Smart park management method and system based on multi-system integration

Abstract

The invention discloses a smart park management method and system based on multi-system integration, and relates to the technical field of smart parks, when the actual behavior of a certain subsystem deviates from a normal model, the subsystem is judged as a problem subsystem, and a basis is provided for subsequent anomaly propagation analysis. After a problem subsystem is detected, a problem event matrix is constructed and is used for depicting an abnormal occurrence relation of each subsystem. By analyzing the matrix, the prediction capability of each subsystem can be calculated, that is, whether the abnormality of a certain subsystem can predict the abnormality of other subsystems in advance is judged. An identification value is calculated through a statistical method, and if the identification value is significantly higher than a threshold value, it is indicated that the subsystem may be a key link of abnormal propagation. And based on the calculated identification value, the system further identifies the abnormal propagation path so as to construct a data-driven optimization model. The model adjusts the control parameters of the subsystems through reinforcement learning to minimize the overall abnormal influence.

Description

Technical Field

[0001] The present invention relates to the technical field of smart park technologies, and in particular to a smart park management method and system based on multi-system integration. Background Art

[0002] With the development of smart cities and digital park management, multi-system integrated smart park management has gradually become an important trend in park operations. Smart park management involves multiple subsystems such as energy management, security monitoring, environmental monitoring, facility operation and maintenance, smart office, and parking management. These subsystems operate independently and have complex interactions with each other. In traditional park management, each subsystem is often managed in a decentralized manner, lacking systematic global collaboration and data sharing, which limits the level of intelligent park management. When an abnormality occurs in a subsystem, due to the data barriers and management mode separation between the systems, the operating status of other related subsystems may not be analyzed in real time, affecting the efficiency and stability of overall management.

[0003] In the current smart park management system, due to the lack of intelligent analysis capabilities for multi-system integration, the abnormal management of each subsystem relies on single-point monitoring and isolated analysis. Once an abnormality occurs in a subsystem, its impact may gradually spread to other subsystems, but existing methods often find it difficult to accurately identify abnormal subsystems, which leads to lags in abnormal handling and local optimization problems. Summary of the invention

[0004] In view of the above-mentioned problems existing in the prior art, the present application provides a smart park management method and system based on multi-system integration.

[0005] On the one hand, the embodiments of the present disclosure provide a smart park management method and system based on multi-system integration, including the following steps:

[0006] S1: Use sensor readings to collect real-time operating data of each subsystem in the park. After data preprocessing, construct the feature vector X of each subsystem, and combine it with the historical data set to build a normal behavior model , to identify the problem subsystem;

[0007] S2: Construct the problem event matrix E, and analyze the foresight capabilities between the subsystems based on the problem event matrix E, and obtain the identification value ZS after verification;

[0008] S3: Based on the identification value ZS, identify the abnormal propagation path and build a data-driven optimization model to minimize the impact of system abnormalities.

[0009] Optionally, the specific steps of S1 include:

[0010] S11: The subsystems in the park include energy management subsystem, safety monitoring and security subsystem, environmental monitoring subsystem, vehicle and smart parking management subsystem, facility operation and equipment management subsystem, smart office and personnel management subsystem, and data analysis and AI decision support subsystem. The operating data of each subsystem in the park are cleaned, outliers are eliminated, and data is standardized to generate the characteristic vector X of each subsystem. The expression of the characteristic vector X of each subsystem is: , n is the number of subsystems, They are respectively the eigenvector of the first subsystem, the eigenvector of the second subsystem, ..., the eigenvector of the nth subsystem.

[0011] Optionally, the specific steps of S1 also include:

[0012] S12: Based on the real-time acquisition of the operating data of each subsystem in the park by S11, the operating data of each subsystem in the past day is stored in the NoSQL database, and the data in the NoSQL database is used as the historical data group;

[0013] S13: extracting historical data of each subsystem in normal operation from the historical data group, and generating a normal feature vector Y. The expression of the normal feature vector Y is: , N is the number of characteristic components, i is the subsystem number, t is the historical moment number, are respectively the first characteristic component in the normal characteristic vector of the ith subsystem at the tth historical moment, the second characteristic component in the normal characteristic vector of the ith subsystem at the tth historical moment, ..., the Nth characteristic component in the normal characteristic vector of the ith subsystem at the tth historical moment;

[0014] S14: Use machine learning techniques and train the model using the normal feature vector Y to build a normal behavior model , specifically: ;

[0015] In the formula, is the output value of the normal behavior model constructed according to the normal feature vector of the i-th subsystem, is the intercept term, , ,..., is the regression coefficient, , ,..., are the eigencomponents in the normal eigenvector of the i-th subsystem.

[0016] Optionally, the specific steps of S1 also include:

[0017] S15: Based on historical data sets and normal behavior models , calculate the residual , specifically: ;in, is the actual value, and the abnormal critical value V is pre-set, and the residual Compare with the abnormal critical value V to identify the problem subsystem, specifically: if the residual If the absolute value of is greater than the abnormal critical value V, the corresponding subsystem is judged to be abnormal, and the corresponding subsystem is marked as a problem subsystem.

[0018] Optionally, the specific steps of S2 include:

[0019] S21: Based on the problem subsystem determined in S15, the problem subsystems corresponding to several groups of historical moments are obtained, and according to the problem subsystems corresponding to several groups of historical moments, the problem event matrix E of several groups of historical moments is constructed, and each row and each column of the problem event matrix E represents a subsystem; the problem event matrix E is filled by the problem subsystems corresponding to several groups of historical moments, specifically: if the subsystems are not marked as problem subsystems at the same time, the value of the corresponding position in the problem event matrix E is 1, otherwise, the value is 0;

[0020] S22: Extract features from the positions corresponding to the values ​​of 1 in the problem event matrix E to obtain several groups of prediction pairs to be analyzed.

[0021] Optionally, the specific steps of S2 also include:

[0022] S23: Mark the problem subsystems in several groups of foreseen pairs to be analyzed as subsystems to be analyzed, and obtain the normal behavior model through S14 By using the method of , and combined with the predicted pair to be analyzed, the two groups of subsystems to be analyzed involved in the predicted pair to be analyzed are analyzed through the corresponding normal behavior model , after expansion, obtain the expanded model , specifically: ;

[0023] In the formula, To contain only Autoregressive model with own lag; ;

[0024] In the formula, is the output value of the extended model constructed based on the normal feature vector of the ith subsystem at the tth historical moment, is the intercept term, , , and are regression coefficients, ,..., are the output values ​​of the normal behavior model constructed according to the normal feature vectors of the ith subsystem at the t-1th historical moment, ..., tqth historical moment respectively; is the error term, q is The lag order, p, is The lag order of ,..., are respectively the output values ​​of the normal behavior model constructed according to the normal feature vectors of the h-th subsystem at the t-1-th historical moment, ..., tp-th historical moment; h and i are the numbers of the subsystems.

[0025] Optionally, the specific steps of S2 also include:

[0026] S24: By comparing only Autoregressive Model with Its Own Lagged Term and the output value of the expanded model constructed based on the normal feature vector of the i-th subsystem at the t-th historical moment The residual sum of squares is used to determine the addition Does the lag term improve the The predictive power is: ;

[0027] Where ZS is the identification value, Indicates that only Autoregressive Model with Its Own Lagged Term The residual sum of squares, Represents the output value of the expanded model constructed based on the normal feature vector of the ith subsystem at the tth historical moment The residual sum of squares, T represents the number of samples;

[0028] S25: preset the recognition threshold, and compare it with the recognition value ZS. If the recognition value ZS> the recognition threshold, it means adding The lag term will increase the The prediction ability of the feature set is obtained by marking the corresponding predicted pairs to be analyzed as predicted pairs and extracting them to obtain the feature set.

[0029] Optional, S3: The specific steps include:

[0030] S31: Based on the extraction set, a set of problem event matrices E is re-established and filled, where each row and each column of the re-established problem event matrix E represents a subsystem to be analyzed, and the specific content of the filling is: each prediction pair is found in the re-established problem event matrix E, and the position corresponding to each prediction pair is marked as a value 1, and the values ​​of the remaining positions are 0;

[0031] S32: Based on the content of S31, taking the subsystem to be analyzed as the node, if the position in the re-established problem event matrix E , then draw from point to After traversing the directed edges, a set of directed graphs is generated; among them, and are different subsystems to be analyzed, s and c are the numbers of the subsystems to be analyzed;

[0032] S33: According to the directed graph, the abnormal propagation path of the corresponding subsystem to be analyzed is identified, and the source of the abnormality is determined. According to the abnormal propagation path of the corresponding subsystem to be analyzed, the abnormal propagation time delay of each propagation segment in the abnormal propagation path of the corresponding subsystem to be analyzed is counted, and the minimum abnormal propagation time delay of each propagation segment is extracted.

[0033] Optionally, the S3 specific steps include:

[0034] S34: Determine the minimum abnormal propagation time delay of each propagation segment. When an abnormality occurs, predict the cascading abnormality in each system to statistically After an exception occurs, within the minimum exception propagation time of the corresponding propagation segment, Probability of abnormality ;

[0035] S35: Use reinforcement learning to train optimization strategies based on historical data to build a data-driven optimization model, specifically: ;

[0036] Where L is the optimization objective function, is the decision variable, is the economic loss of the park caused by the abnormality of the jth subsystem to be analyzed, To remove the After an abnormality occurs in the subsystem to be analyzed other than The probability of anomaly; J is the number of subsystems to be analyzed; j is the number of subsystems to be analyzed.

[0037] On the other hand, the embodiment of the present disclosure provides a smart park management system based on multi-system integration, including: a primary identification module, a secondary identification module and an optimization module;

[0038] The primary recognition module uses sensor readings to collect real-time operating data of each subsystem in the park. After data preprocessing, it constructs the feature vector X of each subsystem and combines it with the historical data set to build a normal behavior model. , to identify the problem subsystem;

[0039] The secondary identification module constructs the problem event matrix E, and analyzes the foresight capabilities between each subsystem based on the problem event matrix E. After testing, the identification value ZS is obtained;

[0040] The optimization module, based on the identification value ZS, identifies the abnormal propagation path and builds a data-driven optimization model to minimize the impact of system abnormalities.

[0041] The present invention provides a smart park management method and system based on multi-system integration, which has the following beneficial effects:

[0042] (1) This method first uses a historical data set and a machine learning model to train the normal behavior model of each subsystem. When the actual behavior of a subsystem deviates from the normal model, it is determined to be a problem subsystem, providing a basis for subsequent abnormal propagation analysis. After detecting the problem subsystem, a problem event matrix is ​​constructed to characterize the relationship between the occurrence of abnormalities in each subsystem. By analyzing the matrix, the predictive ability of each subsystem can be calculated, that is, whether the abnormality of a subsystem can predict the abnormality of other subsystems in advance. The identification value is calculated by statistical methods. If it is significantly higher than the threshold, it indicates that the subsystem may be a key link in the abnormal propagation. Based on the calculated identification value, the system further identifies the abnormal propagation path and constructs a data-driven optimization model. The model adjusts the control parameters of the subsystem through reinforcement learning to minimize the overall abnormal impact. For example, in a smart park, if the power load of the energy management system is abnormal, resulting in server overload, this method can identify the energy management abnormality as the main cause, and reduce the risk of server crashes by optimizing the energy allocation strategy, thereby improving the overall stability of the park.

[0043] (2) Construct the problem event matrix E through the abnormal records in the historical data group. The rows and columns of the matrix represent different subsystems and are filled with historical abnormal information. If the two subsystems are not abnormal at the same time, the matrix value is set to 1, indicating that there may be abnormal correlation; otherwise, it is set to 0, indicating that the corresponding subsystems may be independent abnormalities. Matrix analysis can discover the potential correlation between subsystems and avoid misjudgment of single-point abnormalities. Through matrix analysis, subsystem pairs with high abnormal correlation are extracted to form foresight pairs to be analyzed, and the foresight ability between systems is further analyzed to predict the abnormal propagation path.

[0044] (3) Through step S23, the normal behavior model of the subsystem to be analyzed at different historical moments is calculated, and an extended model is constructed. Compared with the traditional autoregressive model based only on a single subsystem, this method can more comprehensively reflect the dynamic influence relationship between systems by introducing a cross-subsystem lag term, thereby improving the accuracy of anomaly analysis. In the extended model, not only the historical characteristics of the subsystem itself are considered, but also the historical data of other subsystems are introduced as input variables, thereby establishing a cross-system anomaly prediction model.

[0045] (4) By reconstructing the problem event matrix E, the system can reanalyze the abnormal correlation between each subsystem to dynamically identify the linkage relationship between systems, thereby predicting the diffusion trend of the abnormality, and drawing a directed graph of abnormal propagation by foreseeing the impact relationship between the marked subsystems. In the directed graph, it is possible to clearly identify which subsystems are the "driving factors" of the abnormality (i.e., the source of abnormal propagation) and which are the "response factors" of the abnormality (the affected subsystems), thereby more accurately locking the root cause of the system abnormality. By calculating the abnormal propagation time delay, the present invention can analyze the shortest time required for an abnormality to propagate from one subsystem to another, and use this information to predict the probability of cascading abnormalities. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application.

[0047] Figure 1 A schematic diagram of a process flow of a smart park management method based on multi-system integration according to the present invention;

[0048] Figure 2 This is a block diagram of a smart park management system based on multi-system integration of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical solution and advantages of this application clearer, the technical solution of this application will be described clearly and completely in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, but not all of them. Based on the embodiments in this application, ordinary technicians in this field can make the following inventions without creative work.

[0050] All other embodiments obtained under the premise of the above-mentioned embodiment belong to the scope of protection of this application.

[0051] Example 1

[0052] See also Figure 1The present invention provides a smart park management method based on multi-system integration, comprising the following steps:

[0053] S1: Use sensor readings to collect real-time operating data of each subsystem in the park. After data preprocessing, construct the feature vector X of each subsystem, and combine it with the historical data set to build a normal behavior model , to identify the problem subsystem;

[0054] S2: Construct the problem event matrix E, and analyze the foresight capabilities between the subsystems based on the problem event matrix E, and obtain the identification value ZS after verification;

[0055] S3: Based on the identification value ZS, identify the abnormal propagation path and build a data-driven optimization model to minimize the impact of system abnormalities.

[0056] In the embodiment of the present method, the smart park management method based on multi-system integration provided by the present invention can realize real-time abnormal monitoring, abnormal correlation analysis and intelligent optimization decision-making of subsystems in the park by constructing feature vectors, problem event matrices and data-driven optimization models, thereby improving the intelligent level of park management. Compared with the traditional independent abnormal detection method, the present method has the following significant advantages: through the S1 step, the system collects subsystem operation data in real time, combines historical data groups, and constructs a normal behavior model. When the behavior of a subsystem deviates from the normal model, the abnormality can be quickly detected, which avoids the false alarm and missed alarm problems caused by the traditional monitoring method based on fixed thresholds, and improves the accuracy of abnormal identification. Through the S2 step, the system constructs a problem event matrix to analyze the abnormal correlation between each subsystem. Traditional methods often only find the abnormality of a single subsystem, while the present method can reveal the causal relationship between different subsystems through matrix calculation and identification values, determine which abnormalities have predictive ability, and provide data support for further optimization. In the S3 step, the present method calculates the abnormal propagation path based on the identification value, constructs a data-driven optimization model, and dynamically adjusts the system parameters to minimize the impact of the abnormality. For example, in a smart park, if the power load anomaly of the energy management subsystem causes a server overload in the equipment management subsystem, the traditional method may only handle the server anomaly separately and fail to identify the real source of the problem. The method of the present invention can accurately identify the energy management system as the source of the anomaly, and adjust the power supply plan through optimization strategies to avoid server failures in advance and improve the overall stability of the park. In summary, the present invention can not only improve the accuracy of anomaly detection, but also optimize the linkage management between subsystems, reduce the impact of park anomalies, and improve operation and maintenance efficiency.

[0057] Example 2

[0058] Please refer to Figure 1 , specifically: S1 specific steps include:

[0059] S11: The subsystems in the park include energy management subsystem, safety monitoring and security subsystem, environmental monitoring subsystem, vehicle and smart parking management subsystem, facility operation and equipment management subsystem, smart office and personnel management subsystem, and data analysis and AI decision support subsystem. The operating data of each subsystem in the park are cleaned, outliers are eliminated, and data is standardized to generate the characteristic vector X of each subsystem. The expression of the characteristic vector X of each subsystem is: , n is the number of subsystems, They are respectively the eigenvector of the first subsystem, the eigenvector of the second subsystem, ..., the eigenvector of the nth subsystem.

[0060] The specific steps of S1 also include:

[0061] S12: Based on the real-time acquisition of the operating data of each subsystem in the park by S11, the operating data of each subsystem in the past day is stored in the NoSQL database, and the data in the NoSQL database is used as the historical data group;

[0062] S12's data storage and management ensures that the operating data of each subsystem in the park can be stored for a long time and used for subsequent analysis; since the amount of data generated by the park's multiple subsystems is large and the data format is complex (such as time series data, log data, sensor data, etc.), NoSQL databases (such as MongoDB and HBase) are selected for storage, which can provide efficient data reading and expansion capabilities. The stored data is divided into time windows (such as the past 24 hours) to form historical data groups to support data analysis and model training.

[0063] S13: extracting historical data of each subsystem in normal operation from the historical data group, and generating a normal feature vector Y. The expression of the normal feature vector Y is: , N is the number of characteristic components, i is the subsystem number, t is the historical moment number, are respectively the first characteristic component in the normal characteristic vector of the ith subsystem at the tth historical moment, the second characteristic component in the normal characteristic vector of the ith subsystem at the tth historical moment, ..., the Nth characteristic component in the normal characteristic vector of the ith subsystem at the tth historical moment;

[0064] Among them, the normal operating status data of each subsystem is extracted from the historical data to construct a feature vector to provide input for subsequent model training.

[0065] S14: Use machine learning techniques and train the model using the normal feature vector Y to build a normal behavior model , specifically: ;

[0066] In the formula, is the output value of the normal behavior model constructed according to the normal feature vector of the i-th subsystem, is the intercept term, , ,..., is the regression coefficient, which indicates the effect of each feature on the target variable ( ), , ,..., are the eigencomponents in the normal eigenvector of the i-th subsystem.

[0067] The output value of the normal behavior model constructed according to the normal feature vector of the i-th subsystem It is to prepare for the subsequent abnormal analysis, which represents the normal behavior model of the subsystem and is used to predict the target variable;

[0068] It should be noted that in the smart park management system, the dependent variable It is the target variable we want to predict, usually representing the key performance indicator (KPI) of a subsystem, which is used to measure whether the operating status of the subsystem is normal.

[0069] Specifically, the main purpose of this section is to build a normal behavior model based on historical data to support subsequent anomaly detection and analysis. Specifically, this section covers three core steps: data storage, feature extraction, and model training. The purpose is to establish an accurate normal operation baseline to achieve intelligent anomaly monitoring. When the actual operating state of the subsystem deviates from the normal model, the anomaly can be quickly identified, accurate early warning can be achieved, and the intelligent level of park management can be improved.

[0070] Example 3

[0071] Please refer to Figure 1 Specifically: S1 includes the following steps:

[0072] S15: Based on historical data sets and normal behavior models , calculate the residual , specifically: ;in, is the actual value, the abnormal critical value V is set in advance, and the residual Compare with the abnormal critical value V to identify the problem subsystem, specifically: if the residual If the absolute value of is greater than the abnormal critical value V, the corresponding subsystem is judged to be abnormal, and the corresponding subsystem is marked as a problem subsystem.

[0073] By establishing a normal behavior model and monitoring the residuals, the smart park management system can effectively detect abnormal behaviors of subsystems and ensure the stable operation of the system.

[0074] Specifically, the Grubbs test method can be used to set the threshold;

[0075] The specific steps of S2 include:

[0076] S21: Based on the problem subsystem determined in S15, the problem subsystems corresponding to several groups of historical moments are obtained, and according to the problem subsystems corresponding to several groups of historical moments, the problem event matrix E of several groups of historical moments is constructed, and each row and each column of the problem event matrix E represents a subsystem; the problem event matrix E is filled by the problem subsystems corresponding to several groups of historical moments, specifically: if the subsystems are not marked as problem subsystems at the same time, the value of the corresponding position in the problem event matrix E is 1, otherwise, the value is 0;

[0077] S22: Extract features from the positions corresponding to the values ​​of 1 in the problem event matrix E to obtain several groups of prediction pairs to be analyzed.

[0078] The abnormal event matrix E is used to characterize the simultaneous occurrence of abnormal events between subsystems;

[0079] Specifically, this method can effectively identify isolated anomalies and system-level anomalies through residual analysis + problem event matrix, and improve the accuracy of anomaly detection. For example, in a smart park, if the air-conditioning system has a temperature anomaly (S15 identifies the problem subsystem), and the server cooling system is also found to be abnormal at multiple historical moments (the matrix constructed by S21), the system can automatically identify that the air-conditioning failure may affect the server cooling (S22 predictive extraction). Traditional methods may only alarm when the server temperature exceeds the limit, but this method can detect related anomalies in advance, adjust the air-conditioning load in advance, avoid server downtime due to high temperature, and improve the level of intelligent operation and maintenance of the park. In summary, this method can accurately analyze the internal relationship of subsystem anomalies, optimize anomaly management, and improve system stability and operation and maintenance efficiency.

[0080] Example 4

[0081] Please refer to Figure 1 Specifically: S2 includes the following specific steps:

[0082] S23: Mark the problem subsystems in several groups of foreseen pairs to be analyzed as subsystems to be analyzed, and obtain the normal behavior model through S14 By using the method of , and combined with the predicted pair to be analyzed, the two groups of subsystems to be analyzed involved in the predicted pair to be analyzed are analyzed through the corresponding normal behavior model , after expansion, obtain the expanded model , specifically: ;

[0083] In the formula, To contain only Autoregressive model with own lag; ;

[0084] In the formula, is the output value of the extended model constructed based on the normal feature vector of the ith subsystem at the tth historical moment, is the intercept term, indicating the intercept of the model;

[0085] , , and are regression coefficients, representing The influence of the first lag term on the current value, The influence of the qth lag term on the current value, The influence of the first lag term on the current value and The influence of the pth lag term on the current value; ,..., are the output values ​​of the normal behavior model constructed according to the normal feature vectors of the ith subsystem at the t-1th historical moment, ..., tqth historical moment respectively;

[0086] is the error term, q is The lag order, p, is The lag order of ,..., are the output values ​​of the normal behavior model constructed according to the normal feature vectors of the h-th subsystem at the t-1th historical moment, ..., and the tpth historical moment; h and i are the numbers of the subsystems. For the determination of the lag order, the Akaike Information Criterion (AIC) or the Bayesian Information Criterion (BIC) can be used to select the optimal lag order.

[0087] Specifically, traditional anomaly detection methods often only focus on the historical state of a single subsystem, making it difficult to identify the potential impact relationship between subsystems. This method first marks the problem subsystem based on the predicted pair to be analyzed, and uses historical data to train the normal behavior model to build a prediction model based on the autoregressive model (AR). Then, by introducing multiple lag orders, the model not only relies on the historical data of the subsystem itself, but also considers the state changes of other subsystems, thereby optimizing the accuracy of anomaly detection. By extending the behavior model, the normal behavior models of the two subsystems in the predicted pair to be analyzed are combined to establish a set of equations containing mutual influence relationships. For example, if the anomaly of the energy management subsystem may affect the facility operation and maintenance subsystem, the extended model will simultaneously consider the impact of the lag state of the energy management subsystem on the facility operation and maintenance subsystem, thereby improving the ability to identify the anomaly propagation path.

[0088] Example 5

[0089] Please refer to Figure 1 Specifically: S2 includes the following specific steps:

[0090] S24: By comparing only Autoregressive Model with Its Own Lagged Term and the output value of the expanded model constructed based on the normal feature vector of the i-th subsystem at the t-th historical moment The residual sum of squares is used to determine the addition Does the lag term improve the The predictive power is: ;

[0091] Where ZS is the identification value, Indicates that only Autoregressive Model with Its Own Lagged Term The residual sum of squares (i.e., excluding ), Represents the output value of the expanded model constructed based on the normal feature vector of the ith subsystem at the tth historical moment The residual sum of squares (i.e., lag term), T represents the sample size; The molecular part represents the introduction After that, the degree of reduction of error is Larger, indicating the addition After that, the error is reduced significantly, indicating Maybe Play a predictive effect; The denominator part is used to standardize the numerator part so that the identification value ZS follows the F distribution;

[0092] By obtaining the identification value ZS, the potential correlation and causal relationship between abnormal events can be effectively identified, thereby revealing deep-seated problems.

[0093] Among them, the F distribution is a probability distribution used for hypothesis testing, which is often used to compare the goodness of fit of two variances or models;

[0094] S25: preset the recognition threshold, and compare it with the recognition value ZS. If the recognition value ZS> the recognition threshold, it means adding The lag term will increase the The prediction ability of , and mark the corresponding prediction pairs to be analyzed as prediction pairs, and extract them to obtain the feature set, anyway, it cannot;

[0095] Specifically, the method first constructs an autoregressive model containing only its own lag term, and then expands it to a model containing lag terms of other subsystems, and calculates the residual sum of squares respectively. If the error of the extended model is significantly reduced, it means that the introduction of lag terms of other subsystems can improve the prediction effect, indicating that there may be abnormal correlation between the two subsystems, thereby improving the accuracy of anomaly detection. By calculating the recognition value and performing the F distribution test, it is determined whether the newly added lag term has really improved the prediction ability. This statistical test method can effectively filter out irrelevant variables, retain only the lag terms of the subsystem that contribute to the abnormal prediction, avoid the increase in model complexity or misjudgment caused by too many parameters, and improve the stability of the model. If the recognition value exceeds the set threshold, the system marks the subsystem pair as a predicted pair and extracts its features for in-depth analysis. For example, in a smart park, there may be an implicit relationship between the voltage fluctuation of the energy management subsystem and the abnormal heat dissipation of the server. The traditional method may only alarm when the server temperature is abnormal, but this method can identify the delayed effect of voltage fluctuation on server heat dissipation, adjust the power supply strategy in advance, and reduce the risk of server failure. In summary, this method can accurately identify abnormal correlations between subsystems through residual comparison, F distribution test and prediction pair extraction, improve the abnormal warning capability, thereby optimizing the management of smart parks, reducing sudden failures and improving system stability.

[0096] Example 6

[0097] Please refer to Figure 1 , specifically: S3: The specific steps include:

[0098] S31: Based on the extraction set, a set of problem event matrices E is re-established and filled, where each row and each column of the re-established problem event matrix E represents a subsystem to be analyzed, and the specific content of the filling is: each prediction pair is found in the re-established problem event matrix E, and the position corresponding to each prediction pair is marked as a value 1, and the values ​​of the remaining positions are 0;

[0099] S32: Based on the content of S31, taking the subsystem to be analyzed as the node, if the position in the re-established problem event matrix E , then draw from point to After traversing the directed edges, a set of directed graphs is generated; among them, and are different subsystems to be analyzed, s and c are the numbers of the subsystems to be analyzed;

[0100] This process can help us clarify which subsystems are "driving factors" and which are "response factors", providing guidance for subsequent decision-making.

[0101] S33: According to the directed graph, the abnormal propagation path of the corresponding subsystem to be analyzed is identified, and the source of the abnormality is determined. According to the abnormal propagation path of the corresponding subsystem to be analyzed, the abnormal propagation time delay of each propagation segment in the abnormal propagation path of the corresponding subsystem to be analyzed is counted, and the minimum abnormal propagation time delay of each propagation segment is extracted.

[0102] The specific steps of S3 include:

[0103] S34: Determine the minimum abnormal propagation time delay of each propagation segment. When an abnormality occurs, predict the cascading abnormality in each system to statistically After an exception occurs, within the minimum exception propagation time of the corresponding propagation segment, Probability of abnormality ;

[0104] S35: Use reinforcement learning (RL) to train optimization strategies based on historical data to build a data-driven optimization model, specifically: ;

[0105] Where L is the optimization objective function (the total amount of abnormal impact of the entire park), are decision variables (control strategy parameters), is the economic loss of the park caused by the abnormality of the jth subsystem to be analyzed, To remove the After an abnormality occurs in the subsystem to be analyzed other than The probability of abnormality; J is the number of subsystems to be analyzed; j is the number of subsystems to be analyzed. The purpose of this formula is to minimize the sum of the product of abnormal probability and economic loss of all subsystems to be analyzed by adjusting the control strategy θ, that is, to reduce the overall risk of the system.

[0106] The economic loss of the park caused by the abnormality of the jth subsystem to be analyzed It can be determined by business data or expert experience, such as production loss caused by equipment downtime, additional costs caused by abnormal energy consumption, etc.

[0107] The purpose of adjusting θ is to optimize the system control strategy to reduce the probability of abnormalities or mitigate the impact of abnormalities. Different subsystems have different abnormal risks. Adjusting the decision variable θ can optimize resource allocation and minimize the overall abnormal impact. Some abnormal risks are difficult to completely eliminate, but by adjusting θ, subsystems with high loss risks can be given priority protection.

[0108] Specifically, through steps S31-S33, the system reconstructs the problem event matrix E based on the extraction set, identifies the foresight pairs, and generates an abnormal propagation directed graph between subsystems. This method can reveal the abnormal transmission relationship between subsystems, and identify which subsystems are abnormal driving factors and which are response factors by traversing the directed graph. Compared with traditional independent anomaly detection, this method can analyze the abnormal propagation path more comprehensively to ensure that the abnormal management strategy is more targeted. In step S34, the system infers the probability of cascading effects in different subsystems after the abnormality occurs based on the minimum abnormal propagation time delay of each propagation segment. For example, in a smart park, if the energy management subsystem (X1) has a voltage fluctuation, it usually causes abnormal heat dissipation of the server of the equipment management subsystem (X2) after 5 minutes, and may affect the office automation system (X3) after 10 minutes. The traditional method may trigger an alarm only after the server temperature is abnormal, while this method can predict the probability of future abnormalities of X2 and X3 in advance, providing support for preventive maintenance. In step S35, the system uses reinforcement learning (RL) to dynamically adjust the subsystem control parameters θ based on historical data training optimization strategies, so as to minimize the overall abnormal impact of the park. For example, if it is detected that the energy management subsystem may cause abnormal heat dissipation of the server, the system can automatically adjust the air conditioning load or optimize the power distribution to prevent excessive temperature and avoid server downtime. In summary, this method can not only accurately identify the abnormal propagation path, but also predict cascade anomalies based on propagation time delays, and use reinforcement learning to dynamically optimize the control strategy, ultimately reducing the economic losses caused by anomalies and improving the stability, prediction ability and automation management level of the smart park.

[0109] Example 7

[0110] Please refer to Figure 2 ,Specifically: A smart park management system based on multi-system integration includes a primary recognition module, a secondary recognition module and an optimization module;

[0111] The primary recognition module uses sensor readings to collect real-time operating data of each subsystem in the park. After data preprocessing, it constructs the feature vector X of each subsystem and combines it with the historical data set to build a normal behavior model. , to identify the problem subsystem;

[0112] The secondary identification module constructs the problem event matrix E, and analyzes the foresight capabilities between each subsystem based on the problem event matrix E. After testing, the identification value ZS is obtained;

[0113] The optimization module, based on the identification value ZS, identifies the abnormal propagation path and builds a data-driven optimization model to minimize the impact of system abnormalities.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A smart park management method based on multi-system integration, characterized by: The following steps are involved: S1: Use sensor readings to collect real-time operating data of each subsystem in the park. After data preprocessing, construct the feature vector X of each subsystem, and combine it with the historical data set to build a normal behavior model , to identify the problem subsystem; S2: Construct the problem event matrix E, and analyze the foresight capabilities between the subsystems based on the problem event matrix E, and obtain the identification value ZS after verification; S3: Based on the identification value ZS, identify the abnormal propagation path and build a data-driven optimization model to minimize the impact of system abnormalities.

2. According to claim 1, a smart park management method based on multi-system integration is characterized in that: The specific steps of S1 include: S11: The subsystems in the park include energy management subsystem, safety monitoring and security subsystem, environmental monitoring subsystem, vehicle and smart parking management subsystem, facility operation and equipment management subsystem, smart office and personnel management subsystem, and data analysis and AI decision support subsystem. The operating data of each subsystem in the park are cleaned, outliers are eliminated, and data is standardized to generate the characteristic vector X of each subsystem. The expression of the characteristic vector X of each subsystem is: , n is the number of subsystems, They are respectively the eigenvector of the first subsystem, the eigenvector of the second subsystem, ..., the eigenvector of the nth subsystem.

3. According to claim 2, a smart park management method based on multi-system integration is characterized in that: The specific steps of S1 also include: S12: Based on the real-time acquisition of the operating data of each subsystem in the park by S11, the operating data of each subsystem in the past day is stored in the NoSQL database, and the data in the NoSQL database is used as the historical data group; S13: extracting historical data of each subsystem in normal operation from the historical data group, and generating a normal feature vector Y. The expression of the normal feature vector Y is: , N is the number of characteristic components, i is the subsystem number, t is the historical moment number, are respectively the first characteristic component in the normal characteristic vector of the ith subsystem at the tth historical moment, the second characteristic component in the normal characteristic vector of the ith subsystem at the tth historical moment, ..., the Nth characteristic component in the normal characteristic vector of the ith subsystem at the tth historical moment; S14: Use machine learning techniques and train the model using the normal feature vector Y to build a normal behavior model , specifically: ； In the formula, is the output value of the normal behavior model constructed according to the normal feature vector of the i-th subsystem, is the intercept term, , ,..., is the regression coefficient, , ,..., are the eigencomponents in the normal eigenvector of the i-th subsystem.

4. The smart park management method based on multi-system integration according to claim 3 is characterized in that: The specific steps of S1 also include: S15: Based on historical data sets and normal behavior models , calculate the residual , specifically: ;in, is the actual value, and the abnormal critical value V is pre-set, and the residual Compared with the abnormal critical value V, to identify the problem subsystem, specifically: if the residual If the absolute value of is greater than the abnormal critical value V, the corresponding subsystem is judged to be abnormal, and the corresponding subsystem is marked as a problem subsystem.

5. The smart park management method based on multi-system integration according to claim 4 is characterized in that: The specific steps of S2 include: S21: Based on the problem subsystem determined in S15, the problem subsystems corresponding to several groups of historical moments are obtained, and according to the problem subsystems corresponding to several groups of historical moments, the problem event matrix E of several groups of historical moments is constructed, and each row and each column of the problem event matrix E represents a subsystem; the problem event matrix E is filled by the problem subsystems corresponding to several groups of historical moments, specifically: if the subsystems are not marked as problem subsystems at the same time, the value of the corresponding position in the problem event matrix E is 1, otherwise, the value is 0; S22: Extract features from the positions corresponding to the values ​​of 1 in the problem event matrix E to obtain several groups of prediction pairs to be analyzed.

6. The smart park management method based on multi-system integration according to claim 5 is characterized by: The specific steps of S2 also include: S23: Mark the problem subsystems in several groups of foreseen pairs to be analyzed as subsystems to be analyzed, and obtain the normal behavior model through S14 By using the method of , and combined with the predicted pair to be analyzed, the two groups of subsystems to be analyzed involved in the predicted pair to be analyzed are analyzed through the corresponding normal behavior model , after expansion, obtain the expanded model , specifically: ； In the formula, To contain only Autoregressive model with own lag; ； In the formula, is the output value of the extended model constructed based on the normal feature vector of the ith subsystem at the tth historical moment, is the intercept term, , , and are regression coefficients, ,..., are the output values ​​of the normal behavior model constructed according to the normal feature vectors of the ith subsystem at the t-1th historical moment, ..., tqth historical moment respectively; is the error term, q is The lag order, p, is The lag order of ,..., are respectively the output values ​​of the normal behavior model constructed according to the normal feature vectors of the h-th subsystem at the t-1-th historical moment, ..., tp-th historical moment; h and i are the numbers of the subsystems.

7. The smart park management method based on multi-system integration according to claim 6 is characterized in that: The specific steps of S2 also include: S24: By comparing only Autoregressive Model with Its Own Lagged Term and the output value of the expanded model constructed based on the normal feature vector of the i-th subsystem at the t-th historical moment The residual sum of squares is used to determine the addition Does the lag term improve the The predictive power of ； Where ZS is the identification value, Indicates that only Autoregressive Model with Its Own Lagged Term The residual sum of squares is Represents the output value of the expanded model constructed based on the normal feature vector of the ith subsystem at the tth historical moment The residual sum of squares, T represents the number of samples; S25: preset the recognition threshold, and compare it with the recognition value ZS. If the recognition value ZS> the recognition threshold, it means adding The lag term will increase the The prediction ability of the feature set is obtained by marking the corresponding predicted pairs to be analyzed as predicted pairs and extracting them to obtain the feature set.

8. The smart park management method based on multi-system integration according to claim 7 is characterized in that: S3: The specific steps include: S31: Based on the extraction set, a set of problem event matrices E is re-established and filled, where each row and each column of the re-established problem event matrix E represents a subsystem to be analyzed, and the specific content of the filling is: each prediction pair is found in the re-established problem event matrix E, and the position corresponding to each prediction pair is marked as a value 1, and the values ​​of the remaining positions are 0; S32: Based on the content of S31, taking the subsystem to be analyzed as the node, if the position in the re-established problem event matrix E , then draw from point to After traversing the directed edges, a set of directed graphs is generated; among them, and are different subsystems to be analyzed, s and c are the numbers of the subsystems to be analyzed; S33: According to the directed graph, the abnormal propagation path of the corresponding subsystem to be analyzed is identified, and the source of the abnormality is determined. According to the abnormal propagation path of the corresponding subsystem to be analyzed, the abnormal propagation time delay of each propagation segment in the abnormal propagation path of the corresponding subsystem to be analyzed is counted, and the minimum abnormal propagation time delay of each propagation segment is extracted.

9. The smart park management method based on multi-system integration according to claim 8 is characterized in that: The specific steps of S3 include: S34: Determine the minimum abnormal propagation time delay of each propagation segment. When an abnormality occurs, predict the cascading abnormality in each system to statistically After an exception occurs, within the minimum exception propagation time of the corresponding propagation segment, Probability of abnormality ; S35: Use reinforcement learning to train optimization strategies based on historical data to build a data-driven optimization model, specifically: ； Where L is the optimization objective function, is the decision variable, is the economic loss of the park caused by the abnormality of the jth subsystem to be analyzed, To remove the After an abnormality occurs in the subsystem to be analyzed other than The probability of anomaly; J is the number of subsystems to be analyzed; j is the number of subsystems to be analyzed.

10. A smart park management system based on multi-system integration, applied to a smart park management method based on multi-system integration as claimed in any one of claims 1 to 9, characterized in that: It includes a primary recognition module, a secondary recognition module and an optimization module; The primary recognition module uses sensor readings to collect real-time operating data of each subsystem in the park. After data preprocessing, it constructs the feature vector X of each subsystem and combines it with the historical data set to build a normal behavior model. , to identify the problem subsystem; The secondary identification module constructs the problem event matrix E, and analyzes the foresight capabilities between each subsystem based on the problem event matrix E. After testing, the identification value ZS is obtained; The optimization module, based on the identification value ZS, identifies the abnormal propagation path and builds a data-driven optimization model to minimize the impact of system abnormalities.

Images