Smart city data center system based on digital twinning and method thereof

Abstract

The invention relates to the technical field of smart cities, in particular to a smart city data center system based on digital twinning and a method thereof, which are used for solving the problems that in the prior art, a water affair management prediction model cannot be constructed to realize accurate prediction of water affair management related data, and a scientific basis cannot be provided for water affair management; the problems that requirements of different water affair systems cannot be adapted, potential problems cannot be recognized in advance through a prediction model, resource waste and operation and maintenance cost are increased, and system operation efficiency is reduced are solved. According to the invention, a prediction model of water demand, water quality change and pipe network fault risk is constructed based on historical and real-time data through the digital twinborn prediction module, accurate prediction is realized by using a machine learning algorithm, a scientific basis is provided for water affair management, and modular design supports flexible expansion, adapts to demand changes of different water affair systems, and has a wide application prospect. Potential problems are recognized in advance through the prediction model, resource waste and operation and maintenance cost are reduced, and system operation efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of smart city technology, and more specifically, to a smart city data middle platform system and method based on digital twins. Background Art

[0002] In the existing smart city construction, data integration among departments is insufficient, and there is a lack of in-depth analysis and data modeling capabilities, resulting in a large amount of information resources not being fully utilized; at the same time, urban governance information is initially integrated but is severely fragmented, and there is a lack of effective control over the comprehensive situation of the city.

[0003] The patent application with reference publication number CN118470991A discloses a 5G-based smart city intelligent traffic control system and method. By acquiring various data of the target city, a city digital twin model is constructed, and video and environmental data acquisition terminals are arranged on the target roads; these terminals upload the video and environmental data to the edge computing node via 5G, which uses object detection, semantic segmentation and other technologies to identify traffic and environmental objects from the video, and predicts the risk index of each road section in real time through deep learning; if the risk index exceeds the limit, the edge computing node will send the control instruction to the traffic light via 5G to adjust the traffic flow; at the same time, the edge computing node will upload the congestion warning results to the encrypted blockchain and send it to the vehicle-mounted equipment; for emergency events, the edge computing node will use blockchain and federated learning to find the optimal troubleshooting solution and continuously optimize it. The solution of the present invention utilizes edge computing, blockchain and multidisciplinary technology integration to improve the analysis efficiency and service security of traffic management work;

[0004] However, the above-mentioned reference patent utilizes 5G, edge computing and blockchain technology to analyze traffic data in real time, intelligently predict traffic conditions, automatically optimize signal control, ensure secure information distribution and efficient handling of emergencies, and improve the intelligence and safety of smart city traffic management. However, it is unable to build a water management prediction model to achieve accurate prediction of water management-related data, and cannot provide a scientific basis for water management. It cannot adapt to changes in demand for different water systems, and cannot identify potential problems in advance through prediction models, which increases resource waste and operation and maintenance costs, and reduces system operation efficiency. At the same time, it is unable to perform high-precision monitoring of multiple parameters of water equipment, cannot accurately evaluate the operating status of water equipment, cannot respond quickly to potential problems, cannot reduce unexpected shutdowns of water equipment, and cannot ensure the safe and stable operation of water equipment.

[0005] To this end, we propose a smart city data middle-office system and method based on digital twins to address the above problems. Summary of the invention

[0006] The purpose of the present invention is to provide a smart city data middle-end system and method based on digital twins, which solves the problems that the prior art cannot construct a water management prediction model to achieve accurate prediction of water management related data, cannot provide a scientific basis for water management, cannot adapt to changes in demand for different water systems, cannot identify potential problems in advance through prediction models, increase resource waste and operation and maintenance costs, and reduce system operation efficiency. At the same time, it is impossible to perform high-precision monitoring of multiple parameters of water equipment, cannot accurately evaluate the operating status of water equipment, cannot quickly respond to potential problems, cannot reduce unexpected shutdowns of water equipment, and cannot ensure the safe and stable operation of water equipment.

[0007] The purpose of the present invention is achieved through the following technical solutions:

[0008] A digital twin-based smart city data middle platform system, including a cloud management platform, a data acquisition module, a digital twin prediction module, a water affairs scheduling optimization module, and an abnormal warning module;

[0009] The data collection module is used to collect water service data in the water service system in real time, pre-process the collected water service data, and send the pre-processed water service data to the cloud management platform;

[0010] The digital twin prediction module is used to build a virtual water system, synchronize water data in real time, simulate the operating status of different scenarios through dynamic simulation, and predict future water demand, water quality changes and pipe network failure risks based on historical and real-time data;

[0011] The water dispatch optimization module is used to formulate water dispatch strategies and optimize water resource allocation and pipe network operation based on water demand, water quality changes, and pipe network failure risk prediction results;

[0012] The abnormal warning module is used to monitor the operating parameters of water equipment in real time. When an abnormal situation is detected, an early warning is immediately triggered and corresponding measures are taken to deal with it.

[0013] As a preferred embodiment of the present invention, the specific process of the digital twin prediction module simulating the operating status of different scenarios and predicting future water demand, water quality changes and pipe network failure risks is as follows:

[0014] Obtain pre-processed historical water service data, which includes equipment operation parameters, water quality parameters, environmental parameters, water use parameters and pipe network parameters. Equipment operation parameters include water pump pressure, reservoir water level, operating voltage, operating current and abnormal sound value of key equipment. Water quality parameters include water pH value, water dissolved oxygen and water conductivity. Environmental parameters include ambient temperature, ambient humidity and rainfall. Water use parameters include water consumption, water use time and water use frequency. Pipe network parameters include pipe network pressure and pipe network flow rate.

[0015] The specific steps to build a virtual water system through digital twin technology are as follows:

[0016] A physical model is built based on the pipe network topology and equipment operation principle in the water system. Fluid dynamics equations are used to simulate water flow dynamics. The pipe network hydraulic model is used to simulate the pipe network pressure distribution and flow distribution. A 3D visualization model of the water system is built using geographic information system and building information modeling technology.

[0017] As a preferred embodiment of the present invention, pre-processed real-time water service data is obtained, and the real-time data synchronization technology is used to synchronize the real-time water service data to the 3D visualization model of the water service system, and the operation status of the water service system is displayed in real time through the 3D visualization model of the water service system;

[0018] The specific steps for simulating the operating status of different scenarios through dynamic simulation are as follows:

[0019] Define simulation scenarios and divide them into normal water supply scenarios, extreme weather scenarios, equipment failure scenarios, and water consumption peak scenarios;

[0020] Use finite element analysis and computational fluid dynamics technology to simulate the operating status of the water system, adjust the parameters of the simulation model, and simulate the operating status under different scenarios.

[0021] As a preferred implementation of the present invention, a collection cycle is generated, the collection cycle duration is one year, the collection cycle is equally divided into m continuous sub-cycles, and the midpoint moment of each sub-cycle is marked to obtain m midpoint moments;

[0022] Taking m midpoint moments as the base point, expand forward and backward by equal time lengths, mark s-1 expansion moments, and then summarize the midpoint moments and s-1 expansion moments to obtain s detection moments;

[0023] The ambient temperature values ​​at s detection moments are measured by the sensor to obtain s detection ambient temperature values, and the s detection ambient temperature values ​​are accumulated and averaged to obtain m sub-ambient temperature values;

[0024] The expression of the sub-ambient temperature value is:

[0025] Where ZHW zm is the sub-environment temperature value of the mth sub-cycle, ZHW jcmn is the nth detected ambient temperature value in the mth sub-cycle;

[0026] Remove the maximum and minimum values ​​of the sub-environment temperature values, accumulate the remaining m-2 sub-environment temperature values ​​and calculate the average to obtain the mean value of the environment temperature;

[0027] The expression of the mean ambient temperature is:

[0028] Where HW jz is the mean ambient temperature, ZHW zp is the sub-environment temperature value of the pth sub-period.

[0029] As a preferred embodiment of the present invention, the method of calculating the mean value of the ambient temperature can be used to obtain the mean value of the ambient humidity HS jz , mean rainfall JY jz , mean pump pressure SY jz , mean reservoir water level height SSG jz , average operating voltage DY jz , average operating current DL jz , Average value of abnormal noise of key equipment GSY jz , water body pH mean SP jz 、Mean dissolved oxygen value SR jz , mean SD of water conductivity jz , average water consumption YL jz , average water usage time YS jz , average water use frequency YP jz , mean network pressure GY jz And the mean flow rate of the pipeline network GL jz ;

[0030] Set the ambient temperature HW jz , average ambient humidity HS jz , mean rainfall JY jz , average water consumption YL jz , average water usage time YS jz and the mean water use frequency YP jz The water demand prediction matrix YXJ is constructed by combination, and the water demand prediction matrix YXJ is used as the input of the machine learning model. The water demand in the next day corresponding to each group of water demand prediction matrices YXJ is used as the output of the machine learning model. The water demand in the next day is used as the prediction target, and the sum of the prediction errors of all training data is minimized as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence and the training is stopped. The water demand prediction model is obtained. The expression formula of the water demand prediction model is as follows:

[0031] YXL=η1·YL jz +η2·YS jz +η3·YP jz +η4·HW jz +η5·HS jz +η6·JY jz +λ;

[0032] Among them, η1, η2, η3, η4, η5 and η6 are regression coefficients, λ is the random error term, and YXL represents the water demand in the next day.

[0033] As a preferred embodiment of the present invention, the water body pH value SP jz 、Mean dissolved oxygen value SR jz , mean SD of water conductivity jz 、Ambient temperature average HW jz , average ambient humidity HS jz And the average rainfall JY jz The water quality change prediction matrix SZJ is constructed by combination, and the water quality change prediction matrix SZJ is used as the input of the machine learning model, and the water quality change matrix in the next day corresponding to each group of water quality change prediction matrices SZJ is used as the output of the machine learning model. The water quality change matrix in the next day is used as the prediction target, and the sum of the prediction errors of all training data is minimized as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence and the training is stopped to obtain the water quality change prediction model. The expression formula of the water quality change prediction model is as follows:

[0034] SZJ=β1·SP jz +β2·SR jz +β3·SD jz +β4·HW jz +β5·HS jz +β6·JY jz +ε;

[0035]

[0036] Among them, SZJ represents the water quality change matrix within the next day, β1, β2, β3, β4, β5 and β6 are all regression coefficients, ε is the random error term, ΔPV represents the change value of water pH value, ΔDO represents the change value of water dissolved oxygen, and ΔEC represents the change value of water conductivity;

[0037] The pump pressure average SY jz , mean reservoir water level height SSG jz , average operating voltage DY jz , average operating current DL jz , Average value of abnormal noise of key equipment GSY jz , mean network pressure GY jz And the mean flow rate of the pipeline network GL jzThe pipeline failure risk prediction matrix GFJ is constructed in combination, and the pipeline failure risk prediction matrix GFJ is used as the input of the machine learning model. The risk probability value of pipeline failure in the next day corresponding to each group of pipeline failure risk prediction matrices GFJ is used as the output of the machine learning model. The risk probability value of pipeline failure in the next day is used as the prediction target, and the sum of the prediction errors of all training data is minimized as the training target. The machine learning model is trained until the sum of the prediction errors converges and the training is stopped. The pipeline failure risk prediction model is obtained. The expression formula of the pipeline failure risk prediction model is as follows:

[0038]

[0039] Where P GF represents the risk probability value of pipeline network failure within the next day, w0 represents the intercept term, which represents the baseline probability of pipeline network failure, and w1, w2, w3, w4, w5, w6 and w7 are all regression coefficients.

[0040] As a preferred embodiment of the present invention, the specific process of the water dispatch optimization module formulating the water dispatch strategy according to the water demand, water quality change and pipe network failure risk prediction results is as follows:

[0041] The overall water supply area of ​​the city is divided into multiple water supply areas. The water demand, water quality change matrix and risk probability value of pipeline failure in each area in the next day are obtained. According to the prediction results, a water dispatching strategy is formulated to optimize water resource allocation and pipeline operation. The specific contents of the water dispatching strategy are as follows:

[0042] First, set the priority of water dispatch: give priority to ensuring water for residents and industry, ensuring living and production needs, maintaining water quality in areas with poor water quality, reducing the impact on residents, giving priority to dispatching high-quality water sources, focusing on monitoring and dispatching areas with high probability of failure, and formulating emergency plans in advance;

[0043] Formulate a water supply plan: Based on water demand, determine the optimal total water supply for each area to ensure a balance between supply and demand. Select a suitable water source based on water quality forecasts. Based on real-time monitoring data, dynamically adjust the water supply plan to ensure the flexibility and responsiveness of the water supply system.

[0044] Finally, the pipeline network is dispatched: according to the water demand of each area and the risk of pipeline failure, the pipeline flow is flexibly adjusted to ensure reasonable energy distribution. In areas with potential pipeline failures, emergency water supply routes are planned in advance to ensure that the water supply is quickly switched when a failure occurs. By adjusting the pump station and valves, the pipeline pressure is controlled to reduce the risk of pipe burst.

[0045] As a preferred embodiment of the present invention, the specific process of the abnormal warning module monitoring the operating parameters of water equipment in real time is as follows:

[0046] Obtaining the operating parameters of the water equipment, including the operating current, operating temperature and operating vibration intensity, generating a monitoring cycle, and dividing the monitoring cycle T into a plurality of monitoring time periods {t1, t2, ..., tn}, that is, T = {t1, t2, ..., tn};

[0047] Obtain the operating current change rate of the water equipment in multiple monitoring periods. The operating current change rate represents the ratio between the operating current change and the duration of the corresponding time period. This is used to construct a set A of operating current change rates, and the mean of the difference between the largest subset and the smallest subset in set A is recorded as the operating current change rate difference DLC;

[0048] Obtain the operating temperature change rate of the water equipment in multiple monitoring periods. The operating temperature change rate represents the ratio between the operating temperature change and the duration of the corresponding time period. This is used to construct a set B of operating temperature change rates, and the average of the difference between the maximum subset and the minimum subset in set B is recorded as the operating temperature change rate difference WDC;

[0049] The operating vibration intensity change rate of the water equipment in multiple monitoring periods is obtained. The operating vibration intensity change rate represents the ratio between the change in operating vibration intensity and the duration of the corresponding time period. The set C of operating vibration intensity is constructed in this way, and the mean of the difference between the largest subset and the smallest subset in set C is recorded as the operating vibration intensity change rate difference ZQC.

[0050] As a preferred embodiment of the present invention, the operating current change rate difference DLC, the operating temperature change rate difference WDC and the operating vibration intensity change rate difference ZQC are obtained, and the operating status assessment coefficient YZP is calculated by the following formula:

[0051]

[0052] Where d1, d2 and d3 are all preset proportional factor coefficients, d3>d2>d1>0, and the operating status assessment coefficient YZP is compared with the preset operating status assessment coefficient threshold:

[0053] If the operating status assessment coefficient YZP is less than the preset operating status assessment coefficient threshold, it indicates that the water equipment is operating normally;

[0054] If the operating status assessment coefficient YZP is greater than or equal to the preset operating status assessment coefficient threshold, it indicates that the operating status of the water equipment is abnormal, and an equipment operation abnormality signal is generated and sent to the cloud management platform;

[0055] Upon receiving a signal of abnormal equipment operation, the cloud management platform immediately triggers an audible and visual warning and sends a warning notification to relevant operation and maintenance personnel, and immediately takes corresponding measures to deal with it.

[0056] As a preferred embodiment of the present invention, a smart city data middle platform method based on digital twins includes the following steps:

[0057] Step 1: Collect water service data from the water service system in real time, pre-process the collected water service data, and send the pre-processed water service data to the cloud management platform;

[0058] Step 2: Build a virtual water system, synchronize water data in real time, simulate the operating status of different scenarios through dynamic simulation, and predict future water demand, water quality changes and pipe network failure risks based on historical and real-time data;

[0059] Step 3: Develop water dispatch strategies to optimize water resource allocation and pipe network operation based on water demand, water quality changes, and pipe network failure risk prediction results;

[0060] Step 4: Monitor the operating parameters of water equipment in real time. When an abnormal situation is detected, trigger an early warning immediately and take appropriate measures to deal with it.

[0061] Compared with the prior art, the advantages of the present invention are:

[0062] (1) In the present invention, a prediction model for water demand, water quality changes and pipe network failure risks is constructed based on historical and real-time data through a digital twin prediction module, and accurate prediction is achieved using a machine learning algorithm to provide a scientific basis for water management. By dividing the collection period, calculating the data mean and removing outliers, data quality is improved, noise interference is reduced, and the accuracy and stability of the prediction model are guaranteed. The modular design supports flexible expansion and adapts to changes in demand for different water systems. Potential problems are identified in advance through the prediction model, reducing resource waste and operation and maintenance costs, and improving system operation efficiency.

[0063] (2) In the present invention, high-precision multi-parameter monitoring is performed through the abnormal warning module to capture subtle changes, and the operating status assessment coefficient is introduced to realize intelligent evaluation. The system immediately triggers the warning and notifies the operation and maintenance personnel, quickly responds to potential problems, reduces unexpected downtime, improves system reliability, accurately locates the fault point and optimizes maintenance work, and uses accumulated data to support scientific decision-making and continuous optimization, ensuring the safe and stable operation of water equipment and significantly improving operational efficiency and reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 This is a system block diagram of Embodiment 1 of the present invention;

[0065] Figure 2This is a system block diagram of Embodiment 2 of the present invention;

[0066] Figure 3 This is a logical flow diagram of the second embodiment of the present invention;

[0067] Figure 4 This is a flow chart of the smart city data middle platform method in Example 3 of the present invention. DETAILED DESCRIPTION

[0068] The following will combine the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all the embodiments. All other embodiments obtained by ordinary technicians in this field without creative work based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0069] Embodiment 1: Figure 1 As shown, the present invention proposes a smart city data middle platform system based on digital twins, including a cloud management platform, a data acquisition module, a digital twin prediction module, a water affairs scheduling optimization module and an abnormal warning module;

[0070] The data collection module is used to collect water data from the water system in real time and perform preprocessing operations on the collected water data, including data cleaning, denoising and format conversion, and send the preprocessed water data to the cloud management platform;

[0071] The data acquisition module has the advantages of real-time, high efficiency and security in the water system. Through data cleaning, denoising and format conversion, it improves data quality and unifies standards for easy storage and analysis. Preprocessing reduces data volume, improves transmission efficiency, and reduces costs and delays. Modular design supports expansion, and automated operation reduces manual intervention.

[0072] The digital twin prediction module is used to build a virtual water system, synchronize water data in real time, simulate the operating status of different scenarios through dynamic simulation, and predict future water demand, water quality changes and pipe network failure risks based on historical and real-time data;

[0073] The specific process of the digital twin prediction module simulating the operating status of different scenarios and predicting future water demand, water quality changes and pipe network failure risks is as follows:

[0074] Obtain pre-processed historical water service data, which includes equipment operation parameters, water quality parameters, environmental parameters, water use parameters and pipe network parameters. Equipment operation parameters include water pump pressure, reservoir water level, operating voltage, operating current and abnormal sound value of key equipment. Water quality parameters include water pH value, water dissolved oxygen and water conductivity. Environmental parameters include ambient temperature, ambient humidity and rainfall. Water use parameters include water consumption, water use time and water use frequency. Pipe network parameters include pipe network pressure and pipe network flow rate.

[0075] The specific steps to build a virtual water system through digital twin technology are as follows:

[0076] Build a physical model based on the network topology and equipment operation principle in the water system, use fluid dynamics equations to simulate water flow dynamics, use the network hydraulic model to simulate the network pressure distribution and flow distribution, and use geographic information system and building information modeling technology to build a 3D visualization model of the water system;

[0077] Obtain pre-processed real-time water data, use real-time data synchronization technology to synchronize the real-time water data to the 3D visualization model of the water system, and display the operating status of the water system in real time through the 3D visualization model of the water system;

[0078] The specific steps for simulating the operating status of different scenarios through dynamic simulation are as follows:

[0079] Define simulation scenarios and divide them into normal water supply scenarios, extreme weather scenarios, equipment failure scenarios, and water consumption peak scenarios;

[0080] Finite element analysis and computational fluid dynamics technology are used to simulate the operating status of the water system, adjust the parameters of the simulation model, and simulate the operating status under different scenarios. The above process of building a virtual water system and performing simulation is a mature technology in the prior art, which will not be elaborated here. Here, an implementation method is described;

[0081] The digital twin prediction module uses real-time data synchronization technology to synchronize the pre-processed water data to the 3D visualization model in real time, ensuring that the virtual system is consistent with the actual system status and supporting real-time monitoring and decision-making. At the same time, it supports the definition of multiple scenarios and simulates the operating status under different scenarios through finite element analysis and computational fluid dynamics technology, providing a basis for system optimization and emergency response.

[0082] Generate a collection cycle with a duration of one year. Divide the collection cycle into m consecutive sub-cycles and mark the midpoint of each sub-cycle to obtain m midpoints.

[0083] Taking m midpoint moments as the base point, expand forward and backward by equal time lengths, mark s-1 expansion moments, and then summarize the midpoint moments and s-1 expansion moments to obtain s detection moments;

[0084] The ambient temperature values ​​at s detection moments are measured by the sensor to obtain s detection ambient temperature values, and the s detection ambient temperature values ​​are accumulated and averaged to obtain m sub-ambient temperature values;

[0085] The expression of the sub-ambient temperature value is:

[0086] Where ZHW zm is the sub-environment temperature value of the mth sub-cycle, ZHW jcmn is the nth detected ambient temperature value in the mth sub-cycle;

[0087] Remove the maximum and minimum values ​​of the sub-environment temperature values, accumulate the remaining m-2 sub-environment temperature values ​​and calculate the average to obtain the mean value of the environment temperature;

[0088] The expression of the mean ambient temperature is:

[0089] Where HW jz is the mean ambient temperature, ZHW zp is the sub-environment temperature value of the pth sub-period;

[0090] The mean ambient humidity HS can be obtained by using the method of calculating the mean ambient temperature. jz , mean rainfall JY jz , mean pump pressure SY jz , mean reservoir water level height SSG jz , average operating voltage DY jz , average operating current DL jz , Average value of abnormal noise of key equipment GSY jz , water body pH mean SP jz 、Mean dissolved oxygen value SR jz , mean SD of water conductivity jz , average water consumption YL jz , average water usage time YS jz , average water use frequency YP jz , mean network pressure GY jz And the mean flow rate of the pipeline network GL jz ;

[0091] Set the ambient temperature HW jz , average ambient humidity HS jz , mean rainfall JY jz , average water consumption YL jz, average water usage time YS jz and the mean water use frequency YP jz The water demand prediction matrix YXJ is constructed by combination, and the water demand prediction matrix YXJ is used as the input of the machine learning model. The water demand in the next day corresponding to each group of water demand prediction matrices YXJ is used as the output of the machine learning model. The water demand in the next day is used as the prediction target, and the sum of the prediction errors of all training data is minimized as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence and the training is stopped. The water demand prediction model is obtained. The expression formula of the water demand prediction model is as follows:

[0092] YXL=η1·YL jz +η2·YS jz +η3·YP jz +η4·HW jz +η5·HS jz +η6·JY jz +λ;

[0093] Among them, η1, η2, η3, η4, η5 and η6 are regression coefficients, λ is a random error term, and YXL represents the water demand in the next day;

[0094] The average pH value of water body SP jz 、Mean dissolved oxygen value SR jz , mean SD of water conductivity jz 、Ambient temperature average HW jz , average ambient humidity HS jz And the average rainfall JY jz The water quality change prediction matrix SZJ is constructed by combination, and the water quality change prediction matrix SZJ is used as the input of the machine learning model, and the water quality change matrix in the next day corresponding to each group of water quality change prediction matrices SZJ is used as the output of the machine learning model. The water quality change matrix in the next day is used as the prediction target, and the sum of the prediction errors of all training data is minimized as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence and the training is stopped to obtain the water quality change prediction model. The expression formula of the water quality change prediction model is as follows:

[0095] SZJ=β1·SP jz +β2·SR jz +β3·SD jz +β4·HW jz +β5·HS jz +β6·JY jz +ε;

[0096]

[0097] Among them, SZJ represents the water quality change matrix within the next day, β1, β2, β3, β4, β5 and β6 are all regression coefficients, ε is the random error term, ΔPV represents the change value of water pH value, ΔDO represents the change value of water dissolved oxygen, and ΔEC represents the change value of water conductivity;

[0098] The pump pressure average SY jz , mean reservoir water level height SSG jz , average operating voltage DY jz , average operating current DL jz , Average value of abnormal noise of key equipment GSY jz , mean network pressure GY jz And the mean flow rate of the pipeline network GL jz The pipeline failure risk prediction matrix GFJ is constructed in combination, and the pipeline failure risk prediction matrix GFJ is used as the input of the machine learning model. The risk probability value of pipeline failure in the next day corresponding to each group of pipeline failure risk prediction matrices GFJ is used as the output of the machine learning model. The risk probability value of pipeline failure in the next day is used as the prediction target, and the sum of the prediction errors of all training data is minimized as the training target. The machine learning model is trained until the sum of the prediction errors converges and the training is stopped. The pipeline failure risk prediction model is obtained. The expression formula of the pipeline failure risk prediction model is as follows:

[0099]

[0100] Where P GF represents the risk probability value of pipeline failure within the next day, w0 represents the intercept term, which represents the baseline probability of pipeline failure, and w1, w2, w3, w4, w5, w6 and w7 are all regression coefficients;

[0101] Through the digital twin prediction module, based on historical and real-time data, a prediction model for water demand, water quality changes and pipeline failure risks is constructed. The machine learning algorithm is used to achieve accurate prediction, providing a scientific basis for water management. By dividing the collection period, calculating the data mean and eliminating outliers, the data quality is improved, the noise interference is reduced, and the accuracy and stability of the prediction model are guaranteed. The modular design supports flexible expansion to adapt to the changes in demand of different water systems. Potential problems can be identified in advance through the prediction model, which reduces resource waste and operation and maintenance costs, and improves system operation efficiency.

[0102] The water dispatch optimization module is used to formulate water dispatch strategies and optimize water resource allocation and pipe network operation based on water demand, water quality changes, and pipe network failure risk prediction results;

[0103] The specific process of the water dispatch optimization module formulating water dispatch strategies based on water demand, water quality changes, and pipeline failure risk prediction results is as follows:

[0104] The overall water supply area of ​​the city is divided into multiple water supply areas. The water demand, water quality change matrix and risk probability value of pipeline failure in each area in the next day are obtained. According to the prediction results, a water dispatching strategy is formulated to optimize water resource allocation and pipeline operation. The specific contents of the water dispatching strategy are as follows:

[0105] First, set the priority of water dispatch: give priority to ensuring water for residents and industry, ensuring living and production needs, maintaining water quality in areas with poor water quality, reducing the impact on residents, giving priority to dispatching high-quality water sources, focusing on monitoring and dispatching areas with high probability of failure, and formulating emergency plans in advance;

[0106] Formulate a water supply plan: Based on water demand, determine the optimal total water supply for each area to ensure a balance between supply and demand. Select a suitable water source based on water quality forecasts. Based on real-time monitoring data, dynamically adjust the water supply plan to ensure the flexibility and responsiveness of the water supply system.

[0107] Finally, the network is dispatched: according to the water demand of each area and the risk of network failure, the pipeline flow is flexibly adjusted to ensure reasonable energy distribution. In areas with potential network failures, emergency water supply routes are planned in advance to ensure rapid switching of water supply when failures occur. By adjusting pump stations and valves, the network pressure is controlled to reduce the risk of pipe bursts.

[0108] The water dispatch optimization module not only improves the efficiency and fairness of water resource allocation, but also significantly enhances the safety and reliability of the water system by setting reasonable priorities, formulating scientific water supply plans, and flexible pipeline network scheduling.

[0109] Embodiment 2: The technical solution of the embodiment of the present invention is different from that of embodiment 1 in that:

[0110] like Figure 2 and Figure 3 As shown, the abnormal warning module is used to monitor the operating parameters of water equipment in real time. When an abnormal situation is detected, an early warning is immediately triggered and corresponding measures are taken to deal with it;

[0111] The specific process of the abnormal warning module to monitor the operating parameters of water equipment in real time is as follows:

[0112] Obtaining the operating parameters of the water equipment, including the operating current, operating temperature and operating vibration intensity, generating a monitoring cycle, and dividing the monitoring cycle T into a plurality of monitoring time periods {t1, t2, ..., tn}, that is, T = {t1, t2, ..., tn};

[0113] Obtain the operating current change rate of the water equipment in multiple monitoring periods. The operating current change rate represents the ratio between the operating current change and the duration of the corresponding time period. This is used to construct a set A of operating current change rates, and the mean of the difference between the largest subset and the smallest subset in set A is recorded as the operating current change rate difference DLC;

[0114] Obtain the operating temperature change rate of the water equipment in multiple monitoring periods. The operating temperature change rate represents the ratio between the operating temperature change and the duration of the corresponding time period. This is used to construct a set B of operating temperature change rates, and the average of the difference between the maximum subset and the minimum subset in set B is recorded as the operating temperature change rate difference WDC;

[0115] Obtain the change rate of the operating vibration intensity of the water equipment in multiple monitoring periods. The change rate of the operating vibration intensity represents the ratio between the change in the operating vibration intensity and the length of the corresponding time period. This is used to construct a set C of operating vibration intensities, and the mean of the difference between the largest subset and the smallest subset in set C is recorded as the difference ZQC of the change rate of the operating vibration intensity.

[0116] The operating current change rate difference DLC, operating temperature change rate difference WDC and operating vibration intensity change rate difference ZQC are obtained, and the operating status assessment coefficient YZP is calculated using the following formula:

[0117]

[0118] Where d1, d2 and d3 are all preset proportional factor coefficients, d3>d2>d1>0, and the operating status assessment coefficient YZP is compared with the preset operating status assessment coefficient threshold:

[0119] If the operating status assessment coefficient YZP is less than the preset operating status assessment coefficient threshold, it indicates that the water equipment is operating normally;

[0120] If the operating status assessment coefficient YZP is greater than or equal to the preset operating status assessment coefficient threshold, it indicates that the operating status of the water equipment is abnormal, and an equipment operation abnormality signal is generated and sent to the cloud management platform;

[0121] Upon receiving the abnormal operation signal of the equipment, the cloud management platform immediately triggers the sound and light warning and sends a warning notification to the relevant operation and maintenance personnel, and immediately takes corresponding measures to deal with it. The specific contents of the handling measures are as follows:

[0122] Parameter adjustment: adjust the valve opening or pump station operating parameters to return the flow rate to the normal range, adjust the pump station operating parameters or valve opening to return the pressure to the normal range, and adjust the water treatment process parameters to return the water quality to the normal range;

[0123] Equipment maintenance: Conduct detailed inspections on relevant equipment, eliminate potential problems, repair or replace faulty equipment to ensure normal operation of the equipment, enable backup equipment, and ensure normal operation of the system;

[0124] Emergency response: immediately stop the operation of related equipment to prevent the accident from expanding, isolate the abnormal area to prevent the problem from spreading to other areas, organize an emergency team to handle the problem on site and quickly solve the problem;

[0125] Scheduling adjustment: adjust the water supply plan to reduce the impact on abnormal areas, deploy resources from other areas, and support the handling of abnormal areas;

[0126] Through the abnormal warning module, high-precision multi-parameter monitoring is carried out to capture subtle changes, and the operating status assessment coefficient is introduced to realize intelligent evaluation. The system immediately triggers the warning and notifies the operation and maintenance personnel, quickly responds to potential problems, reduces unexpected downtime, improves system reliability, accurately locates the fault point and optimizes maintenance work, and uses the accumulated data to support scientific decision-making and continuous optimization, ensuring the safe and stable operation of water equipment and significantly improving operational efficiency and reliability.

[0127] Embodiment 3: The technical solution of the embodiment of the present invention is different from that of Embodiment 1 and Embodiment 2 in that:

[0128] like Figure 4 As shown, a smart city data middle platform method based on digital twins includes the following steps:

[0129] Step 1: Collect water service data from the water service system in real time, pre-process the collected water service data, and send the pre-processed water service data to the cloud management platform;

[0130] Step 2: Build a virtual water system, synchronize water data in real time, simulate the operating status of different scenarios through dynamic simulation, and predict future water demand, water quality changes and pipe network failure risks based on historical and real-time data;

[0131] Step 3: Develop water dispatch strategies to optimize water resource allocation and pipe network operation based on water demand, water quality changes, and pipe network failure risk prediction results;

[0132] Step 4: Monitor the operating parameters of water equipment in real time. When abnormal conditions are detected, trigger an early warning and take appropriate measures to deal with them.

[0133] This digital twin-based smart city data middle platform method ensures efficient resource utilization through real-time data collection, dynamic simulation prediction and optimized scheduling, and monitors equipment operation in real time, immediately warns of abnormal situations, enhances system stability and security, and continuously optimizes algorithms using accumulated data to support scientific decision-making. Integrating digital twins, big data and artificial intelligence technologies to achieve full-process intelligent management, significantly improve the intelligence level and reliability of water systems, and build a smarter and greener urban water environment.

[0134] The above are only preferred specific implementation modes of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical solutions and improved concepts of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A smart city data middle platform system based on digital twins, characterized in that: It includes cloud management platform, data acquisition module, digital twin prediction module, water dispatch optimization module and abnormal warning module; The data collection module is used to collect water service data in the water service system in real time, pre-process the collected water service data, and send the pre-processed water service data to the cloud management platform; The digital twin prediction module is used to build a virtual water system, synchronize water data in real time, simulate the operating status of different scenarios through dynamic simulation, and predict future water demand, water quality changes and pipe network failure risks based on historical and real-time data; The water dispatch optimization module is used to formulate water dispatch strategies and optimize water resource allocation and pipe network operation based on water demand, water quality changes, and pipe network failure risk prediction results; The abnormal warning module is used to monitor the operating parameters of water equipment in real time. When an abnormal situation is detected, an early warning is immediately triggered and corresponding measures are taken to deal with it.

2. According to the digital twin-based smart city data middle platform system of claim 1, it is characterized in that: The specific process of the digital twin prediction module simulating the operating status of different scenarios and predicting future water demand, water quality changes and pipe network failure risks is as follows: Obtain pre-processed historical water service data, which includes equipment operation parameters, water quality parameters, environmental parameters, water use parameters and pipe network parameters. Equipment operation parameters include water pump pressure, reservoir water level, operating voltage, operating current and abnormal sound values ​​of key equipment. Water quality parameters include water pH value, water dissolved oxygen and water conductivity. Environmental parameters include ambient temperature, ambient humidity and rainfall. Water use parameters include water consumption, water use time and water use frequency. Pipe network parameters include pipe network pressure and pipe network flow rate. The specific steps to build a virtual water system through digital twin technology are as follows: A physical model is built based on the pipe network topology and equipment operation principle in the water system. Fluid dynamics equations are used to simulate water flow dynamics. The pipe network hydraulic model is used to simulate the pipe network pressure distribution and flow distribution. A 3D visualization model of the water system is built using geographic information system and building information modeling technology.

3. According to a digital twin-based smart city data middle platform system according to claim 2, it is characterized in that: Obtain pre-processed real-time water data, use real-time data synchronization technology to synchronize the real-time water data to the 3D visualization model of the water system, and display the operating status of the water system in real time through the 3D visualization model of the water system; The specific steps for simulating the operating status of different scenarios through dynamic simulation are as follows: Define simulation scenarios and divide them into normal water supply scenarios, extreme weather scenarios, equipment failure scenarios, and water consumption peak scenarios; Finite element analysis and computational fluid dynamics technology are used to simulate the operating status of the water system, adjust the parameters of the simulation model, and simulate the operating status under different scenarios.

4. According to the digital twin-based smart city data middle platform system of claim 3, it is characterized in that: Generate a collection cycle with a duration of one year. Divide the collection cycle into m consecutive sub-cycles and mark the midpoint of each sub-cycle to obtain m midpoints. Taking m midpoint moments as the base point, expand forward and backward by equal time lengths, mark s-1 expansion moments, and then summarize the midpoint moments and s-1 expansion moments to obtain s detection moments; The ambient temperature values ​​at s detection moments are measured by the sensor to obtain s detection ambient temperature values, and the s detection ambient temperature values ​​are accumulated and averaged to obtain m sub-ambient temperature values; The expression of the sub-ambient temperature value is: Where ZHW zm is the sub-environment temperature value of the mth sub-cycle, ZHW jcmn is the nth detected ambient temperature value in the mth sub-cycle; Remove the maximum and minimum values ​​of the sub-environment temperature values, accumulate the remaining m-2 sub-environment temperature values ​​and calculate the average to obtain the mean value of the environment temperature; The expression of the mean ambient temperature is: Where HW jz is the mean ambient temperature, ZHW zp is the sub-environment temperature value of the pth sub-period.

5. According to the digital twin-based smart city data middle platform system of claim 4, it is characterized in that: The mean ambient humidity HS can be obtained by using the method of calculating the mean ambient temperature. jz , mean rainfall JY jz , mean pump pressure SY jz , mean reservoir water level height SSG jz , average operating voltage DY jz 、Running current average DL jz , Average value of abnormal noise of key equipment GSY jz , water body pH mean SP jz 、Mean dissolved oxygen value SR jz , mean SD of water conductivity jz , average water consumption YL jz , average water usage time YS jz , average water use frequency YP jz , mean network pressure GY jz And the mean flow rate of the pipeline network GL jz ; Set the ambient temperature HW jz , Ambient humidity average HS jz , mean rainfall JY jz , average water consumption YL jz , average water usage time YS jz and the mean water use frequency YP jz The water demand prediction matrix YXJ is constructed by combination, and the water demand prediction matrix YXJ is used as the input of the machine learning model. The water demand in the next day corresponding to each group of water demand prediction matrices YXJ is used as the output of the machine learning model. The water demand in the next day is used as the prediction target, and the sum of the prediction errors of all training data is minimized as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence and the training is stopped. The water demand prediction model is obtained. The expression formula of the water demand prediction model is as follows: YXL=η1·YL jz +η2·YS jz +η3·YP jz +η4·HW jz +η5·HS jz +η6·JY jz +λ; Among them, η1, η2, η3, η4, η5 and η6 are regression coefficients, λ is the random error term, and YXL represents the water demand in the next day.

6. According to the digital twin-based smart city data middle platform system of claim 5, it is characterized in that: The average pH value of water body SP jz 、Mean dissolved oxygen value SR jz , mean SD of water conductivity jz 、Ambient temperature average HW jz , Ambient humidity average HS jz And the average rainfall JY jz The water quality change prediction matrix SZJ is constructed by combination, and the water quality change prediction matrix SZJ is used as the input of the machine learning model, and the water quality change matrix in the next day corresponding to each group of water quality change prediction matrices SZJ is used as the output of the machine learning model. The water quality change matrix in the next day is used as the prediction target, and the sum of the prediction errors of all training data is minimized as the training target. The machine learning model is trained until the sum of the prediction errors reaches convergence and the training is stopped to obtain the water quality change prediction model. The expression formula of the water quality change prediction model is as follows: SZJ=β1·SP jz +β2·SR jz +β3·SD jz +β4·HW jz +β5·HS jz +β6·JY jz +e; Among them, SZJ represents the water quality change matrix within the next day, β1, β2, β3, β4, β5 and β6 are all regression coefficients, ε is the random error term, ΔPV represents the change value of water pH value, ΔDO represents the change value of water dissolved oxygen, and ΔEC represents the change value of water conductivity; The pump pressure average SY jz , mean reservoir water level height SSG jz , average operating voltage DY jz 、Running current average DL jz , Average value of abnormal noise of key equipment GSY jz , mean network pressure GY jz And the mean flow rate of the pipeline network GL jz The pipeline failure risk prediction matrix GFJ is constructed in combination, and the pipeline failure risk prediction matrix GFJ is used as the input of the machine learning model. The risk probability value of pipeline failure in the next day corresponding to each group of pipeline failure risk prediction matrices GFJ is used as the output of the machine learning model. The risk probability value of pipeline failure in the next day is used as the prediction target, and the sum of the prediction errors of all training data is minimized as the training target. The machine learning model is trained until the sum of the prediction errors converges and the training is stopped. The pipeline failure risk prediction model is obtained. The expression formula of the pipeline failure risk prediction model is as follows: Where P GF represents the risk probability value of pipeline network failure within the next day, w0 represents the intercept term, which represents the baseline probability of pipeline network failure, and w1, w2, w3, w4, w5, w6 and w7 are all regression coefficients.

7. According to a digital twin-based smart city data middle platform system according to claim 1, it is characterized in that: The specific process of the water dispatch optimization module formulating a water dispatch strategy based on water demand, water quality changes, and pipe network failure risk prediction results is as follows: The overall water supply area of ​​the city is divided into multiple water supply areas. The water demand, water quality change matrix and risk probability value of pipeline failure in each area in the next day are obtained. According to the prediction results, a water dispatching strategy is formulated to optimize water resource allocation and pipeline operation. The specific contents of the water dispatching strategy are as follows: First, set the priority of water dispatch: give priority to ensuring water for residents and industry, ensuring living and production needs, maintaining water quality in areas with poor water quality, reducing the impact on residents, giving priority to dispatching high-quality water sources, focusing on monitoring and dispatching areas with high probability of failure, and formulating emergency plans in advance; Formulate a water supply plan: Based on water demand, determine the optimal total water supply for each area to ensure a balance between supply and demand. Select a suitable water source based on water quality forecasts. Based on real-time monitoring data, dynamically adjust the water supply plan to ensure the flexibility and responsiveness of the water supply system. Finally, the pipeline network is dispatched: according to the water demand of each area and the risk of pipeline failure, the pipeline flow is flexibly adjusted to ensure reasonable energy distribution. In areas with potential pipeline failures, emergency water supply routes are planned in advance to ensure that the water supply is quickly switched when a failure occurs. By adjusting the pump station and valves, the pipeline pressure is controlled to reduce the risk of pipe burst.

8. According to the digital twin-based smart city data middle platform system of claim 1, it is characterized in that: The specific process of the abnormal warning module to monitor the operating parameters of water equipment in real time is as follows: Obtaining the operating parameters of the water equipment, including the operating current, operating temperature and operating vibration intensity, generating a monitoring cycle, and dividing the monitoring cycle T into a plurality of monitoring time periods {t1, t2, ..., tn}, that is, T = {t1, t2, ..., tn}; Obtain the operating current change rate of the water equipment in multiple monitoring periods. The operating current change rate represents the ratio between the operating current change and the duration of the corresponding time period. This is used to construct a set A of operating current change rates, and the mean of the difference between the largest subset and the smallest subset in set A is recorded as the operating current change rate difference DLC; Obtain the operating temperature change rate of the water equipment in multiple monitoring periods. The operating temperature change rate represents the ratio between the operating temperature change and the duration of the corresponding time period. This is used to construct a set B of operating temperature change rates, and the average of the difference between the maximum subset and the minimum subset in set B is recorded as the operating temperature change rate difference WDC; The operating vibration intensity change rate of the water equipment in multiple monitoring periods is obtained. The operating vibration intensity change rate represents the ratio between the change in operating vibration intensity and the duration of the corresponding time period. The set C of operating vibration intensity is constructed in this way, and the mean of the difference between the largest subset and the smallest subset in set C is recorded as the operating vibration intensity change rate difference ZQC.

9. According to the digital twin-based smart city data middle platform system of claim 8, it is characterized in that: The operating current change rate difference DLC, operating temperature change rate difference WDC and operating vibration intensity change rate difference ZQC are obtained, and the operating status assessment coefficient YZP is calculated using the following formula: Where d1, d2 and d3 are all preset proportional factor coefficients, d3>d2>d1>0, and the operating status assessment coefficient YZP is compared with the preset operating status assessment coefficient threshold: If the operating status assessment coefficient YZP is less than the preset operating status assessment coefficient threshold, it indicates that the water equipment is operating normally; If the operating status assessment coefficient YZP is greater than or equal to the preset operating status assessment coefficient threshold, it indicates that the operating status of the water equipment is abnormal, and an equipment operation abnormality signal is generated and sent to the cloud management platform; Upon receiving a signal of abnormal equipment operation, the cloud management platform immediately triggers an audible and visual warning and sends a warning notification to relevant operation and maintenance personnel, and immediately takes corresponding measures to deal with it.

10. A smart city data middle platform method based on digital twins, characterized in that: The following steps are involved: Step 1: Collect water service data from the water service system in real time, pre-process the collected water service data, and send the pre-processed water service data to the cloud management platform; Step 2: Build a virtual water system, synchronize water data in real time, simulate the operating status of different scenarios through dynamic simulation, and predict future water demand, water quality changes and pipe network failure risks based on historical and real-time data; Step 3: Develop water dispatch strategies to optimize water resource allocation and pipe network operation based on water demand, water quality changes, and pipe network failure risk prediction results; Step 4: Monitor the operating parameters of water equipment in real time. When an abnormal situation is detected, trigger an early warning immediately and take appropriate measures to deal with it.

Images