Analyzing and monitoring method and system based on operation state of smart park

Abstract

The invention relates to the technical field of smart park management, in particular to an analysis monitoring method and system based on a smart park operation state, and the method comprises the following steps: obtaining park equipment operation data parameters and associated environment parameters, analyzing the equipment operation difference, recognizing the equipment operation state, and predicting the state change trend. And calculating an abnormal event propagation path and rate, determining a response priority, constructing early warning information, and obtaining a smart park monitoring record. According to the method, the accuracy of park equipment state identification is improved by combining the fluctuation analysis of the equipment operation data and the environmental parameters, early warning of abnormal states is realized by combining the time sequence correlation analysis and equipment state prediction, the influence of equipment faults on park operation is reduced, and the accuracy of park equipment state identification is improved through the topological structure and propagation probability calculation. The method achieves the prediction of an abnormal event propagation path, enhances the accuracy and timeliness of emergency response, improves the real-time monitoring and fault prediction capability, and guarantees the efficient operation and safety of a park.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent park management, and particularly to an analysis and monitoring method and system based on the operating status of an intelligent park. Background Art

[0002] The technical field of intelligent park management includes a technical system for centralized monitoring, unified scheduling, and intelligent management of various resources and facilities within the park. The core content includes the perception and data collection of operating elements such as park buildings, energy, security, environment, transportation, personnel, and equipment, as well as the integrated processing and logical judgment of the perceived information to support the operation management, emergency response, and service optimization of the park. The intelligent park management technology conducts real-time monitoring and centralized management of the park operating status by deploying perception terminals, constructing information networks, setting up data platforms, and combining preset operating rules, ensuring the efficient operation of the park functions and improving the overall management level. Its overall systematicness is reflected in the hierarchical construction and coordinated linkage of the perception layer, network layer, platform layer, and application layer. Data penetration and instruction transmission are achieved between layers through standardized interfaces, forming a closed-loop management logic.

[0003] Among them, the analysis and monitoring method based on the operating status of an intelligent park refers to a method that, for the status information generated during the operation of each subsystem and facility within the park, screens, classifies, and standardizes different types of operating statuses by setting classification criteria, and establishes a status recognition model and judgment process based on preset rules. The method covers the acquisition of specific information such as environmental monitoring data, power system parameters, security alarm information, equipment working hours, and start-stop status. The analysis and monitoring of key statuses are realized by setting elements such as data reading frequency, status recognition threshold, and logical judgment rules. The analysis and monitoring process is completed by setting monitoring points, arranging data reading modules, using a unified data format interface, and introducing a rule logic judgment mechanism to achieve the acquisition and judgment of the operating status.

[0004] Traditional intelligent park operating status analysis and monitoring technologies rely on fixed-rule-based monitoring and data processing for the real-time monitoring and abnormal early warning of equipment status. They have weak capabilities in the refined status analysis and abnormal status prediction of equipment within the park, lack sufficient data correlation and trend prediction functions for the differential analysis between different equipment. When the equipment status changes, only preset thresholds are relied on for alarm, without considering the dynamic trend of equipment status changes and the correlation between equipment. Subtle status changes in the initial stage are not detected in time, missing the best treatment opportunity, resulting in the gradual expansion of equipment failures and triggering systematic failures. There are deficiencies in the prediction of the abnormal status propagation path, unable to accurately simulate the propagation of abnormal events on multiple paths, and unable to provide effective decision support for emergency response. Summary of the Invention

[0005] To solve the technical problems existing in the prior art, an embodiment of the present invention provides an analysis and monitoring method and system based on the operation status of an intelligent park. The technical solution is as follows:

[0006] To achieve the above object, the present invention adopts the following technical solution. The analysis and monitoring method based on the operation status of an intelligent park includes the following steps:

[0007] S1: Obtain the monitoring data of park devices, including operation data parameters and associated environmental parameters. By analyzing the fluctuation amplitude and range of each parameter, identify the operation differences of each device relative to similar devices during the current period. Combine the device type and deployment location to obtain the device operation feature dataset;

[0008] S2: Call the device operation feature dataset, extract the change rate of operation data parameters and the fluctuation trend of associated environmental parameters. According to the temporal correlation of various parameter change characteristics, identify the device status, detect abnormal operation status, and obtain the operation status node set;

[0009] S3: Based on the operation status node set, analyze the change direction, status span, and node order between multiple consecutive status nodes, identify the status evolution trend, predict the status change trend of each device in the park, predict the abnormal status time point, and generate status trend prediction information;

[0010] S4: Call the operation status node set and status trend prediction information, obtain the topological structure of the park operation devices, calculate the propagation probability of abnormal events in multiple directions, predict the propagation path of abnormal events, and generate a path propagation node list.

[0011] As a further solution of the present invention, the device operation feature dataset is specifically the device deployment location, operation data parameters, and associated environmental parameters. The operation status node set specifically refers to device status nodes, abnormal status nodes, and temporal change patterns. The status trend prediction information includes the predicted status change path, abnormal status occurrence period, and status change trend type. The path propagation node list includes node sequences, propagation probability values, and propagation path priorities.

[0012] As a further solution of the present invention, the step of obtaining the monitoring data of park devices, including operation data parameters and associated environmental parameters, by analyzing the fluctuation amplitude and range of each parameter, identifying the operation differences of each device relative to similar devices during the current period, and combining the device type and deployment location to obtain the device operation feature dataset is specifically as follows:

[0013] S101: Obtain the monitoring data of park devices, extract the operation data parameters of each park device, obtain the associated environmental parameters corresponding to the devices, analyze the fluctuation amplitude and fluctuation range of each item of data, and obtain the operation parameter data set;

[0014] S102: According to the operation parameter data set, extract the average value of the fluctuation amplitude of the operation data parameters of the same type of devices and the average value of the fluctuation range of the environmental parameters. Calculate the deviation degree according to the operation differences between each device and the same type of devices, and generate the operation difference identification result;

[0015] S103: Invoke the operation difference identification result, and combine the type information and deployment location coordinates of the devices to generate the device operation feature data set.

[0016] As a further solution of the present invention, the steps of invoking the device operation feature data set, extracting the change rate of the operation data parameters and the fluctuation trend of the associated environmental parameters, and identifying the device status according to the time series correlation of various parameter change characteristics, detecting the abnormal operation status, and obtaining the operation status node set are specifically as follows:

[0017] S201: Invoke the device operation feature data set, extract the change rate and fluctuation range of the operation data parameters within a continuous time window, calculate the gradient change amount and time series fluctuation amplitude of the associated environmental parameters, and generate the parameter dynamic change feature set;

[0018] S202: Invoke the parameter dynamic change feature set, divide the change rate sequence of the operation data parameters according to the sliding window, calculate the correlation coefficient of the change rate of the operation parameters and the fluctuation amplitude of the associated environmental parameters within adjacent windows, analyze the lag correlation weight of the change rate of the operation parameters and the gradient of the associated environmental parameters, and generate the cross-parameter time series correlation matrix;

[0019] S203: Invoke the cross-parameter time series correlation matrix, analyze the deviation level of the device operation status according to the time series correlation of various device parameters, identify the device status, detect the abnormal operation status in real time, and generate the operation status node set.

[0020] As a further solution of the present invention, the steps of analyzing the change direction, state span and node order between multiple consecutive state nodes based on the operation status node set, identifying the state evolution trend, predicting the state change trend of each device in the park, predicting the abnormal state time point, and generating the state trend prediction information are specifically as follows:

[0021] S301: Based on the operation status node set, extract the change direction, state span and node order of multiple consecutive state nodes, analyze the state change amount and change direction of each node, and generate the node change direction data set;

[0022] S302: Analyze the order of device status change nodes according to the node change direction data set, analyze the time relationship between the node orders, identify the trend characteristics of state evolution, and generate state evolution trend characteristics;

[0023] S303: Based on the state evolution trend characteristics, predict the device state change trend according to the trend characteristics and time series of state changes, calculate the abnormal state time points, and generate state trend prediction information.

[0024] As a further solution of the present invention, the steps of calling the operating state node set and the state trend prediction information, obtaining the topological structure of the park operating equipment, calculating the propagation probability of abnormal events in multiple directions, predicting the propagation path of abnormal events, and generating a path propagation node list are specifically as follows:

[0025] S401: Based on the operating state node set and the state trend prediction information, obtain the topological structure of the smart park equipment, including the power supply, water supply, and network structures, according to the physical distance and network connection strength between devices, and generate a device topological structure data set;

[0026] S402: Based on the device topological structure data set, extract the node positions of abnormal state devices, and calculate the propagation probability of abnormal events in multiple directions by combining the connection strength, distance, and device status between devices, and generate an event propagation probability data set;

[0027] S403: Call the event propagation probability data set, predict the propagation path of abnormal events, identify the propagation nodes of abnormal events, and generate a path propagation node list.

[0028] As a further solution of the present invention, the specific formula for calculating the propagation probability of abnormal events in multiple directions is:

[0029]

[0030] Calculate the propagation probability;

[0031] Among them, P i′,j′ represents the abnormal propagation probability from device i' to device j', C i′,j′ represents the connection strength between device i' and device j', D i′,j′ represents the influence on directionality, S i and S j represent the state sensitivities of device i' and device j', d i′,j′ represents the physical distance between device i' and device j', i' represents the index of the starting device, and j' represents the index of the target device.

[0032] As a further solution of the present invention, the method further includes:

[0033] S5: Based on the list of path propagation nodes, obtain the status change information of each node, calculate the abnormal propagation time interval, analyze the propagation rate of the abnormal status on multiple paths, calculate the response priorities of multiple nodes, match the processing measures in combination with the device type of the target node and construct early warning information, and obtain the monitoring records of the intelligent park;

[0034] The monitoring records of the intelligent park include early warning information, response priorities, and the propagation rate of the abnormal status.

[0035] As a further solution of the present invention, the steps of obtaining the status change information of each node based on the list of path propagation nodes, calculating the abnormal propagation time interval, analyzing the propagation rate of the abnormal status on multiple paths, calculating the response priorities of multiple nodes, matching the processing measures in combination with the device type of the target node and constructing early warning information, and obtaining the monitoring records of the intelligent park are specifically as follows:

[0036] S501: Based on the list of path propagation nodes, obtain the status change information of each node, analyze the status change rate between device nodes by calculating the time interval of status changes, and generate a status change data set;

[0037] S502: According to the status change data set, analyze the propagation rate of the abnormal status on multiple paths, calculate the propagation delay of the abnormal event between nodes by comparing the propagation duration on multiple paths and the connection strength between devices, and generate an abnormal propagation rate data set;

[0038] S503: Call the abnormal propagation rate data set, combine the node characteristics in the list of path propagation nodes, calculate the response priority of each node, match the type of processing measures according to the device type, construct early warning information, and generate the monitoring records of the intelligent park;

[0039] The specific formula for calculating the response priority of each node is:

[0040]

[0041] Calculate the response priority of each node;

[0042] where, P z is the response priority of the z-th node, V z is the device status change rate of the z-th node, ΔT z is the abnormal propagation time interval of the z-th node, C z is the device importance coefficient of the z-th node, D z is the device health coefficient of the z-th node, and z is the index of the node.

[0043] On the other hand, an analysis and monitoring system based on the operation status of an intelligent park is provided. This system is applied to the analysis and monitoring method based on the operation status of an intelligent park. The system includes:

[0044] The device feature extraction module extracts operation data parameters and associated environment parameters based on the park device monitoring data, calculates the fluctuation amplitude and fluctuation range of each parameter, analyzes the operation differences of each device relative to similar devices in real time, and combines the device type and location coordinates to obtain a device operation feature data set;

[0045] The state coupling analysis module calls the device operation feature data set, analyzes the operation data parameters and associated environment parameters, extracts the data change rate and fluctuation trend, analyzes the time series correlation of various parameter change characteristics, identifies the operation state of the device in real time, and establishes an operation state node set;

[0046] The trend evolution prediction module analyzes the change direction, state span and node order of each state node based on the operation state node set, predicts the state change trends of multiple devices, predicts and calculates the abnormal state time points, and generates state trend prediction information;

[0047] The abnormal propagation deduction module calls the operation state node set and the state trend prediction information, obtains the topological connection structure between each device in the park, analyzes the probability and direction of abnormal event propagation between nodes in the topological structure, predicts the propagation path of abnormal events, and establishes a path propagation node list;

[0048] The intelligent response decision module obtains the real-time state change information of each node based on the path propagation node list, calculates the time interval of abnormal state propagation between nodes, analyzes the propagation rate of multi-path abnormal states, calculates the response priorities of multiple nodes, matches the processing measures in combination with the node device type, constructs early warning information, and generates an intelligent park monitoring record.

[0049] The beneficial effects brought by the technical solution provided by the embodiments of the present invention at least include:

[0050] By combining the fluctuation analysis of device operation data and environmental parameters, the accuracy of park device state identification is improved. By combining time series correlation analysis and device state prediction, early warning of abnormal states is realized, reducing the impact of device failures on park operations. By calculating the topological structure and propagation probability, the prediction of the propagation path of abnormal events is realized, enhancing the accuracy and timeliness of emergency response, improving the real-time monitoring and fault prediction capabilities, and ensuring the efficient operation and safety of the park. Brief Description of the Drawings

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0052] Figure 1 is a schematic diagram of the work flow of the present invention;

[0053] Figure 2 is a system flow chart of the present invention. Specific embodiments

[0054] The following will describe the technical solutions in the present invention with reference to the accompanying drawings.

[0055] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to represent examples, illustrations or explanations. Any embodiment or design solution described as "example" in the present invention should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly, the use of the word "example" is intended to present concepts in a specific manner. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one of the two.

[0056] In the embodiments of the present invention, "image" and "picture" can sometimes be used interchangeably. It should be noted that when the difference is not emphasized, the meaning they express is the same. "(of)", "corresponding", and "corresponding" can sometimes be used interchangeably. It should be noted that when the difference is not emphasized, the meaning they express is the same.

[0057] In the embodiments of the present invention, sometimes subscripts such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0058] To make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0059] Please refer to Figure 1 , the present invention provides a technical solution, an analysis and monitoring method based on the operation status of a smart park, including the following steps:

[0060] S1: Obtain the monitoring data of park equipment, including operation data parameters and associated environmental parameters. By analyzing the fluctuation amplitude and fluctuation range of each parameter, identify the operation differences of each device relative to similar devices in the current period. Combine the device type and deployment location to obtain the device operation characteristic data set;

[0061] S2: Call the device operation feature data set, extract the change rate of operation data parameters and the fluctuation trend of associated environmental parameters, identify the device status according to the temporal correlation of various parameter change characteristics, detect abnormal operation status, and obtain the operation status node set;

[0062] S3: Based on the operation status node set, analyze the change direction, status span, and node order between multiple consecutive status nodes, identify the status evolution trend, predict the status change trend of each device in the park, predict the time point of abnormal status, and generate status trend prediction information;

[0063] S4: Call the operation status node set and status trend prediction information, obtain the topological structure of the park operation devices, calculate the propagation probability of abnormal events in multiple directions, predict the propagation path of abnormal events, and generate a path propagation node list;

[0064] S5: Based on the path propagation node list, obtain the status change information of each node, calculate the abnormal propagation time interval, analyze the propagation rate of abnormal status on multiple paths, calculate the response priority of multiple nodes, match the processing measures in combination with the device type of the target node and construct warning information, and obtain the intelligent park monitoring record.

[0065] The device operation feature data set specifically refers to the device deployment location, operation data parameters, and associated environmental parameters. The operation status node set specifically refers to device status nodes, abnormal status nodes, and temporal change patterns. The status trend prediction information includes the predicted status change path, abnormal status occurrence period, and status change trend type. The path propagation node list includes node sequence, propagation probability value, and propagation path priority. The intelligent park monitoring record includes warning information, response priority, and abnormal status propagation rate.

[0066] Obtain the park device monitoring data, including operation data parameters and associated environmental parameters. By analyzing the fluctuation amplitude and fluctuation range of each parameter, identify the operation differences of each device relative to similar devices in the current period. Combining the device type and deployment location, the steps to obtain the device operation feature data set are specifically as follows:

[0067] S101: Obtain the park device monitoring data, extract the operation data parameters of each park device, and obtain the associated environmental parameters corresponding to the device. Analyze the fluctuation amplitude and fluctuation range of each data item to obtain the operation parameter data set;

[0068] The park data acquisition submodule first collects real-time data from operating units such as air-conditioning units, power cabinets, and exhaust systems by deploying environmental and equipment monitoring devices in the building partitions. Taking air-conditioning units numbered E01, E02, and E03 as examples, the monitoring time period is set to 10:00 to 10:05, and the equipment operating temperature and power consumption values are recorded every minute. A total of 5 sets of data are collected for each device. After extracting the parameter sequence of each device, the maximum and minimum values of the temperature and power consumption data are extracted respectively, and the difference calculation operation is performed to obtain the fluctuation range of each device parameter, and then Get the running stability status in the current time window. Specifically, the temperature sampling values of the E01 device are 22.1, 22.3, 22.0, 22.2, and 22.4. By subtracting 22.4 from 22.0, the temperature fluctuation amplitude is 0.4. The power consumption sampling values are 60, 62, 61, 60, and 63, and the fluctuation amplitude is 63 minus 60, which is 3. Repeat the above operation to obtain the corresponding fluctuation values of the E02 and E03 devices. After uniformly recording the fluctuation amplitudes of various parameters of the same type of equipment, calculate the average of the temperature and power consumption fluctuations as the standard reference, using the formula:

[0069]

[0070] Calculate the fluctuation mean, where is the average fluctuation range of a parameter, x i is the fluctuation range of the corresponding parameter of the i-th device, and n is the number of devices involved in the calculation. Assuming that the temperature fluctuation range is 0.4, 0.5, and 0.4, substitute it into the formula to get:

[0071]

[0072] The results show that the temperature fluctuations of the equipment group are concentrated around 0.433, which provides a benchmark for the subsequent deviation degree judgment. The temperature, power consumption and corresponding fluctuation mean information are stored in a unified structured manner to form an operation parameter data set. This data set provides basic input for subsequent operation difference judgment. The parameter items contained in it have been fully collected, processed and archived, with traceability and classification identification. The parameters are bound to the equipment number, and the time period structure and equipment spatial location information are recorded.

[0073] S102: extracting the mean fluctuation amplitude of the operating data parameters and the mean fluctuation range of the environmental parameters of the same type of equipment according to the operating parameter data set, calculating the deviation according to the operating difference between each device and the same type of equipment, and generating an operating difference identification result;

[0074] After obtaining the operating parameter dataset, the device difference evaluation sub-module extracts the average temperature fluctuation and average power consumption fluctuation of all similar devices in the current time period as the reference standard, and performs a difference deviation calculation between the current fluctuation amplitude of each device and the mean value. Taking the temperature fluctuation of device E02 as 0.5 and the population mean as 0.433, the fluctuation difference degree is expressed in the form of the ratio of the difference value to the mean value, and the deviation degree calculation is performed using the formula:

[0075]

[0076] where D is the deviation degree, x is the parameter fluctuation amplitude of a certain device, is the population mean of the corresponding parameter. Assuming the temperature fluctuation of device E02 is 0.5 and substituting the above mean value 0.433 for calculation:

[0077]

[0078] It indicates that device E02 has a deviation degree of 0.1548 in the temperature parameter compared with the average level. By analogy, the difference deviation degree calculation is completed for all devices and their parameter items, and the interval judgment threshold is set. The deviation degree greater than 0.1 is defined as the "significant difference interval", and less than or equal to 0.1 is classified into the "slight difference interval", and different judgment intervals are mapped to difference labels. The difference level, specific deviation value, device number, and corresponding parameter item are uniformly encapsulated into a structure item to form the operation difference recognition result, which is used for subsequent combination with device type and location information to support the extraction of operation behavior patterns.

[0079] S103: Call the operation difference recognition result, and combine the type information and deployment location coordinates of the device to generate the device operation feature dataset;

[0080] After the device feature labeling sub-module calls the operation difference recognition result and obtains the difference label and deviation degree value of each device, it consults the device registration form to extract the device number and device type information. Taking E02 as an example, the number corresponds to the B series constant temperature device, and the deployment location is inside the east side floor of Building 2. According to the difference label being "significant difference", the location coordinate being the east side partition, and the type being the B type air-conditioning device, the operation feature information of this device is marked as "east side - B type - temperature deviation", while the device numbered E01 is marked as "west side - B type - slight deviation". According to the combination of the three elements of spatial position, device type, and difference degree, the complete feature record of each device is extracted, and the same type of devices are classified and sorted according to the location aggregation degree using the formula:

[0081]

[0082] Among them, R represents the spatial distance between equipment rooms, and (x1, y1) and (x2, y2) are the coordinates of the deployment points of two pieces of equipment. Assuming the coordinates of E01 are (10, 20) and those of E02 are (15, 20), substituting into the calculation gives:

[0083]

[0084] It indicates that the distance between E01 and E02 is 5 meters, which can be used for subsequent feature clustering processing to construct the operation feature labels of each piece of equipment under a specific deployment background. After jointly encoding information such as type, location, and difference level, an equipment operation feature dataset is formed, and a structured record is constructed for use as the input source in the trend recognition and path analysis stages.

[0085] The steps of calling the equipment operation feature dataset, extracting the change rate of operation data parameters and the fluctuation trend of associated environmental parameters, and identifying the equipment status according to the temporal correlation of various parameter change characteristics, detecting abnormal operation status, and obtaining the set of operation status nodes are specifically as follows:

[0086] S201: Call the equipment operation feature dataset, extract the change rate and fluctuation range of operation data parameters within a continuous time window, calculate the gradient change amount and temporal fluctuation amplitude of associated environmental parameters, and generate a parameter dynamic change feature set;

[0087] Call the equipment operation feature dataset. First, read the operation data parameters of the equipment within a continuous time period. Select the operation data of the cooling fan F01 from 9:00 to 9:10 as the analysis object, collect the power consumption value and temperature value once per minute, for a total of 10 groups of data. The power consumptions are 220, 225, 223, 228, 230, 231, 232, 234, 233, 235 watts respectively, and the temperature values are 31.1, 30.9, 30.7, 30.5, 30.2, 30.3, 30.1, 30.0, 29.8, 29.9 degrees Celsius. Divide this time period into 5 two-minute windows, and calculate the change rate between each two groups of data within the window. Taking the power consumption parameter as an example, the change rate of power consumption within the first window is (225 - 220) / 220 = 0.0227, and the second window is (228 - 223) / 223 = 0.0224, and so on to calculate the continuous change rate. The temperature change rate is the same, and the relative change rate sequence of each parameter within each window is obtained; at the same time, after the same window division of the sampled values of environmental temperature and humidity parameters, calculate the difference between the maximum value and the minimum value as the fluctuation range, and further arrange the change amplitude of the differences in chronological order. Taking humidity as an example, the humidity values in the first window are 41.2% and 42.0%, and the fluctuation amplitude is 0.8%. The second window is 42.0% and 42.5%, and the fluctuation is 0.5%. The formula for calculating the fluctuation change amount within its window is:

[0088] G = |Vi -V i-1 |;

[0089] where G is the gradient change of the i-th window, and V i represents the fluctuation amplitude of the current window environment parameter, and V i-1 is the fluctuation amplitude of the corresponding parameter in the previous window. Assume that the humidity fluctuation in the 3rd window is 1.0%, and the fluctuation in the previous window is 0.5%, then the gradient change is:

[0090] G = |1.0 - 0.5| = 0.5;

[0091] This gradient change value represents the severity of the fluctuation change trend of the environmental parameters. Record the change rate of each operating parameter and the gradient change value of each environmental parameter under all windows as parameter structure items, and combine them to form a parameter dynamic change feature set. The feature set records the continuous change behavior information of each period's parameters in the time dimension, which can be used as the basic input for subsequent sequence calculations.

[0092] S202: Call the parameter dynamic change feature set, divide the running data parameter change rate sequence according to the sliding window, calculate the correlation coefficient between the running parameter change rate and the associated environmental parameter fluctuation amplitude within adjacent windows, analyze the lag correlation weight between the running parameter change rate and the associated environmental parameter gradient, and generate a cross-parameter time series correlation matrix;

[0093] Call the parameter dynamic change feature set, perform sliding window division analysis on the power consumption change rate and the environmental humidity gradient within continuous windows. Set the sliding window span to 2 minutes and the step size to 1 minute. Construct a sequence group for the running parameter change rate. Take the power consumption change rate sequence of device F01 as 0.0227, 0.0224, 0.0216, 0.0215, 0.0210, 0.0209, 0.0205, 0.0203, and the humidity change sequence as 0.8%, 0.5%, 1.0%, 1.3%, 1.1%, 0.9%, 0.6%, 0.7%. Each two windows form a group and construct them into a one-to-one corresponding sliding window data pair. First, take the power consumption change rate of the first sliding window data pair as 0.0227, 0.0224, and the corresponding humidity change as 0.8%, 0.5%, and calculate their correlation coefficient. Use the Pearson correlation coefficient formula:

[0094]

[0095] where r is the correlation coefficient, x ix is the running parameter change rate of the ix-th window, y ix is the gradient change amount of the corresponding window environmental parameter, is the mean of their respective sequences.

[0096] Assume that the power consumption change rates of the first sliding window are 0.0227 and 0.0224, and the humidity changes are 0.8% and 0.5%. Calculate the mean value:

[0097]

[0098] The correlation coefficient r = 0.0317 indicates that the linear relationship between the two is very weak.

[0099] Continue to perform the same calculation on subsequent sliding windows, calculate the correlation coefficient of each pair of window data, and record each calculation result in the correlation coefficient matrix.

[0100] Further determine the number of steps by which the change in operating parameters lags behind the environmental parameter gradient in adjacent windows, record the proportion of the frequency of occurrence of this lag phenomenon, set the weight ratio threshold, and when the lag ratio exceeds the set threshold (such as 50%), record it as having a lag correlation relationship. In this way, construct a time-series lag matching relationship matrix between multi-dimensional parameters, record the correlation coefficients and lag steps between all device parameter pairs, and form a cross-parameter time-series correlation matrix, which is used as the input basis for running behavior correlation analysis.

[0101] S203: Invoke the cross-parameter time-series correlation matrix, analyze the deviation level of the device operating state according to the time-series correlation of various device parameters, identify the device state, detect abnormal operating states in real time, and generate a set of operating state nodes;

[0102] Invoke the cross-parameter time-series correlation matrix, extract the time-series relationships between parameter items such as power consumption, temperature, vibration, and humidity in the same device one by one, analyze the correlation coefficients of each pair of parameters in different time windows, and include the parameter pairs with a correlation coefficient greater than 0.7 in the strong correlation items. Expand the time window segments composed of these strong correlation items by minutes, calculate the window deviation value, in the form of the mean absolute deviation, and the formula is as follows:

[0103]

[0104] Among them, M is the device operating state deviation value, x iy is the comprehensive change rate of operating parameters at the iy-th moment, μ is the mean value of the parameter change rate of the same device during this period, and ny is the number of sampling points in the total duration of the time period. Assume that the comprehensive change rates of operating parameters of a certain air-conditioning device at 6 time points are 0.05, 0.06, 0.07, 0.05, 0.06, and 0.09, and the mean value is 0.0633. Substitute into the formula to get the deviation value:

[0105]

[0106] Set the deviation level threshold according to the device type. When the M value is less than 0.01, it is classified into level I; when it is between 0.01 and 0.02, it is classified into level II; when it is greater than 0.02, it is classified into level III. Mark the device status level and classify it into the status node, and finally output the set of operating status nodes for subsequent evolution trend analysis and status prediction.

[0107] Based on the set of operating status nodes, analyze the change direction, status span, and node order between multiple consecutive status nodes, identify the status evolution trend, predict the status change trend of each device in the park, predict the abnormal status time point, and the steps to generate the status trend prediction information are specifically as follows:

[0108] S301: Based on the set of operating status nodes, extract the change direction, status span, and node order of multiple consecutive status nodes, analyze the status change amount and change direction of each node, and generate a node change direction data set;

[0109] Call the device operation feature data set. First, extract the status change data of each device within a continuous period from the set of operating status nodes of the device. Assume that the status of device F01 is "normal" from 9:00 to 9:05, "slightly abnormal" from 9:05 to 9:10, and "severely abnormal" from 9:10 to 9:15. These status changes are judged according to the fluctuations of parameters such as the temperature and power consumption of the device. Further extract the change direction of the status. For example, from "normal" to "slightly abnormal" indicates that the change direction is "rising", from "slightly abnormal" to "severely abnormal" indicates that the change direction is "intensifying", and from "severely abnormal" to "return to normal" indicates that the change direction is "falling". The calculation of the status span is the time the device persists in each status. For example, from 9:00 to 9:05 is 5 minutes, from 9:05 to 9:10 is 5 minutes, and from 9:10 to 9:15 is 5 minutes. After extracting these data, analyze the status change amount of each node (that is, the magnitude of the status change within each status span). For example, the change amount from "normal" to "slightly abnormal" is 0.2, indicating a slight fluctuation in the device status; the change amount from "slightly abnormal" to "severely abnormal" is 0.5, indicating a significant fluctuation in the device status. Use the formula:

[0110] ΔS=S end -S start ;

[0111] where ΔS represents the status change amount, S end is the status value at the end of the status, and S start is the status value at the start of the status. Assume that the change amount of the device from "normal" to "slightly abnormal" is 0.2, and the change amount of the device from "slightly abnormal" to "severely abnormal" is 0.5. Substitute into the formula for calculation:

[0112] ΔS1 = 0.2, ΔS2 = 0.5;

[0113] Generate a node change direction dataset for each node, including the change direction, state span, and change amount, for subsequent analysis.

[0114] S302: Analyze the order of device state change nodes based on the node change direction dataset, analyze the time relationship between the node orders, identify the trend characteristics of state evolution, and generate state evolution trend characteristics;

[0115] Based on the node change direction dataset, analyze the order of device state change nodes. First, extract the state change order of each device in different time periods. Taking device F01 as an example, assume its state is "normal" from 9:00 to 9:05, "slightly abnormal" from 9:05 to 9:10, and "severely abnormal" from 9:10 to 9:15. According to the time order of state changes, construct a time relationship matrix and calculate the time differences (5 minutes from 9:00 to 9:05, 5 minutes from 9:05 to 9:10, and 5 minutes from 9:10 to 9:15). Then, analyze the trend characteristics of state changes. For example, the trend from "normal" to "slightly abnormal" is "rising", the trend from "slightly abnormal" to "severely abnormal" is "intensifying", and the trend from "severely abnormal" to "return to normal" is "falling". To calculate the trend intensity of state changes, use the formula:

[0116]

[0117] where T is the trend intensity, S end is the state end value, S start is the state start value, and Δt is the time span of state change. Assume the state change amount from "normal" to "slightly abnormal" is 0.2 and the change time is 5 minutes. Substitute into the formula for calculation:

[0118]

[0119] Calibrate the device state according to this trend intensity. If T ≥ 0.05, it is defined as a strong trend; if T < 0.05, it is a weak trend. By calculating the trend intensity of each device's state change, identify the characteristics of state evolution, and generate the state evolution trend characteristics of the device, providing data support for subsequent state prediction.

[0120] S303: Based on the state evolution trend characteristics, predict the device state change trend according to the trend characteristics and time series of state changes, calculate the abnormal state time point, and generate state trend prediction information;

[0121] Predict the change trend of the device status based on the status evolution trend characteristics and time series. First, based on the trend patterns extracted from the status evolution trend characteristics (such as "rising", "intensifying", "falling"), combined with the time series of the device operation, assume that device F01 has experienced a status evolution from "normal" to "slightly abnormal" to "severely abnormal" in the past 1 hour. Then, through the known time series information and status evolution trend characteristics, predict its future status. For example, it is predicted that device F01 may recover from the "severely abnormal" status to the "slightly abnormal" status within the next 5 minutes. To calculate the occurrence time point of the abnormal status, the formula is used:

[0122]

[0123] where Δt pred is the prediction time, ΔS is the status change amount, and T is the trend intensity. Assume that the change amount of the status of device F01 from "severely abnormal" to "slightly abnormal" is 0.5 and the trend intensity is 0.04. Substitute these values into the formula for calculation:

[0124]

[0125] The result shows that device F01 is expected to recover from the "severely abnormal" status to the "slightly abnormal" status within the next 12.5 minutes. Based on this prediction information, generate status trend prediction information to further provide dynamic decision support for device operation management.

[0126] The steps of calling the set of running status nodes and status trend prediction information, obtaining the topological structure of the devices operating in the park, calculating the propagation probability of abnormal events in multiple directions, predicting the propagation path of abnormal events, and generating a list of path propagation nodes are specifically as follows:

[0127] S401: Based on the set of running status nodes and status trend prediction information, obtain the topological structure of the intelligent park devices, including the energy supply, water supply, and network structure, according to the physical distance and network connection strength between devices, and generate a device topological structure data set;

[0128] Based on the set of nodes in the operating state and the status trend prediction information, first obtain the physical distance and network connection strength between devices. Assume that there are devices F01, F02, and F03 in different floors in the park. Device F01 is located in Building 1, device F02 is located in Building 2, and device F03 is located in Building 3. The physical distance between devices is the straight-line distance between every two devices. Taking the physical distance between device F01 and F02 as an example, assume the distance between them is 50 meters, and the distance between F02 and F03 is 30 meters. Next, evaluate according to the connection strength between devices. Assume that the network connection strength between F01 and F02 is 0.8 (representing a strong connection), and the connection strength between F02 and F03 is 0.5 (representing a medium connection). These data will help generate the topological structure of the devices in the park, draw a topological map through the connection strength and physical distance between devices, further refine the energy supply, water supply, and network structure of the park, and generate a device topological structure data set according to these data. This data set contains attributes such as the location information of each device, the physical distance between devices, and the network connection strength. Combining the detailed information of the energy supply, water supply, and network structure, it forms a comprehensive relationship diagram between the devices in the park, reflecting the interconnection relationship between the devices in the park.

[0129] S402: Based on the device topological structure data set, extract the node positions of the devices in the abnormal state. Combining the connection strength, distance, and device status between devices, calculate the propagation probability of the abnormal event in multiple directions, and generate an event propagation probability data set;

[0130] The specific formula for calculating the propagation probability of the abnormal event in multiple directions is:

[0131]

[0132] Calculate the propagation probability;

[0133] Among them, P i′,j′ represents the abnormal propagation probability from device i' to device j', C i′,j′ represents the connection strength between device i' and device j', D i′,j′ represents the influence directionality, S i and S j represent the state sensitivity of device i' and device j', d i′,j′ represents the physical distance between device i' and device j', i' represents the index of the starting device, and j' represents the index of the target device as the target device.

[0134] Formula:

[0135]

[0136] Detailed explanation of the formula and the derivation process of the formula calculation:

[0137] This formula is used to calculate the propagation probability of abnormal events between devices.

[0138] Parameter meanings and set values:

[0139] S i and S j : The state sensitivities of devices i' and j', which represent the ability of devices to respond to external abnormalities. There are two devices set, an air conditioner and a generator. Among them, S i = 3, S j = 15;

[0140] C i′,j′ : The connection strength between devices i' and j'. This parameter measures the strength of the physical connection or dependence relationship between devices, such as power supply, control signal transmission, etc. Set C i′,j′ = 0.85, indicating a strong power supply connection between the air conditioner and the generator;

[0141] D i′,j′ : The influence directionality, which represents the influence direction from device i' to device j'. The upstream device has a greater influence on the downstream device. Assuming that the generator is the upstream device of the air conditioner, set D i′,j′ = 1;

[0142] d i′,j′ : The physical distance between devices i' and j'. Set d i′,j′ = 50;

[0143] Substitute the parameters into the formula for calculation:

[0144]

[0145] 0.85·1·15 = 12.75;

[0146] S i +S j = 3 + 15 = 18;

[0147]

[0148] 18·7.071 ≈ 127.28;

[0149]

[0150] The result P i′,j′ ≈ 0.1001 indicates that when the generator is abnormal, the propagation probability of the air conditioner being abnormal is 10.01%. The propagation probability takes into account the connection strength between devices, the power supply dependence relationship, the fluctuation sensitivity of device states, and the physical distance. The result reflects the possibility of abnormal propagation between devices and can provide decision-making support for park management to identify which devices have a higher risk of abnormality.

[0151] S403: Call the event propagation probability dataset, predict the propagation path of the abnormal event, identify the propagation nodes of the abnormal event, and generate a path propagation node list;

[0152] Call the event propagation probability dataset to further predict the propagation path of the abnormal event. Assume that device F01 has an abnormality, and the propagation probabilities of F02 and F03 are calculated to be 0.016 and 0.0167 respectively. Then analyze the propagation path of the F01 abnormal event. Based on the propagation probability, assume that the abnormal state of device F01 first propagates to F02, and then from F02 to F03. By analyzing each node in the propagation path (such as F02, F03, etc.), record the propagation nodes and their state changes, and generate a path propagation node list. For example, if device F01 has an abnormality, according to the calculated propagation probability, device F02 becomes the first device to receive the abnormality and further transmits it to F03. Through the propagation probability between devices, F02 and F03 can be identified as the propagation nodes of the abnormal state, and a path propagation node list is generated. The final propagation path is F01->F02->F03, providing a decision basis for subsequent abnormal management and monitoring.

[0153] Based on the path propagation node list, obtain the state change information of each node, calculate the abnormal propagation time interval, analyze the propagation rate of the abnormal state on multiple paths, calculate the response priorities of multiple nodes, match the processing measures in combination with the device type of the target node and construct early warning information. The steps to obtain the monitoring records of the smart park are specifically as follows:

[0154] S501: Based on the path propagation node list, obtain the state change information of each node. By calculating the time interval of the state change, analyze the state change rate between device nodes, and generate a state change dataset;

[0155] Based on the path propagation node list, obtain the state change information of each node. First, extract the state change data of each device node. Assume that the states of devices F01, F02, and F03 change at different time periods. For example, the state of F01 is "normal" from 9:00 to 9:05, "slightly abnormal" from 9:05 to 9:10, and "severely abnormal" from 9:10 to 9:15. By calculating the time intervals of these state changes, assume that the time interval of the state change of F01 between 9:00 and 9:10 is 10 minutes, and the time interval between 9:10 and 9:15 is 5 minutes. Then, analyze the state change rate between device nodes. The rate can be calculated by dividing the state change amount by the time interval. Assume that the change amount of F01 from "normal" to "slightly abnormal" is 0.2 and the change time is 10 minutes. The state change rate of F01 is calculated as:

[0156]

[0157] Among them, R F01 is the state change rate, ΔS is the state change amount, and Δt is the time interval of the state change. Substitute into the calculation:

[0158]

[0159] This result shows that the state change rate of F01 is 0.02 per minute, and by analogy, calculate the state change rates of other devices. Generate a state change data set, record the state change information and the corresponding state change rates of each device node, and provide data support for subsequent propagation analysis.

[0160] S502: According to the state change data set, analyze the propagation rate of the abnormal state on multiple paths. By comparing the propagation duration on multiple paths and the connection strength between devices, calculate the propagation delay of the abnormal event between nodes, and generate an abnormal propagation rate data set;

[0161] According to the state change data set, analyze the propagation delay of the abnormal state propagating on multiple paths. First, extract the device node information, the connection strength between devices, and the physical distance in the state change data set. For example, assume that there are different physical distances and network connection strengths between F01, F02, and F03. The distance between F01 and F02 is 50 meters, the connection strength is 0.8, the distance between F02 and F03 is 30 meters, and the connection strength is 0.5. Next, calculate the propagation delay of the abnormal event from one node to the next node. The propagation delay can be estimated by the proportional relationship between the physical distance between devices and the connection strength. Assume the formula is:

[0162]

[0163] Among them, T ij represents the propagation delay from device i to device j, d ij is the physical distance between device i and device j, and C ij is the connection strength between devices.

[0164] Assume that the distance between F01 and F02 is 50 meters and the connection strength is 0.8, and substitute into the formula for calculation:

[0165]

[0166] For F02 and F03, assume the distance is 30 meters and the connection strength is 0.5, and substitute into the formula for calculation:

[0167]

[0168] The calculation shows that the propagation delay between F01 and F02 is 62.5 seconds, and the propagation delay between F02 and F03 is 60 seconds. By calculating the propagation delay of each path, an abnormal propagation rate data set can be generated for each node, recording the propagation delay information of each path. This data set provides basic data for subsequent abnormal event propagation predictions, helping managers understand the propagation of abnormal conditions between campus devices.

[0169] S503: Call the abnormal propagation rate data set, combine the node characteristics in the path propagation node list, calculate the response priority of each node, match the processing measure type according to the device type, build warning information, and generate a smart park monitoring record;

[0170] The specific formula for calculating the response priority of each node is:

[0171]

[0172] Calculate the response priority of each node;

[0173] Among them, P z is the response priority of the zth node, V z is the device state change rate of the zth node, ΔT z is the abnormal propagation time interval of the zth node, C z is the equipment importance coefficient of the zth node, D z is the equipment health coefficient of the z-th node, and z is the index of the node.

[0174] formula:

[0175]

[0176] Detailed explanation of the formula and the process of formula calculation and derivation:

[0177] The formula is used to calculate the response priority of node z and determine the priority response order of the devices.

[0178] Parameter meaning and setting value:

[0179] V z is the state change rate of device z, reflecting the speed of the device state change in unit time. If the power consumption of the device changes from 3000W to 3500W in unit time, then V z 500W / min;

[0180] ΔT z is the abnormal propagation time interval of device z, which represents the propagation delay of abnormal events between nodes. Set ΔT z = 10 minutes;

[0181] Cz The device importance coefficient of device z, which reflects the relative importance of the device within the park, is set as C z = 0.8;

[0182] D z The health coefficient of device z, which represents the current health status of the device, is set as D z = 0.9;

[0183] Substitute the parameters into the formula for calculation:

[0184] Set the parameter values as follows:

[0185]

[0186] P z = 4.1322 × 1.8889;

[0187] P z ≈ 7.8;

[0188] The calculation results show that the response priority P of the device z is 7.8. A higher response priority value indicates that when an abnormal state occurs for this device, it needs to be responded to and processed preferentially. This priority value will affect the scheduling order of the system when multiple devices have abnormalities simultaneously, ensuring timely intervention for the most important devices with a greater potential for fault spread.

[0189] Please refer to Figure 2 , the analysis and monitoring system based on the operation status of the smart park. The analysis and monitoring system based on the operation status of the smart park is used to execute the above-mentioned analysis and monitoring method based on the operation status of the smart park. The system includes:

[0190] The device feature extraction module extracts operation data parameters and associated environmental parameters based on the park device monitoring data, calculates the fluctuation amplitude and fluctuation range of each parameter, analyzes the operation differences of each device relative to similar devices in real time, and combines the device type and location coordinates to obtain the device operation feature data set;

[0191] The state coupling analysis module calls the device operation feature data set, analyzes the operation data parameters and associated environmental parameters, extracts the data change rate and fluctuation trend, analyzes the temporal correlation of various parameter change characteristics, identifies the operation state of the device in real time, and establishes an operation state node set;

[0192] The trend evolution prediction module analyzes the change direction, state span, and node order of each state node based on the operation state node set, predicts the state change trends of multiple devices, predicts and calculates the abnormal state time points, and generates state trend prediction information;

[0193] The abnormal propagation deduction module calls the set of operation status nodes and the status trend prediction information, obtains the topological connection structure of each equipment room in the park, analyzes the probability and direction of abnormal event propagation between nodes in the topological structure, predicts the propagation path of abnormal events, and establishes a list of path propagation nodes;

[0194] Based on the list of path propagation nodes, the intelligent response decision module obtains the real-time status change information of each node, calculates the time interval of abnormal status propagation between nodes, analyzes the propagation rate of multi-path abnormal status, calculates the response priorities of multiple nodes, matches the processing measures in combination with the node device types, constructs early warning information, and generates a monitoring record of the intelligent park.

[0195] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware, or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions according to the embodiments of the present invention are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that includes one or more sets of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0196] It should be understood that the term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Among them, A and B can be singular or plural. In addition, the character " / " in this article generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood with reference to the context before and after.

[0197] In the present invention, "at least one" means one or more, and "a plurality of" means two or more than two. "At least one of the following" or a similar expression means any combination of these items, including any combination of single item or plural items. For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0198] It should be understood that in various embodiments of the present invention, the sequence numbers of the above - mentioned processes do not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0199] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0200] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above - described devices, apparatuses, and units can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0201] In several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of the devices or units can be in an electrical, mechanical, or other form.

[0202] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0203] In addition, in each embodiment of the present invention, each functional unit can be integrated into a processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0204] If the above-mentioned function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs.

[0205] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A method for analyzing and monitoring the operating status of an intelligent park, characterized in that, The method includes: S1: Obtain the monitoring data of park devices, including operation data parameters and associated environmental parameters. By analyzing the fluctuation amplitude and range of each parameter, identify the operation differences of each device relative to similar devices during the current period. Combine the device type and deployment location to obtain the device operation feature dataset. S2: Invoke the device operation feature dataset, extract the change rate of operation data parameters and the fluctuation trend of associated environmental parameters. According to the temporal correlation of various parameter change characteristics, identify the device status, detect abnormal operation status, and obtain the operation status node set. S3: Based on the operation status node set, analyze the change direction, status span, and node order between multiple consecutive status nodes, identify the status evolution trend, predict the status change trend of each device in the park, predict the abnormal status time point, and generate status trend prediction information. S4: Invoke the operation status node set and status trend prediction information, obtain the topological structure of the operating devices in the park, calculate the propagation probability of abnormal events in multiple directions, predict the propagation path of abnormal events, and generate a path propagation node list.

2. The analysis and monitoring method based on the operation status of the smart park according to claim 1, wherein The device operation feature dataset specifically refers to the device deployment location, operation data parameters, and associated environmental parameters. The operation status node set specifically refers to device status nodes, abnormal status nodes, and temporal change patterns. The status trend prediction information includes the predicted status change path, abnormal status occurrence period, and status change trend type. The path propagation node list includes node sequence, propagation probability value, and propagation path priority.

3. The analysis and monitoring method based on the operation status of the smart park according to claim 1, wherein, The steps for obtaining the monitoring data of park devices, including operation data parameters and associated environmental parameters, and identifying the operation differences of each device relative to similar devices during the current period by analyzing the fluctuation amplitude and range of each parameter, and then combining the device type and deployment location to obtain the device operation feature dataset are as follows: S101: Obtain the monitoring data of park devices, extract the operation data parameters of each park device, and obtain the associated environmental parameters corresponding to the device. Analyze the fluctuation amplitude and range of each piece of data to obtain the operation parameter dataset. S102: According to the operation parameter dataset, extract the mean value of the fluctuation amplitude of the operation data parameters of similar devices and the mean value of the fluctuation range of environmental parameters. Calculate the deviation degree based on the operation differences between each device and similar devices, and generate an operation difference identification result. S103: Invoke the operation difference identification result, and combine the device type information and deployment location coordinates to generate the device operation feature dataset.

4. The analysis and monitoring method based on the operation status of the intelligent park according to claim 3, wherein, The steps for invoking the device operation feature dataset, extracting the change rate of operation data parameters and the fluctuation trend of associated environmental parameters, and then identifying the device status, detecting abnormal operation status, and obtaining the operation status node set according to the temporal correlation of various parameter change characteristics are as follows: S201: Invoke the device operation feature dataset, extract the change rate and fluctuation range of operation data parameters within a continuous time window, calculate the gradient change amount and temporal fluctuation amplitude of associated environmental parameters, and generate a parameter dynamic change feature set. S202: Invoke the dynamically changing feature set of the parameters, divide the sequence of the change rate of the operation data parameters according to a sliding window, calculate the correlation coefficient between the change rate of the operation parameters within adjacent windows and the fluctuation amplitude of the associated environmental parameters, analyze the lag correlation weight between the change rate of the operation parameters and the gradient of the associated environmental parameters, and generate a cross-parameter time-series correlation matrix; S203: Invoke the cross-parameter time-series correlation matrix, analyze the deviation level of the device operation state according to the time-series correlation of multiple device parameters, identify the device state, detect the abnormal operation state in real time, and generate a set of operation state nodes.

5. The analysis and monitoring method based on the operation status of the intelligent park according to claim 4, characterized in that, The steps of analyzing the change direction, state span, and node order between multiple consecutive state nodes based on the set of operation state nodes, identifying the state evolution trend, predicting the state change trend of each device in the park, predicting the time point of the abnormal state, and generating state trend prediction information are specifically as follows: S301: Based on the set of operation state nodes, extract the change direction, state span, and node order of multiple consecutive state nodes, analyze the state change amount and change direction of each node, and generate a dataset of node change directions; S302: According to the dataset of node change directions, analyze the order of the device state change nodes, analyze the time relationship between the node orders, identify the trend characteristics of the state evolution, and generate state evolution trend characteristics; S303: Based on the state evolution trend characteristics, predict the device state change trend according to the trend characteristics and time series of the state change, calculate the time point of the abnormal state, and generate state trend prediction information.

6. The analysis and monitoring method based on the operation status of the smart park according to claim 5, characterized in that The steps of invoking the set of operation state nodes and the state trend prediction information, obtaining the topological structure of the operation devices in the park, calculating the propagation probability of the abnormal event in multiple directions, predicting the propagation path of the abnormal event, and generating a list of path propagation nodes are specifically as follows: S401: Based on the set of operation state nodes and the state trend prediction information, obtain the topological structure of the smart park devices, including the power supply, water supply, and network structures, according to the physical distance and network connection strength between the devices, and generate a dataset of device topological structures; S402: Based on the dataset of device topological structures, extract the node positions of the devices in the abnormal state, combine the connection strength, distance, and device state between the devices, calculate the propagation probability of the abnormal event in multiple directions, and generate a dataset of event propagation probabilities; S403: Invoke the dataset of event propagation probabilities, predict the propagation path of the abnormal event, identify the propagation nodes of the abnormal event, and generate a list of path propagation nodes.

7. The analysis and monitoring method based on the operation status of the intelligent park according to claim 6, wherein, The specific formula for calculating the propagation probability of the abnormal event in multiple directions is: Calculate the propagation probability; Among them, P i′,j′ represents the abnormal propagation probability from device i' to device j', C i′,j′ represents the connection strength between device i' and device j', D i′,j′ represents the influence of directivity, S i and S j represent the state sensitivity of device i' and device j', d i′,j′ represents the physical distance between device i' and device j', where i' represents the index of the starting device and j' represents the index of the target device.

8. The analysis and monitoring method based on the operation status of the intelligent park according to claim 1, characterized in that, The method further includes: S5: Based on the list of path propagation nodes, obtain the state change information of each node, calculate the abnormal propagation time interval, analyze the propagation rate of the abnormal state on multiple paths, calculate the response priorities of multiple nodes, match the processing measures according to the device type of the target node and construct warning information, and obtain the monitoring records of the smart park; The monitoring records of the smart park include warning information, response priorities, and the propagation rate of the abnormal state.

9. The analysis and monitoring method based on the operation status of the intelligent park according to claim 8, characterized in that Based on the list of path propagation nodes, obtain the status change information of each node, calculate the abnormal propagation time interval, analyze the propagation rate of the abnormal status on multiple paths, calculate the response priorities of multiple nodes, match processing measures in combination with the device type of the target node, and construct early warning information. The steps for obtaining the intelligent park monitoring records are specifically as follows: S501: Based on the list of path propagation nodes, obtain the status change information of each node. By calculating the time interval of status changes, analyze the status change rate between device nodes, and generate a status change data set; S502: According to the status change data set, analyze the propagation rate of the abnormal status on multiple paths. By comparing the propagation duration on multiple paths and the connection strength between devices, calculate the propagation delay of the abnormal event between nodes, and generate an abnormal propagation rate data set; S503: Invoke the abnormal propagation rate data set, combine the node characteristics in the list of path propagation nodes, calculate the response priority of each node, match the type of processing measures according to the device type, construct early warning information, and generate intelligent park monitoring records; The specific formula for calculating the response priority of each node is: Calculate the response priority of each node; Among them, P z is the response priority of the z-th node, V z is the device status change rate of the z-th node, ΔT z is the abnormal propagation time interval of the z-th node, C z is the device importance coefficient of the z-th node, D z is the device health coefficient of the z-th node, and z is the index of the node.

10. An analysis and monitoring system based on the operating status of an intelligent park, characterized in that, The system is used to implement the analysis and monitoring method based on the operation status of the intelligent park according to any one of claims 1-9. The system includes: The device feature extraction module extracts the operation data parameters and associated environment parameters based on the park device monitoring data, calculates the fluctuation amplitude and fluctuation range of each parameter, analyzes the operation differences of each device relative to the same type of devices in real time, and combines the device type and location coordinates to obtain a device operation feature data set; The status coupling analysis module invokes the device operation feature data set, analyzes the operation data parameters and associated environment parameters, extracts the data change rate and fluctuation trend, analyzes the temporal correlation of various parameter change characteristics, identifies the operation status of the device in real time, and establishes an operation status node set; The trend evolution prediction module analyzes the change direction, status span and node order of each status node based on the operation status node set, predicts the status change trends of multiple devices, predicts and calculates the abnormal status time points, and generates status trend prediction information; The abnormal propagation deduction module invokes the operation status node set and the status trend prediction information, obtains the topological connection structure between each device in the park, analyzes the probability and direction of abnormal event propagation between nodes in the topological structure, predicts the propagation path of the abnormal event, and establishes a list of path propagation nodes; The intelligent response decision module, based on the list of path propagation nodes, obtains the real-time status change information of each node, calculates the time interval of the abnormal status propagation between nodes, analyzes the propagation rate of the multi-path abnormal status, calculates the response priorities of multiple nodes, matches the processing measures in combination with the node device type, constructs early warning information, and generates intelligent park monitoring records.

Images