheme, the control module sends instructions to the pressure regulating station equipment to obtain the dynamically adjusted running state, including: Obtain the expected load increase data of the next period of the water supply system; Compare the expected load increase data with the preset pressure adjustment threshold; If the expected load increase exceeds the pressure adjustment threshold, a neural network model is used to process the water pipe pressure history sequence to obtain a target pressure interval; According to the target pressure interval and the total amount of water reserve, combined with the current inventory of each pressure regulating station, the distribution ratio of each station is calculated; The distribution ratio is used to generate pressure setting instructions for each pressure regulating station and is issued; Get the actual running state data of each pressure regulating station after executing the pressure setting instruction; Compare the actual running state data with the target pressure interval to obtain the corrected pressure value and the corrected distribution ratio.

[0013] In one aspect of the present application, the feedback data is obtained according to the dynamically adjusted running state, and compared with the initial load demand index to judge the matching degree and determine the further fine tuning demand, comprising: Get the feedback data of the dynamically adjusted running state; Compare the feedback data with the initial load index to obtain the matching deviation value; If the matching deviation value exceeds the preset threshold, the fine tuning demand is determined; By collecting real-time pipe network flow sequence, a time series decomposition method is used to process the flow sequence to obtain the load fluctuation amplitude; According to the load fluctuation amplitude and the current water reserve distribution, the supplement distribution amount of each pressure regulating station is calculated; The supplement distribution amount is solved by using a linear programming model to determine the distribution instruction of each pressure regulating station; Generate the distribution instruction and issue it to each pressure regulating station equipment.

[0014] In one aspect of the present application, if the further fine tuning demand exists, the water pipe pressure and water reserve are updated through an iterative optimization cycle to obtain the final water supply load matching result, comprising: Get the current water pipe pressure and water reserve distribution data; Determine the water pipe pressure fluctuation range and reserve distribution uneven area by collecting multi-point pressure sensor readings and reserve amount monitoring data; According to the water pipe pressure fluctuation range, a linear programming model is used to calculate the pressure adjustment amount; If the pressure adjustment amount exceeds the preset threshold, a pressure control instruction is generated and issued to the pressure regulating valve equipment; Get the real-time water pipe pressure feedback data after executing the pressure control instruction; According to the real-time water pipe pressure feedback data and the water reserve distribution, the reserve supplement distribution amount is calculated; If the reserve supplement distribution amount deviates greatly from the current reserve distribution, update the water reserve distribution and generate the supplement distribution instruction.

[0015] In another aspect of the present application, the present application also relates to a multi-pressure regulating station management system based on airport operation data, characterized by: a data acquisition module for acquiring air temperature data and water demand data from the external environment and the terminal interior through the deployment of a sensor network to obtain real-time environmental change indicators and load demand indicators; a high-temperature classification module for classifying high-temperature influences, judging extreme weather levels and obtaining weather influence coefficients by using a support vector machine algorithm according to the real-time environmental change indicators and the load demand indicators; a data transmission module for transmitting the coefficients to a pressure regulating station control system through a data interaction interface after obtaining the weather influence coefficients to determine the expected increase of the water supply load; an increase judgment module for predicting water pipe pressure adjustment requirements by using a neural network model to obtain optimized pressure values and water reserve allocation schemes if the expected increase exceeds a preset threshold; a pressure prediction module for extracting key parameters from the optimized pressure values and the water reserve allocation schemes, sending instructions to the pressure regulating station equipment through a control module and obtaining a dynamically adjusted operating state; an instruction control module for obtaining feedback data and comparing them with the initial load demand indicators according to the dynamically adjusted operating state, judging the matching degree and determining further fine-tuning requirements; a feedback fine-tuning module for judging whether the further fine-tuning requirements exist, and if so, updating the water pipe pressure and the water reserve through an iterative optimization cycle to obtain the final water supply load matching result.

[0016] In another aspect of the present application, the present application also relates to an electronic device, comprising: a memory having a computer program stored thereon; a processor for executing the computer program in the memory to implement the above-mentioned multi-pressure regulating station management method based on airport operation data.

[0017] Compared with the prior art, the present application has the following beneficial effects: The present application classifies high-temperature influences, judges extreme weather levels and calculates weather influence coefficients by using a support vector machine algorithm; transmits the coefficients to a pressure regulating station control system to determine the expected increase of the water supply load; when the increase exceeds a threshold, introduces a neural network model to predict water pipe pressure adjustment requirements, generates optimized pressure values and water reserve allocation schemes; extracts key parameters to issue instructions to pressure regulating station equipment to realize dynamic adjustment of pressure and reserve; compares feedback data with initial requirements to judge the matching degree and perform iterative optimization to finally obtain a precise water supply load matching result. The present application realizes intelligent prediction and dynamic regulation of water supply load under high-temperature extreme weather, effectively guarantees the stability of terminal water demand and energy utilization efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required to be used in the embodiments will be briefly introduced as follows. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be considered as a limitation to the scope. Other related drawings can also be obtained by those skilled in the art without any creative effort on the basis of these drawings.

[0019] Fig. 1 One of the flowcharts of the multi-pressure regulating station management method based on airport operation data of the present application.

[0020] Fig. 2 The second flowchart of the multi-pressure regulating station management method based on airport operation data of the present application.

[0021] Fig. 3 The third flowchart of the multi-pressure regulating station management method based on airport operation data of the present application. DETAILED DESCRIPTION

[0022] The present application will be further described in conjunction with the embodiments. The described embodiments are only some of the embodiments of the present application, and are not all the embodiments. Based on the embodiments of the present application, other embodiments obtained by those skilled in the art without any creative effort belong to the protection scope of the present application.

[0023] Please refer to Figs. 1-3 The present embodiment discloses a multi-pressure regulating station management method and system based on airport operation data, which can specifically include: Step 101: Collecting air temperature data and water consumption demand data from the external environment and the inside of the terminal by deploying a sensor network to obtain real-time environmental change indicators and load demand indicators.

[0024] The air temperature data and water supply load data are obtained from the external environment and the terminal building through the sensor network to construct real-time environmental change indicators and load demand indicators. The obtained air temperature data and water supply load data are cleaned and standardized by the preprocessing module to obtain structured environmental change data sets and load demand data sets. For the structured environmental change data sets and load demand data sets, the time series analysis method is used to determine the periodic pattern of environmental change and the fluctuation trend of load demand. If the periodic pattern of environmental change exceeds the preset threshold range, the data filtering module is used to remove abnormal points to obtain the corrected environmental change data set. According to the corrected environmental change data set, combined with the load demand data set, the regression analysis model is used to judge the correlation strength between the water supply load and the air temperature change. Through the correlation strength result, the monitoring range and data acquisition frequency of the sensor network are dynamically adjusted to obtain the optimized real-time acquisition strategy. After obtaining the optimized real-time acquisition strategy, the sensor configuration of the deployment location is updated to determine more accurate environmental change indicators and load demand indicators.

[0025] Specifically, in the process of deploying the sensor network to collect air temperature data and water demand data, first, high-precision temperature and humidity sensors, such as HTS-300 devices, are installed outside and inside the terminal building. The environmental temperature data is automatically collected every 5 minutes. The external sensors cover 10 key points within a 100-meter range around the terminal building, and the internal sensors are arranged in the departure hall, office area, and other 5 main areas to ensure data comprehensiveness. The collected temperature data is transmitted to the central server through wireless network. The server processes the data using time series analysis algorithm to calculate the average temperature change rate per hour, for example, the external temperature decreases from 26 degrees at 6 am to 40 degrees at 12 pm, with a change rate of 4 degrees / hour, which serves as the basis for real-time environmental change indicators. Then, for water demand data, the system records the energy consumption data of each area in the terminal building through intelligent electric meters, combines historical water consumption records and current temperature data, and uses linear regression algorithm to predict load demand, for example, the prediction model shows that when the temperature is higher than 26 degrees, the water demand in the departure hall increases by about 15.6 kWh per hour. Combined with the current temperature of 40 degrees, the system automatically calculates the current load demand indicator as 18.2 kWh, and compares it with the historical peak value of 20.5 kWh to analyze that the current load is at a moderately high level. Finally, the system integrates the environmental change indicators and load demand indicators into the data visualization platform, and displays the temperature distribution and water demand intensity through dynamic heat maps, for example, red areas represent areas with temperature higher than 20 degrees and load demand exceeding 18 kWh. The system automatically triggers water demand equipment optimization scheduling instructions to ensure that energy utilization efficiency is improved by about 12.3%. At the same time, the system updates the scheduling strategy every 4 hours according to the prediction model and real-time data to form a closed-loop management logic, ensuring continuous optimization of environmental adaptability and water efficiency.

[0026] In step 102, according to the real-time environmental change index and the load demand index, the support vector machine algorithm is used to classify the high temperature influence, determine the extreme weather level and obtain the weather influence coefficient.

[0027] From the data set of environmental change and load demand, the fluctuation of real-time index is obtained, the support vector machine algorithm is used to classify and process the high temperature influence, and the preliminary extreme weather classification result is obtained. According to the preliminary extreme weather classification result, the judgment basis of weather level is obtained, if the classification result exceeds the preset threshold range, the abnormal data points are filtered, and the corrected weather level classification is determined. According to the corrected weather level classification, combined with the calculation logic of the influence coefficient, if the influence coefficient deviates greatly from the historical data, the coefficient is adjusted through the data smoothing technology, and the stable weather influence coefficient is obtained. Through the stable weather influence coefficient, combined with the fluctuation of real-time index, the data comparison method is used to judge the correlation strength between high temperature influence and load demand. According to the correlation strength between high temperature influence and load demand, according to the coverage of the monitoring range, if the correlation strength is higher than the preset standard, the density adjustment of sensor data acquisition is used to determine the optimized monitoring range configuration. Through the optimized monitoring range configuration, combined with the dynamic adjustment strategy logic, the real-time corresponding relationship between environmental change and extreme weather is obtained, and the priority direction of subsequent data acquisition is determined. According to the priority direction of subsequent data acquisition, according to the updating demand of index correlation, the data integration technology is used to obtain more comprehensive comprehensive analysis result of high temperature influence and load demand.

[0028] Specifically, after receiving real-time environmental change indicators such as hourly temperature drop rate 4 degrees / hour and load demand indicators such as current total water demand 22.4 kWh, the system first combines these indicators with historical high temperature data sets to form a feature data set containing similar weather events in the past 30 days, including temperature change rate, wind speed influence factor, and humidity fluctuation value, etc. Then, a multi-classification model is constructed using a support vector machine algorithm, and a radial basis kernel function is used for nonlinear mapping of the data. During the training process, the parameters C=10.5 and γ=0.12 are optimized by grid search, so that the accuracy of the model reaches 92.7% in cross-validation. In actual classification, the system inputs the current feature vector, such as temperature change rate 4 degrees / hour combined with wind speed 8.3 meters / second and humidity increase 12%, and the model outputs a high temperature impact of three extreme weather levels, with a weather impact coefficient of 1.45, indicating that the water demand needs to be enlarged by 45%. Subsequently, the system multiplies this coefficient with the load demand indicator to calculate the adjusted predicted load as 22.4 x 1.45 = 32.48 kWh, and compares it with the threshold library to confirm that it belongs to the high-risk interval, automatically activates the backup water demand plan, and records the classification result and coefficient to the database for incremental learning and updating of the next model, ensuring that the adaptability of the algorithm to new high temperature patterns gradually increases, forming a complete closed-loop analysis logic from indicator input to impact assessment to coefficient application.

[0029] After obtaining the weather impact coefficient, the coefficient is transmitted to the pressure regulating station control system through the data interaction interface to determine the expected increase of the water supply load.

[0030] After obtaining the weather impact coefficient through the data interaction interface, a linear regression algorithm is used to train the corresponding model between the coefficient and the historical water supply load record to obtain a preliminary prediction value of the expected increase of the water supply load. According to the preliminary prediction value of the expected increase of the water supply load, the deviation between the current pressure regulating parameter and the prediction value is obtained according to the real-time running state data of the pressure regulating station control system, and the initial correction direction of the pressure regulation is determined. According to the initial correction direction of the pressure regulation, a threshold comparison method is used, if the deviation exceeds the preset range, the automatic correction of the pressure regulating parameter is triggered, and the corrected pressure set value is obtained. According to the corrected pressure set value, combined with the pipe flow monitoring data of the pressure regulating station control system, the flow distribution adjustment demand of the water supply pipe network is judged through flow balance calculation, and the flow distribution scheme is determined. According to the flow distribution scheme, the matching degree index is obtained according to the matching situation of the expected increase of the water supply load and the actual flow data, the heat supply stability of the water supply pipe network is judged, and the stability evaluation result is obtained. According to the stability evaluation result, a random forest algorithm is used to classify the evaluation result and the historical water supply load record to obtain the final confirmation value of the increase of the water supply load. According to the final confirmation value of the increase of the water supply load, feedback is given to the pressure regulating station control system through the data interaction interface to determine the next period of pressure and flow joint regulation instruction.

[0031] Specifically, after the system obtains the weather impact coefficient 1.45, the coefficient is pushed to the central control system of the pressure regulating station in real time through the data interaction interface of the MQTT protocol. The control system first parses the received JSON format data packet, extracts the coefficient field and verifies the signature integrity. Then, the control system calls the built-in load increment calculation module, multiplies the coefficient by the current pipe network pressure reference value 15.8 Pa to obtain the expected pressure increment value 15.8x1.45=22.91 Pa, and combines the real-time flow monitoring data 42.6 cubic meters / hour to calculate the adjusted target gas supply amount 42.6x1.45=61.77 cubic meters / hour using a linear regression prediction algorithm. Next, the system compares the target value with the pressure regulating station valve opening mapping table, and dynamically adjusts the proportional valve opening to 68.4% through a PID control algorithm, with an integral time constant of 12 seconds and a differential time constant of 3 seconds, ensuring that the pressure fluctuation is controlled within ±0.5 Pa. At the same time, the system writes the calculated expected increment value and valve adjustment record into the distributed time series database for subsequent pipe network balance optimization module calling, forming a closed-loop control logic from coefficient transmission to pressure load dynamic adjustment.

[0032] Step 104, if the expected increment exceeds the preset threshold, a neural network model is used to predict the water pipe pressure adjustment requirement to obtain an optimized pressure value and water quantity reserve distribution scheme.

[0033] Through the data acquisition module, the expected increment data of the water supply system is obtained, and a preliminary comparison is made with the preset threshold to determine whether to trigger the subsequent adjustment process. If the expected increment exceeds the preset threshold, a neural network model is used to analyze the water pipe pressure data in depth to obtain an optimized pressure value domain. According to the optimized pressure value domain, combined with the current capacity data of the water quantity reserve, the availability of the reserve resources is calculated to determine a preliminary distribution scheme. For the preliminary distribution scheme, real-time operation state data of the water supply system is obtained, and the feasibility of the distribution scheme is determined by comparing the matching degree of the scheme. If the scheme matching degree does not meet the preset standard, the distribution scheme is recalculated through the data adjustment module to obtain a corrected distribution scheme. Through the data transmission interface, the corrected distribution scheme and the optimized pressure value domain are synchronized to the water consumption control system to determine the final execution instruction. According to the final execution instruction, the operation feedback data of the water supply system is obtained to determine the stability of the instruction execution, and the process is closed.

[0034] Specifically, when the system detects that the expected load increase exceeds the preset threshold of 25%, the neural network prediction module is triggered immediately, which is based on a three-layer feedforward neural network structure, and the input layer contains the weather influence coefficient 1.62, the current pipe network pressure 18.4 Pa, the real-time flow 56.3 cubic meters / hour, and the temperature sequence data in the past 24 hours. First, the input data is normalized to the range of 0-1, and then the feature extraction is performed through the hidden layer with the ReLU activation function, and the weight matrix has been trained by the historical data set with the loss function MSE of 0.012. The network output layer directly regresses the predicted water pipe pressure adjustment demand, and the calculation shows that the optimized pressure target value is 18.4x1.62+2.7 Pa compensation item=32.51 Pa, and the water reserve allocation scheme is generated, which preferentially allocates 38% of the total reserve to the main pipeline buffer tank, and the remaining 62% is evenly distributed to the three-level branch reserve points, so as to reduce the risk of peak pressure. During the prediction process, the gradient explanation algorithm is run synchronously to calculate the importance weight of each input feature, and it is found that the weather coefficient contribution rate is 54.8% and the flow data contribution rate is 31.2%. The system then encapsulates the optimized pressure value and the reserve allocation scheme as structured data, pushes it to the downstream execution unit through the internal message queue, and records the prediction confidence of 92.7% and the feature contribution distribution to the log database, forming a complete automatic decision link from threshold judgment to reserve optimization.

[0035] Step 105, extract key parameters from the optimized pressure value and water reserve allocation scheme, send instructions to the pressure regulating station equipment through the control module, and obtain the dynamically adjusted running state.

[0036] Obtain the expected load increase data of the next cycle of the water supply system. Compare the expected load increase data with the preset pressure adjustment threshold. If the expected load increase exceeds the pressure adjustment threshold, use the neural network model to process the water pipe pressure history sequence to obtain the target pressure interval. According to the target pressure interval and the total amount of water reserve, combined with the current inventory of each pressure regulating station, calculate the allocation proportion of each station. Use the allocation proportion to generate pressure setting instructions for each pressure regulating station and issue them. Obtain the actual running state data of each pressure regulating station after executing the pressure setting instructions. Compare the actual running state data with the target pressure interval to obtain the corrected pressure value and the corrected allocation proportion.

[0037] Specifically, the control module parses the optimization pressure target value 32.51 Pa and the reserve allocation ratio from the received structured data. First, it extracts key fields through a JSON parser, including the main trunk buffer tank allocation rate of 38%, the third-level branch allocation rate of 62%, and the target pressure increment of 14.11 Pa. Then, it converts these parameters into device control instruction format and encapsulates them into register write command packets using the Modbus TCP protocol, with register address 40001 writing the target pressure value 32.51, address 40002 writing the main trunk tank allocation ratio 0.38, and addresses 40003 to 40005 writing the third-level branch ratios 0.2067 average value, respectively. After adding a CRC16 check code 0xA47F to the command packet, it is sent to the pressure regulating station PLC controller through industrial Ethernet, with an exponential backoff algorithm retrying three times to ensure reliability. After receiving the command, the pressure regulating station device immediately executes proportional valve opening adjustment, calculates the valve opening increment using a PID control algorithm with Kp set to 1.85, Ki to 0.42, and Kd to 0.18, calculates the output control quantity based on the current feedback pressure and the target 32.51 Pa deviation -0.37 Pa, adjusts the valve opening from the current 65% to 78.4%, and simultaneously starts the reserve pump set, according to the allocation ratio, preferentially injects 1732.8 cubic meters of the total reserve 4560 cubic meters into the main trunk buffer tank, and the remaining 2827.2 cubic meters is injected into the third-level branch through a flow distribution algorithm. After adjustment, the PLC collects running state data every 5 seconds, including actual pressure 32.48 Pa, flow rate stabilized at 78.9 cubic meters / hour, and reserve tank liquid level distribution, calculates the pressure deviation rate 0.09% which is lower than the threshold 0.5%, confirms the adjustment success, and encapsulates the state data as a feedback message, returns to the upstream control module through the same message queue, and forms a closed-loop verification link.

[0038] Step 106, according to the dynamically adjusted running state, obtain feedback data and compare with the initial load demand index, judge the matching degree and determine the further fine tuning demand.

[0039] Obtain the feedback data of the dynamically adjusted running state. Compare the feedback data with the initial load index to obtain the matching deviation value. If the matching deviation value exceeds the preset threshold, determine the fine tuning demand. By collecting real-time pipe network flow sequence, use time series decomposition method to process the flow sequence to obtain the load fluctuation amplitude. According to the load fluctuation amplitude and the current water reserve distribution, calculate the supplementary allocation of each pressure regulating station. Use linear programming model to solve the supplementary allocation to determine the allocation instruction of each pressure regulating station. Generate the allocation instruction and send it to each pressure regulating station device.

[0040] Specifically, after the dynamic adjustment of the operation state of the water volume pressure regulating station, the system first obtains feedback data from the field sensors through the data acquisition module, including the current operating pressure value of 31.75 Pa, the total flow of 82.3 cubic meters / hour, and the liquid level proportion of each reserve tank, among which the main reserve tank liquid level is 42.5%, and the average liquid level of the secondary branch line is 57.5%. Then, the system compares these feedback data with the initial load demand indicators, which are set as operating pressure 31.80 Pa, total flow demand 85.0 cubic meters / hour, and main reserve tank liquid level proportion target 45%. Through comparative analysis, it is calculated that the pressure deviation is -0.05 Pa, the deviation rate is only 0.16%, which is lower than the preset threshold of 0.3%, while the flow deviation is -2.7 cubic meters / hour, the deviation rate is 3.18%, which exceeds the allowed range of 2.0%, and the liquid level proportion deviation is -2.5%, which also exceeds the threshold of 1.5%. Therefore, the system starts the matching degree evaluation algorithm, adopts a weighted scoring model, in which the pressure weight is 0.4, the flow weight is 0.35, and the liquid level weight is 0.25, and calculates the comprehensive matching degree as 86.7%, which is lower than the target value of 90%, and it is determined that further fine-tuning is needed. Then, the system automatically generates a fine-tuning demand analysis report, determines that the flow adjustment priority is the highest based on the deviation data, and uses a linear regression algorithm to predict the flow adjustment amount combined with historical operation data, obtains that the pump output power needs to be increased to 108% of the original power, the predicted flow is increased to 84.6 cubic meters / hour, and the main reserve tank liquid level distribution ratio is fine-tuned to 43.8%, and the secondary branch line liquid level target is recalculated to 56.2% through the flow distribution model. Finally, the system packages the fine-tuning parameters as internal scheduling instructions, transmits them to the execution unit through the data bus, synchronously updates the monitoring database for subsequent tracking analysis, and forms a complete closed-loop logic from data acquisition to parameter optimization.

[0041] Step 107, if further fine-tuning demand exists, update the water pipe pressure and water reserve through an iterative optimization loop to obtain the final water supply load matching result.

[0042] Obtain current water pipe pressure and water reserve distribution data. By collecting multi-point pressure sensor readings and reserve amount monitoring data, determine the water pipe pressure fluctuation range and reserve distribution uneven area. According to the water pipe pressure fluctuation range, use a linear programming model to calculate the pressure adjustment amount. If the pressure adjustment amount exceeds the preset threshold, generate a pressure control instruction and issue it to the pressure regulating valve device. Obtain real-time water pipe pressure feedback data after executing the pressure control instruction. According to the real-time water pipe pressure feedback data and the water reserve distribution, calculate the reserve supplement distribution amount. If the reserve supplement distribution amount deviates greatly from the current reserve distribution, update the water reserve distribution and generate a supplement distribution instruction.

[0043] Specifically, after determining the need for further fine-tuning, the system immediately enters an iterative optimization loop, first extracting the water pipe pressure, water reserve volume, and ambient temperature data series within the last 48 hours from the historical database to construct a multivariate time series dataset. Subsequently, an ARIMA model is used to make a short-term prediction of the water pipe pressure, combined with the current measured pipe outlet pressure of 28.4 Pa and the target pressure of 28.6 Pa, a predicted deviation trend coefficient of 0.012 is calculated, and the system accordingly starts the pressure iterative adjustment module, calculates the incremental output in real time through the PID control algorithm, determines that the valve opening degree needs to be increased from the current 72% to 74.8%, and predicts that the pressure in the next cycle can be raised to 28.55 Pa. At the same time, the system processes the water reserve optimization in parallel, collects the current total reserve volume of 18650 cubic meters from the reserve tank group, of which the peak standby tank accounts for 28.4% and the regular tank accounts for 71.6%, compares with the required reserve volume of 19200 cubic meters output by the water supply load prediction model, and calculates a reserve gap of 550 cubic meters. The system uses a dynamic programming algorithm to iteratively solve the reserve allocation, with the minimum transportation energy consumption as the objective function, and the maximum pipe transportation rate of 95 cubic meters / hour and the tank switching time of not more than 6 minutes as the constraint conditions, and finally obtains the optimization scheme: preferentially increasing the input flow to 92 cubic meters / hour from the external supplement line for 45 minutes, and simultaneously transferring 8% of the reserve volume from the peak standby tank to the regular tank. After three iterations, the water pipe pressure stabilizes at 28.59 Pa, the deviation rate decreases to 0.035%, the total reserve volume reaches 19180 cubic meters, the gap narrows to within 20 cubic meters, and the comprehensive water supply load matching degree improves to 98.6%, exceeding the preset threshold of 98%, and the system confirms that the final matching result is reached, immediately locks the current operating parameters and pushes them to the central dispatching platform, and records the entire parameter change curve and convergence index of this iteration process to the optimization log database, providing a reference basis for subsequent load fluctuations and forming a complete iterative closed-loop optimization logic.

[0044] In some different embodiments, the present application provides a multi-pressure regulating station management system based on airport operation data, mainly comprising: A data acquisition module for collecting air temperature data and water demand data from the external environment and the interior of the terminal building through the deployment of a sensor network to obtain real-time environmental change indicators and load demand indicators; A high-temperature classification module for classifying high-temperature influences using a support vector machine algorithm based on real-time environmental change indicators and load demand indicators, determining extreme weather levels, and obtaining weather influence coefficients; A data transmission module for transmitting the weather influence coefficients to the pressure regulating station control system through a data interaction interface after obtaining the coefficients to determine the expected increase in water supply load; The amplification judgment module is configured to, if the expected amplification exceeds a preset threshold, predict water pipe pressure adjustment demand by using a neural network model, and obtain an optimized pressure value and a water reserve allocation scheme; The pressure prediction module is configured to extract key parameters from the optimized pressure value and the water reserve allocation scheme, send instructions to the pressure regulating station equipment through the control module, and obtain a dynamically adjusted operating state. The instruction control module is configured to, according to the dynamically adjusted operating state, acquire feedback data and compare the feedback data with initial load demand indexes, judge a matching degree, and determine further fine-tuning demand. The feedback fine-tuning module is configured to, if the further fine-tuning demand exists, update the water pipe pressure and the water reserve through an iterative optimization cycle, and obtain a final water supply load matching result.

[0045] Although the embodiments of the present application have been shown and described, it can be understood by those skilled in the art that various changes, modifications, replacements and variations can be made to the embodiments without departing from the principles and spirits of the present application, and the scope of the present application is defined by the appended claims and their equivalents.

Claims

1. A multi-pressure station management method based on airport operation data, characterized in that, The method includes: By deploying a sensor network to collect temperature data and water demand data from the external environment and inside the terminal, real-time environmental change indicators and load demand indicators can be obtained. Based on real-time environmental change indicators and load demand indicators, the support vector machine algorithm is used to classify the impact of high temperature, determine the extreme weather level, and obtain the weather impact coefficient. After obtaining the weather impact coefficient, the coefficient is transmitted to the pressure regulating station control system through the data interaction interface to determine the expected increase in water supply load; If the expected increase exceeds the preset threshold, a neural network model is used to predict the water pipe pressure adjustment demand, and an optimized pressure value and water reserve allocation scheme are obtained. Key parameters are extracted from the optimized pressure value and water reserve allocation scheme, and instructions are sent to the pressure regulating station equipment through the control module to obtain the dynamically adjusted operating status. Based on the dynamically adjusted operating status, obtain feedback data and compare it with the initial load demand indicators to determine the degree of matching and identify further fine-tuning requirements. If further fine-tuning is needed, the water pipe pressure and water volume reserve will be updated iteratively to obtain the final water supply load matching result.

2. The multi-pressure station management method based on airport operation data according to claim 1, characterized in that: The method involves deploying a sensor network to collect temperature and water demand data from the external environment and the terminal building, thereby obtaining real-time environmental change indicators and load demand indicators, including: By acquiring temperature and water supply load data from the external environment and the terminal building through a sensor network, real-time environmental change indicators and load demand indicators are constructed. The acquired temperature data and water supply load data are cleaned and standardized by the preprocessing module to obtain structured environmental change datasets and load demand datasets. For structured environmental change datasets and load demand datasets, time series analysis methods are used to determine the periodic patterns of environmental changes and the fluctuation trends of load demand. If the periodic pattern of environmental changes exceeds the preset threshold range, the outlier will be removed through the data filtering module to obtain the corrected environmental change dataset. Based on the corrected environmental change dataset and the load demand dataset, a regression analysis model is used to determine the strength of the correlation between water supply load and temperature changes. By dynamically adjusting the monitoring range and data acquisition frequency of the sensor network based on the correlation strength results, an optimized real-time acquisition strategy can be obtained. After obtaining the optimized real-time data acquisition strategy, the sensor configuration at the deployment location is updated to determine more accurate environmental change indicators and load demand indicators.

3. The multi-pressure station management method based on airport operation data according to claim 1, characterized in that: The process involves classifying the impact of high temperatures using a support vector machine algorithm based on real-time environmental change indicators and load demand indicators, determining the extreme weather level, and obtaining the weather impact coefficient. This includes: By acquiring real-time index fluctuations from the data set of environmental changes and load demand, and using the support vector machine algorithm to classify the impact of high temperatures, preliminary extreme weather classification results are obtained. Based on the preliminary extreme weather classification results, the criteria for determining the weather level are obtained. If the classification results exceed the preset threshold range, the abnormal data points are filtered to determine the corrected weather level classification. Based on the revised weather level classification and the calculation logic of the impact coefficient, if the impact coefficient deviates significantly from historical data, the coefficient is adjusted using data smoothing techniques to obtain a stable weather impact coefficient. By using a stable weather impact coefficient and combining it with real-time indicator fluctuations, a data comparison method is employed to determine the strength of the correlation between the impact of high temperatures and load demand. Based on the strength of the correlation between the impact of high temperature and load demand, and considering the coverage of the monitoring range, if the strength of the correlation is higher than the preset standard, the optimal monitoring range configuration is determined by adjusting the density of sensor data acquisition. By optimizing the monitoring range configuration and combining it with dynamically adjusted strategy logic, the real-time correlation between environmental changes and extreme weather can be obtained to determine the priority direction of subsequent data collection. Based on the priority of subsequent data collection and the need for updating the correlation of indicators, data integration technology is used to obtain a more comprehensive analysis of the impact of high temperature on load demand.

4. The multi-pressure station management method based on airport operation data according to claim 1, characterized in that: After obtaining the weather impact coefficient, the coefficient is transmitted to the pressure regulating station control system through a data interaction interface to determine the expected increase in water supply load, including: The weather impact coefficient is received through the data interaction interface, and a linear regression algorithm is used to train the correspondence model between the coefficient and historical water supply load records to obtain a preliminary prediction of the expected increase in water supply load. Based on the preliminary forecast of the expected increase in water supply load, and using the real-time operating status data of the pressure regulating station control system, the deviation between the current pressure regulation parameters and the forecast values ​​is obtained, and the initial correction direction of pressure regulation is determined. Based on the initial correction direction of the pressure regulation, a threshold comparison method is used. If the deviation exceeds the preset range, the automatic correction of the pressure regulation parameters is triggered to obtain the corrected pressure setpoint. Based on the revised pressure setpoint and combined with the pipeline flow monitoring data of the pressure regulating station control system, the flow balance calculation is used to determine the flow distribution adjustment needs of the water supply network and to determine the flow distribution scheme. Based on the flow allocation plan, the matching degree index is obtained by considering the matching between the expected increase in water supply load and the actual flow data, and the stability of heat supply in the water supply network is judged to obtain the stability assessment results. Based on the stability assessment results, the random forest algorithm is used to classify the assessment results and historical water supply load records to obtain the final confirmed value of the water supply load increase. Based on the final confirmed value of the increase in water supply load, the data is fed back to the pressure regulating station control system through the data interaction interface to determine the pressure and flow joint regulation command for the next cycle.

5. The multi-pressure station management method based on airport operation data according to claim 1, characterized in that: If the expected increase exceeds a preset threshold, a neural network model is used to predict the water pipe pressure adjustment demand, resulting in an optimized pressure value and water reserve allocation scheme, including: The expected increase data of the water supply system is obtained through the data acquisition module, and a preliminary comparison is made with the preset threshold to determine whether to trigger the subsequent adjustment process. If the expected increase exceeds the preset threshold, a neural network model is used to perform in-depth analysis of the water pipe pressure data to obtain the optimized pressure range. Based on the optimized pressure range and the current capacity data of water reserves, the availability of reserve resources is calculated, and a preliminary allocation plan is determined. For the preliminary allocation plan, real-time operating status data of the water supply system is obtained, and the feasibility of the allocation plan is judged by comparing the matching degree of the plan. If the matching degree of the scheme does not meet the preset standard, the allocation scheme will be recalculated through the data adjustment module to obtain the corrected allocation scheme. The revised allocation scheme and optimized pressure range are synchronized to the water consumption control system through the data transmission interface to determine the final execution command. Based on the final execution command, obtain the operational feedback data of the water supply system, determine the stability of command execution, and complete the process loop.

6. The multi-pressure station management method based on airport operation data according to claim 1, characterized in that: The process of extracting key parameters from the optimized pressure value and water reserve allocation scheme, and sending commands to the pressure regulating station equipment through the control module to obtain the dynamically adjusted operating status includes: Obtain data on the expected load increase of the water supply system in the next cycle; Compare the expected load increase data with the preset pressure adjustment threshold; If the expected increase in load exceeds the pressure adjustment threshold, a neural network model is used to process the historical water pipe pressure sequence to obtain the target pressure range. Based on the target pressure range and total water reserves, and combined with the current reserves of each pressure regulating station, the allocation ratio of each station is calculated. The pressure setting instructions for each pressure regulating station are generated and issued using the allocation ratio. Obtain the actual operating status data of each pressure regulating station after executing the pressure setting command; By comparing the actual operating status data with the target pressure range, the corrected pressure value and the corrected distribution ratio are obtained.

7. The multi-pressure station management method based on airport operation data according to claim 1, characterized in that: The process of acquiring feedback data based on the dynamically adjusted operating status and comparing it with the initial load demand indicators to determine the degree of matching and identify further fine-tuning requirements includes: Obtain dynamically adjusted operational status feedback data; The feedback data is compared and analyzed with the initial load indicators to obtain the matching deviation value; If the matching deviation value exceeds the preset threshold, then fine-tuning is required; By collecting real-time pipeline flow sequences and processing the flow sequences using time series decomposition methods, the load fluctuation amplitude can be obtained. Calculate the supplementary allocation for each pressure regulating station based on the load fluctuation range and the current water reserve distribution; A linear programming model is used to solve for the supplementary allocation amount and determine the allocation instructions for each voltage regulating station; Generate allocation instructions and send them to each voltage regulating station device.

8. The multi-pressure station management method based on airport operation data according to claim 1, characterized in that: If further fine-tuning is required, the final water supply load matching result is obtained by iteratively optimizing and updating the water pipe pressure and water volume reserve, including: Obtain current water pipe pressure and water reserve distribution data; By collecting readings from multiple pressure sensors and monitoring data on water reserves, the range of water pipe pressure fluctuations and areas of uneven water reserve distribution can be determined. Based on the pressure fluctuation range of the water pipe, a linear programming model is used to calculate the pressure adjustment amount; If the pressure adjustment exceeds the preset threshold, a pressure control command is generated and sent to the pressure regulating valve device; Obtain real-time water pipe pressure feedback data after the pressure control command is executed; Calculate the reserve replenishment allocation based on real-time water pipe pressure feedback data and water reserve distribution; If the amount of water reserve replenishment allocation deviates significantly from the current reserve distribution, the water reserve distribution will be updated and a replenishment allocation instruction will be generated.

9. A multi-pressure station management system based on airport operation data, characterized in that: The data acquisition module is used to collect temperature data and water demand data from the external environment and the interior of the terminal building by deploying a sensor network, so as to obtain real-time environmental change indicators and load demand indicators. The high temperature classification module is used to classify the impact of high temperature based on real-time environmental change indicators and load demand indicators, using the support vector machine algorithm to determine the extreme weather level and obtain the weather impact coefficient. The data transmission module is used to obtain the weather impact coefficient and then transmit the coefficient to the pressure regulating station control system through the data interaction interface to determine the expected increase in water supply load. The increase judgment module is used to predict the water pipe pressure adjustment demand using a neural network model if the expected increase exceeds the preset threshold, and obtain the optimized pressure value and water reserve allocation plan. The pressure prediction module is used to extract key parameters from the optimized pressure value and water reserve allocation scheme, and send instructions to the pressure regulating station equipment through the control module to obtain the dynamically adjusted operating status. The instruction control module is used to obtain feedback data based on the dynamically adjusted operating status and compare it with the initial load demand index to determine the degree of matching and identify further fine-tuning requirements. The feedback fine-tuning module is used to determine whether there is a need for further fine-tuning. If so, it iteratively optimizes and updates the water pipe pressure and water volume reserve to obtain the final water supply load matching result.

10. An electronic device, characterized in that, include: A memory on which computer programs are stored; A processor is configured to execute the computer program in the memory to implement the multi-pressure station management method based on airport operation data as described in any one of claims 1 to 7.

Images