Smart city monitoring and early warning method based on digital twinning

Abstract

The invention relates to the field of smart cities, in particular to a smart city monitoring and early warning method based on digital twinning. Comprising the following steps: setting a monitoring time point, and collecting and preprocessing original data to obtain monitoring data; calculating the comprehensive state of the smart city based on the detection data; based on the comprehensive state of the smart city, calculating to obtain a state change quantity and a state change predicted quantity; and based on the state change quantity and the state change prediction quantity, obtaining an adjusted threshold value by using a self-adaptive threshold value adjustment mechanism, calculating to obtain a comprehensive monitoring state value, comparing the comprehensive monitoring state value with the adjusted threshold value, and judging whether to trigger an alarm mechanism or not. The technical problems that a traditional method depends on a static threshold value to carry out anomaly detection and cannot adapt to dynamic changes of the state of the smart city in real time, monitoring and early warning are lagged, and emergencies or rapid changes are difficult to recognize in time are solved.

Description

Technical Field

[0001] The present invention relates to the field of smart cities, and particularly to a smart city monitoring and early warning method based on digital twins. Background Art

[0002] With the continuous advancement of smart city construction, monitoring and early warning technologies based on digital twins have been widely applied in various urban management. By deploying a variety of sensors and monitoring devices in smart cities, the city can collect key data related to traffic, environment, infrastructure, etc. in real time. The key data provides important support for resource management, efficiency improvement, and emergency prediction. In recent years, with the continuous development of big data analysis, artificial intelligence, and machine learning technologies, smart city monitoring has gradually shifted from single data collection to more complex data processing and intelligent decision-making processes, enabling more accurate assessment and prediction of the operating conditions of the city, providing reliable data support for decision-makers, and improving the efficiency of urban management and the accuracy of early warning.

[0003] However, the existing smart city monitoring and early warning methods have the following technical problems: traditional methods rely on static thresholds for anomaly detection and cannot adapt to the dynamic changes of smart city states in real time, resulting in lagged early warning responses and difficulty in timely identifying emergencies or rapid changes; existing technologies ignore the comprehensive analysis of multi-dimensional data and can only reflect the situation of a certain aspect of the smart city singly, failing to fully consider the complexity and diversity of smart city operation and lacking the ability to comprehensively evaluate the overall health status of the smart city; traditional methods use simple linear models, ignoring the non-linear characteristics of the changes in the overall health status of the smart city, resulting in the inability to accurately capture the complex fluctuations and potential risks in smart city operation. Summary of the Invention

[0004] The present invention provides a smart city monitoring and early warning method based on digital twins to solve the technical problems that traditional methods rely on static thresholds for anomaly detection and cannot adapt to the dynamic changes of smart city states in real time, resulting in lagged monitoring and early warning and difficulty in timely identifying emergencies or rapid changes; existing technologies ignore the comprehensive analysis of multi-dimensional data and can only reflect the situation of a certain aspect of the smart city singly, failing to fully consider the complexity and diversity of smart city operation and lacking the ability to comprehensively evaluate the overall health status of the smart city; traditional methods use simple linear models, ignoring the non-linear characteristics of the changes in the overall health status of the smart city, resulting in the inability to accurately capture the complex fluctuations and potential risks in smart city operation.

[0005] The smart city monitoring and early warning method based on digital twins of the present invention specifically includes the following technical solutions:

[0006] The smart city monitoring and early warning method based on digital twins includes the following steps:

[0007] S1: Set the monitoring time points, collect and preprocess the original data to obtain the monitoring data; calculate the comprehensive status of the smart city based on the detection data; calculate the amount of state change and the predicted amount of state change based on the comprehensive status of the smart city;

[0008] S2: Based on the amount of state change and the predicted amount of state change, use the adaptive threshold adjustment mechanism to obtain the adjusted threshold, and at the same time calculate the comprehensive monitoring status value, and compare the comprehensive monitoring status value with the adjusted threshold to determine whether to trigger the alarm mechanism.

[0009] Preferably, the S1 specifically includes:

[0010] The calculation formula for the comprehensive status of the smart city is as follows:

[0011]

[0012] Among them, M(t) is the comprehensive status of the smart city at time t; t is the time variable; n is the number of data sources, that is, the number of sensors or monitoring devices; i and k are the index variables of the data sources; θ i (t) is the weight coefficient of the i-th data source; D i (t) is the monitoring data collected from the i-th data source at time t; D k (t) is the monitoring data collected from the k-th data source at time t; ∈ is the smoothing constant.

[0013] Preferably, the S1 specifically includes:

[0014] Based on the comprehensive status of the smart city, introduce an exponential response term, and combine the response sensitivity coefficient and the exponential response coefficient to obtain the amount of state change, which measures the comprehensive state fluctuation of the smart city between different time points. The specific implementation formula is:

[0015] ΔM 1 (t) = M(t) - M(t - 1) + α·(e β·(M(t)-M(t-1)) -1)

[0016] Among them, ΔM 1 (t) is the amount of state change at time t; t is the time variable; M(t) is the comprehensive status of the smart city at time t; M(t - 1) is the comprehensive status of the smart city at time t - 1; α is the response sensitivity coefficient; e β·(M(t)-M(t-1)) is the exponential response term; β is the exponential response coefficient.

[0017] Preferably, the S1 specifically includes:

[0018] The predicted state change amount is the amount of predicted state change at a future time based on the current state change amount and the comprehensive state change rate of the smart city, quantifying the future change trend of the smart city. The specific implementation formula is as follows:

[0019]

[0020] Among them, ΔM 2 (t, t + Δt) is the predicted state change amount; Δt is the time interval, which is the time span of the predicted state change amount; t is the time variable; τ is the virtual time variable in the integration process; is the comprehensive state change rate of the smart city at time τ; M(τ) is the comprehensive state of the smart city at time τ; γ is the attenuation coefficient.

[0021] Preferably, S2 specifically includes:

[0022] The adaptive threshold adjustment mechanism dynamically adjusts the threshold by performing weighted and non-linear adjustment on the state change amount and the predicted state change amount, responding to sudden changes and abnormal situations, and obtaining the adjusted threshold.

[0023] Preferably, S2 specifically includes:

[0024] After the adaptive threshold adjustment mechanism ends, multi-dimensional data stream fusion processing is performed on the monitoring data, the state change amount, and the predicted state change amount, and the comprehensive monitoring state value is calculated.

[0025] Preferably, S2 specifically includes:

[0026] The implementation formula for multi-dimensional data stream fusion is as follows:

[0027]

[0028] Among them, F(t) is the comprehensive monitoring state value of the smart city at time t; t is the time variable; n is the number of data sources; i is the index variable of the data source; w i is the influence degree of the i-th data source on the comprehensive monitoring state of the smart city; D i (t) is the monitoring data collected from the i-th data source at time t; is the time decay term; λ is the time decay coefficient; t i is the timestamp of the monitoring data collected from the i-th data source at time t; ΔM 1 (t) is the state change amount at time t; ζ is the smoothing parameter; tanh is the hyperbolic tangent function; ΔM 2 (t, t + Δt) is the predicted state change amount; is the weighted non-linear adjustment term.

[0029] Preferably, the S2 specifically includes:

[0030] Compare the calculated comprehensive monitoring status value with the adjusted threshold. When the comprehensive monitoring status value is greater than or equal to the adjusted threshold, it indicates that there are abnormalities and potential risks in the smart city; when the comprehensive monitoring status value is lower than the adjusted threshold, it indicates that the operation status of the smart city is normal.

[0031] The beneficial effects of the technical solution of the present invention are:

[0032] 1. By combining the comprehensive status, status change amount, and status change prediction amount of the smart city, the present invention proposes a digital-twin-based smart city monitoring and early warning method, realizing the comprehensive quantification and dynamic tracking of the comprehensive status of the smart city; the status change amount can measure the fluctuations in the operation status of the smart city in real time and identify abnormalities, while the status change prediction amount predicts future trends based on the current change rate, capable of accurately capturing the dynamic changes in the operation of the smart city and avoiding the lag in traditional methods.

[0033] 2. Through the adaptive threshold adjustment mechanism, the present invention overcomes the limitations of static threshold setting in traditional smart city monitoring and early warning. The adaptive threshold adjustment mechanism dynamically adjusts the threshold according to the changes in the comprehensive status and status change prediction amount of the smart city, quickly responding to emergencies and abnormal fluctuations in the operation of the smart city, being able to more precisely adapt to the changes in the smart city status, and improving the accuracy and response efficiency of early warning.

[0034] 3. The present invention performs multi-dimensional data stream fusion processing on the monitoring data, status change amount, and status change prediction amount from different data sources, calculates the comprehensive monitoring status value of the smart city, and through weighting and non-linear adjustment, improves the early warning ability for potential risks, can more accurately reflect the overall status of the smart city, enhances the perception and analysis ability for complex environments, and ensures timely and accurate risk identification and emergency response. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is a flowchart of the digital-twin-based smart city monitoring and early warning method described in the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0036] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0037] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which this invention belongs.

[0038] The following specifically describes the specific solution of the smart city monitoring and early warning method based on digital twin provided by the present invention in conjunction with the accompanying drawings.

[0039] Refer to the attached Figure 1 , which shows the flowchart of the smart city monitoring and early warning method based on digital twin provided by an embodiment of the present invention. The method includes the following steps:

[0040] S1: Set the monitoring time point, collect and preprocess the original data to obtain the monitoring data; based on the detection data, calculate the comprehensive state of the smart city; based on the comprehensive state of the smart city, calculate the state change amount and the predicted state change amount;

[0041] According to the specific implementation scenario, set the monitoring time point, and regularly deploy sensors and monitoring devices in the smart city to collect the original data. The sensors and monitoring devices serve as data sources, specifically including traffic monitoring sensors, environmental quality sensors, infrastructure monitoring, etc. The original data is used to reflect the latest information of various situations in the smart city, and has immediacy and timeliness. For example, various environmental data and facility monitoring data such as traffic flow, vehicle speed, air quality index, temperature and humidity, noise level, energy consumption, and infrastructure operation status. Preprocess the original data such as noise filtering, outlier detection, and data calibration to obtain the monitoring data to ensure the accuracy and stability of the data, and use the monitoring data as the basis for subsequent calculations and analyses. The existing technologies well-known to those skilled in the art are adopted in the preprocessing process, and will not be elaborated here.

[0042] Furthermore, based on the monitoring data, calculate the comprehensive state of the smart city. The comprehensive state of the smart city is a comprehensive index calculated by calculating the monitoring data from different data sources, and is used to quantify and reflect the operation efficiency, resource utilization status, and potential risks of the smart city at a certain moment, so as to provide data support and decision-making basis for the evaluation of the overall health status of the smart city.

[0043] The calculation formula for the comprehensive state of the smart city is as follows:

[0044]

[0045] Among them, M(t) is the comprehensive state of the smart city at time t; t is the time variable; n is the number of data sources, that is, the number of sensors or monitoring devices, and the specific number depends on the scale of the monitoring object and the application scenario; i and k are the index variables of the data sources; θ i(t) is the weight coefficient of the i-th data source, representing the influence degree of the i-th data source on the comprehensive state of the smart city at time t. The specific value of the weight coefficient is set by using the expert experience method according to the quality, stability, and real-time factors of the data source in the specific implementation scenario; D i (t) is the monitoring data collected from the i-th data source at time t; D k (t) is the monitoring data collected from the k-th data source at time t; ∈ is the smoothing constant, which is used to avoid calculation errors caused by the value being zero in the logarithmic function, ensure the stability of the calculation process, and will not have a significant impact on the comprehensive state of the smart city, and is set according to the specific implementation scenario.

[0046] Furthermore, based on the comprehensive state of the smart city, the state change amount and the state change prediction amount are calculated to quantify the dynamic changes in the operation of the smart city. The state change amount is used to measure the comprehensive state fluctuation of the smart city between different time points, help identify anomalies or sudden changes in the operation process of the smart city, and provide a basis for dynamic adjustment and response for monitoring and early warning. The state change prediction amount is the amount of state change predicted for the future moment based on the current moment's state change amount and the comprehensive state change rate of the smart city, and is used to quantify the possible future change trend of the smart city. The comprehensive state change rate of the smart city refers to the change rate of the comprehensive state of the smart city relative to time, which is obtained by taking the derivative of the comprehensive state of the smart city with respect to time, and reflects the change speed of the comprehensive state of the smart city over time.

[0047] The calculation formula for the state change amount is as follows:

[0048] ΔM 1 (t) = M(t) - M(t - 1) + α·(e β·(M(t)-M(t-1)) -1)

[0049] where, ΔM 1 (t) is the state change amount at time t, which is used to measure the change speed of the comprehensive state of the smart city; t is the time variable; M(t) is the comprehensive state of the smart city at time t; M(t - 1) is the comprehensive state of the smart city at time t - 1; α is the response sensitivity coefficient, which is used to adjust the response sensitivity of the state change, and the value range is [0, 10], and is set according to the specific implementation scenario; e β·(M(t)-M(t-1)) is the exponential response term, which is used to capture the non-linear characteristics of the comprehensive state change of the smart city; β is the exponential response coefficient, which determines the sensitivity of the exponential function to the state change, and the value range is [0.1, 5], and is set by using the expert experience method according to the change rate and sensitivity requirements of the comprehensive state of the smart city.

[0050] The calculation formula for the state change prediction amount is as follows:

[0051]

[0052] where, ΔM 2 (t, t + Δt) is the predicted state change amount, representing the state change amount within the time interval [t, t + Δt], which is used to describe the comprehensive state change of the smart city in the future time period and provides prediction information for the monitoring and early warning of the smart city; Δt is the time interval, which is the time span of the predicted state change amount and determines the length of the integration interval, and is set according to specific implementation scenarios; t is the time variable; τ is the virtual time variable in the integration process, which is used to calculate the state change rate of the comprehensive state of the smart city at each moment. is the comprehensive state change rate of the smart city at the moment τ, indicating the state change speed at a certain moment; M(τ) is the comprehensive state of the smart city at the moment τ; γ is the attenuation coefficient, which is used to control the weighted influence of the state prediction at future moments. As τ - t increases, the influence of the state change at future moments on the current predicted state change amount will gradually decrease. A larger γ value means that the influence of the state change at future moments on the current predicted state change amount is further reduced, and it is set according to specific implementation scenarios.

[0053] S2: Based on the state change amount and the predicted state change amount, use the adaptive threshold adjustment mechanism to obtain the adjusted threshold, and at the same time calculate the comprehensive monitoring state value, and compare the comprehensive monitoring state value with the adjusted threshold to determine whether to trigger the alarm mechanism.

[0054] The state change amount reflects the fluctuation of the comprehensive state of the smart city, and the predicted state change amount reflects the possible change trend of the smart city in the future time period. Based on the state change amount and the predicted state change amount, an adaptive threshold adjustment mechanism is introduced to respond to the dynamic changes in the smart city. The adaptive threshold adjustment mechanism is applicable to various monitoring and early warning tasks of the smart city, and can dynamically adjust the threshold according to the real-time fluctuation of the comprehensive state of the smart city and the predicted state change amount, so as to quickly respond to potential abnormal situations or emergencies in the smart city and provide more accurate early warning capabilities.

[0055] The adaptive threshold adjustment mechanism uses the adaptive threshold adjustment formula to dynamically adjust the threshold by weighting and non-linearly adjusting the state change amount and the predicted state change amount, and timely responds to sudden changes or abnormal situations, avoiding the lag response caused by static threshold settings.

[0056] The adaptive threshold adjustment formula is as follows:

[0057]

[0058] where, T adj(t) is the adjusted threshold, representing the adaptive threshold corresponding to time t, which is used for the status monitoring and early warning control of the smart city; t is the time variable; T 0 is the basic threshold level of the smart city without dynamic adjustment, which is set by using the expert experience method and serves as the initial value for the adaptive threshold adjustment; ΔM 1 (t) is the amount of state change at time t, which is used to measure the change speed of the comprehensive state of the smart city; is the exponential decay function, which is used to non-linearly weight the amount of state change at time t; is the exponential decay coefficient, which is used to control the influence degree of the amount of state change on the exponential decay function, and is set according to the implementation scenario requirements by using the expert experience method; ΔM 2 (t, t + Δt) is the predicted amount of state change, representing the amount of state change within the time interval [t, t + Δt]; is the adjustment factor of the predicted amount of state change, which is used to adjust the influence of the predicted amount of state change on the threshold adjustment, and is set according to the specific implementation scenario by using the expert experience method; is the weighting coefficient of the predicted amount of state change, which is used to adjust the influence of the predicted amount of state change on the threshold adjustment, and is set according to the specific implementation scenario by using the expert experience method; is the smoothing weighting term, which is used to adjust the influence of the predicted amount of state change on the adaptive threshold, and enhance the response ability of the adjusted threshold to future state changes; is the state change weighting term, which adjusts the threshold according to the magnitude of the current amount of state change, and responds to the fluctuations of the comprehensive state of the smart city with larger changes.

[0059] After the adaptive threshold adjustment mechanism ends, the monitoring data, the amount of state change, and the predicted amount of state change are processed by multi-dimensional data stream fusion to calculate the comprehensive monitoring state value. The multi-dimensional data stream fusion comprehensively considers the monitoring data, the amount of state change, and the predicted amount of state change from different data sources to provide a comprehensive evaluation of the overall operation state of the smart city, which can more accurately capture potential abnormal changes and improve the accuracy and reliability of early warning. The comprehensive monitoring state value is a comprehensive index calculated by multi-dimensional data stream fusion, which reflects the overall operation status of the smart city at a certain moment and provides data support for early warning and decision-making.

[0060] The implementation formula of multi-dimensional data stream fusion is as follows:

[0061]

[0062] Among them, F(t) is the comprehensive monitoring state value of the smart city at time t, which reflects the overall operation state of the smart city; t is the time variable; n is the number of data sources; i is the index variable of the data source; w iis the influence degree of the i-th data source on the comprehensive monitoring status of the smart city, which is set according to specific implementation scenarios using the expert experience method; D i D(i)(t) is the monitoring data collected from the i-th data source at time t; is the time decay term, indicating the rate at which the influence of the i-th data source decays over time; λ is the time decay coefficient, used to control the rate of influence decay of the i-th data source. A larger time decay coefficient means that the influence of the data source decays faster, while a smaller one means that the influence of the data source lasts longer, which is set according to the expert experience method; t i is the timestamp of the monitoring data collected from the i-th data source at time t, and it is the monitoring data D(i) of the i-th data source i (t) the actual recording time; ΔM 1 ΔM(t) is the state change amount at time t, used to measure the change speed of the comprehensive state of the smart city; ζ is the smoothing parameter, used to limit the growth of the logarithmic term, which is set using the expert experience method; tanh is the hyperbolic tangent function; ΔM 2 ΔM(t, t + Δt) is the predicted state change amount, indicating the state change amount within the time interval [t, t + Δt]; is the weighted non-linear adjustment term, which combines the state change amount and the predicted state change amount, and is used to describe the dynamic fluctuations and trends of the overall operation state of the smart city.

[0063] Compare the calculated comprehensive monitoring status value with the adjusted threshold. If the comprehensive monitoring status value exceeds or is equal to the adjusted threshold, it indicates that the smart city may have anomalies or potential risks, and record the relevant data, including the monitoring data collected from each data source and the timestamps of the monitoring data, the comprehensive state of the smart city, the state change amount, the predicted state change amount, the adjusted threshold, and the comprehensive monitoring status value, for subsequent analysis and response decision-making, and trigger the early warning mechanism to notify relevant personnel for analysis. Take corresponding emergency response measures according to specific implementation scenarios, such as traffic management, environmental monitoring, infrastructure maintenance, etc., to deal with potential risks or abnormal situations.

[0064] If the comprehensive monitoring status value is lower than the adjusted threshold, it indicates that the operation state of the smart city is normal, and there is no need to take immediate emergency measures. Continue to monitor and collect monitoring data, and regularly recalculate the comprehensive monitoring status value and the adjusted threshold according to the preset time points to ensure dynamic monitoring of the city operation state.

[0065] In summary, the smart city monitoring and early warning method based on digital twin is completed.

[0066] The sequence of the invention embodiments is only for description and does not represent the superiority or inferiority of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0067] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments.

[0068] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A smart city monitoring and early warning method based on digital twins, characterized in that: The following steps are involved: S1: Set the monitoring time point, collect and pre-process the raw data, and obtain the monitoring data; based on the detection data, calculate the comprehensive status of the smart city; Based on the comprehensive status of the smart city, the state change amount and the state change prediction amount are calculated; S2: Based on the state change amount and the state change prediction amount, an adaptive threshold adjustment mechanism is used to obtain the adjusted threshold, and a comprehensive monitoring state value is calculated at the same time. The comprehensive monitoring state value is compared with the adjusted threshold to determine whether to trigger the alarm mechanism.

2. The smart city monitoring and early warning method based on digital twin according to claim 1 is characterized in that: The S1 specifically includes: The calculation formula for the comprehensive status of a smart city is as follows: Where M(t) is the comprehensive state of the smart city at time t; t is the time variable; n is the number of data sources, that is, the number of sensors or monitoring devices; i and k are index variables of the data source; θ i (t) is the weight coefficient of the i-th data source; D i (t) is the monitoring data collected from the i-th data source at time t; D k (t) is the monitoring data collected from the kth data source at time t; ∈ is a smoothing constant.

3. The smart city monitoring and early warning method based on digital twin according to claim 2 is characterized in that: The S1 specifically includes: Based on the comprehensive state of the smart city, the exponential response term is introduced, and the state change is obtained by combining the response sensitivity coefficient and the exponential response coefficient to measure the comprehensive state fluctuation of the smart city at different time points. The specific implementation formula is: ΔM1(t)=M(t)-M(t-1)+α·(e β·(M(t)-M(t-1)) -1) Among them, ΔM1(t) is the state change at time t; t is the time variable; M(t) is the comprehensive state of the smart city at time t; M(t-1) is the comprehensive state of the smart city at time t-1; α is the response sensitivity coefficient; e β·(M(t)-M(t-1)) is the exponential response term; β is the exponential response coefficient.

4. The digital twin-based smart city monitoring and early warning method according to claim 3 is characterized in that: The S1 specifically includes: The state change prediction is based on the current state change and the comprehensive state change rate of the smart city, predicting the state change in the future, and quantifying the future change trend of the smart city. The specific implementation formula is: Among them, ΔM2(t, t+Δt) is the state change prediction; Δt is the time interval, which is the time span of the state change prediction; t is the time variable; τ is the virtual time variable in the integration process; is the rate of change of the comprehensive state of the smart city at time τ; M(τ) is the comprehensive state of the smart city at time τ; γ is the attenuation coefficient.

5. The smart city monitoring and early warning method based on digital twin according to claim 1 is characterized in that: The S2 specifically includes: The adaptive threshold adjustment mechanism dynamically adjusts the threshold by weighting and nonlinearly adjusting the state change amount and the state change prediction amount, responding to sudden changes and abnormal situations, and obtaining the adjusted threshold.

6. The digital twin-based smart city monitoring and early warning method according to claim 5 is characterized in that: The S2 specifically includes: After the adaptive threshold adjustment mechanism ends, the monitoring data, state change quantity and state change prediction quantity are processed by multi-dimensional data stream fusion to calculate the comprehensive monitoring state value.

7. The digital twin-based smart city monitoring and early warning method according to claim 6 is characterized in that: The S2 specifically includes: The implementation formula for multi-dimensional data stream fusion is as follows: Where F(t) is the comprehensive monitoring status value of the smart city at time t; t is the time variable; n is the number of data sources; i is the index variable of the data source; w i is the influence of the ith data source on the comprehensive monitoring status of the smart city; D i (t) is the monitoring data collected from the i-th data source at time t; is the time decay term; λ is the time decay coefficient; t i is the timestamp of the monitoring data collected from the i-th data source at time t; ΔM1(t) is the state change at time t; ζ is the smoothing parameter; tanh is the hyperbolic tangent function; ΔM2(t, t+Δt) is the predicted state change; is a weighted nonlinear adjustment term.

8. The digital twin-based smart city monitoring and early warning method according to claim 7 is characterized in that: The S2 specifically includes: The calculated comprehensive monitoring status value is compared with the adjusted threshold. When the comprehensive monitoring status value is greater than or equal to the adjusted threshold, it indicates that the smart city has abnormalities and potential risks; when the comprehensive monitoring status value is lower than the adjusted threshold, it indicates that the smart city is operating normally.

Images