A Smart Water Management Dynamic Monitoring Method and Platform Based on Digital Twin
By building a digital twin model and optimization knowledge map of the water pipeline network, the problem of difficult to predict and deal with future failures of the water pipeline network in the existing technology is solved, and real-time monitoring and optimization of the water pipeline network is realized to ensure its safe operation.
Patent Information
- Application Number
- CN202410609215.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-16
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-05-16
AI Technical Summary
The existing smart water dynamic supervision methods are difficult to effectively predict and respond to potential future failures or performance degradation of water pipelines, and it is difficult to quickly provide response plans when a failure occurs, so it is impossible to achieve rapid processing.
By building a digital twin model of the water pipeline network, monitoring its operating status and communication status, if the load or stability is found to be lower than expected, optimized measures will be taken, and optimization solutions will be provided using the optimization knowledge graph, and the water pipeline network will be screened and optimized by screening the optimization plan. Optimize to improve its operating stability.
Real-time monitoring and prediction of the operating status of the water pipeline network is realized, and the problems of overload and poor communication status can be dealt with in a timely manner, ensuring the safe operation of the water pipeline network, and improving the operating stability and safety of the water pipeline network.
Smart Images

Figure CN118570017B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of dynamic water affairs supervision, and specifically to a smart water affairs dynamic supervision method and platform based on digital twins. Background Art
[0002] Smart water management fully explores the value of data and logical relationships through the deep integration of new-generation information technology and water technology, realizes intelligent control, data resourceization, precise management, and intelligent decision-making of water business systems, ensures the safe operation of water facilities, and makes water business operations more efficient, management more scientific, and services better quality.
[0003] The realization of smart water services relies on the support of multiple key technologies, including the Internet of Things, sensor technology, and big data analysis. Among them, the Internet of Things realizes the remote dispatch and management of water equipment through real-time monitoring and dynamic data collection; sensor technology is used to monitor indicators such as water quality, water volume, and water pressure, obtain data in real time, and conduct analysis and prediction; big data analysis can help predict and optimize the use and allocation of water resources.
[0004] After the water supply network is installed and set up and enters the operating state, if it is under a large operating load for a long time, there may be certain safety hazards in the water supply network. In order to avoid such safety hazards or failures, the existing smart water dynamic supervision methods usually collect the current data of the water supply network and analyze the collected data to determine whether there are any faults or hidden dangers. However, this method can only judge and analyze the current problem and cannot effectively cover the next situation. At the same time, when the water supply network has already failed or its performance has declined, it is difficult to quickly provide a response plan and it is difficult to achieve rapid processing. Summary of the invention
[0005] In view of the shortcomings of the prior art, the present invention provides a smart water dynamic supervision method and platform based on digital twins, which collects relevant data of the water pipe network, trains to obtain a digital twin model of the water pipe network, monitors the communication status of the digital twin model of the water pipe network with the outside, and takes communication enhancement measures if the communication status is lower than expected; simulates and analyzes the water pipe network, and constructs the stability of the water pipe network from the test data; if the stability does not exceed the stability threshold, the water pipe network optimization knowledge graph gives an optimization plan for the water pipe network, and then the output optimization plan is screened, and the water pipe network is optimized by the screened optimization plan; when the water pipe network is optimized, the optimization effect can be guaranteed, thereby solving the technical solution proposed in the background technology.
[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions:
[0007] The present invention provides a smart water affairs dynamic supervision method based on digital twins, including monitoring the operating status of the water affairs network, and constructing the load degree Ho of the water affairs network during operation from the data obtained by monitoring. If the load degree of the water affairs network exceeds expectations, collect relevant data of the water affairs network, and summarize and construct a water affairs data set;
[0008] The digital twin model of the water pipe network is obtained by training the water pipe network and related data, the communication status between the digital twin model of the water pipe network and the outside is monitored, and the communication coefficient Txs is then constructed based on the monitoring data. If the communication coefficient Txs is lower than expected, communication enhancement measures are taken.
[0009] The water pipe network is simulated and analyzed, and the stability Wo of the water pipe network is constructed based on the test data. If the stability Wo does not exceed the stability threshold, a maintenance instruction is issued to the outside. The construction method of the stability Wo is as follows:
[0010]
[0011] Where i = 1, 2, ..., p, p is the number of analysis nodes; Ho i is the load on the i-th analysis node, Ho avg is the mean value of load degree;
[0012] After the water pipe network optimization knowledge graph gives an optimization plan for the water pipe network, the output optimization plan is screened according to the optimization degree Yo constructed by a number of stability degrees Wo, and the water pipe network is optimized by the screened optimization plan;
[0013] Collect water quality data related to the water supply network, and construct the water pollution degree Rxs based on the water quality data. If the pollution degree exceeds the pollution threshold, issue a water quality adjustment instruction to the outside.
[0014] In one embodiment, the operation status of the water pipe network is monitored, and the load degree Ho of the water pipe network during operation is constructed from the data obtained by the monitoring. If the load degree of the water pipe network exceeds expectations, relevant data of the water pipe network is collected and a water data set is constructed in aggregating, including:
[0015] Select several monitoring locations in the water pipe network, monitor the operation status of the water pipe network at each monitoring location, obtain the water flow Sn and operation noise Zn, and after aggregating the acquired monitoring data, construct a water pipe network operation data set;
[0016] The load degree Ho of the water pipe network during operation is constructed from the water pipe network operation data set; if the load degree of the water pipe network exceeds expectations, a data collection instruction is issued to the outside;
[0017] After receiving the data collection instruction, a sensor network and data collection system are built for the water supply network to collect various types of data in the water supply network, and the collected data are summarized to construct a water supply data set.
[0018] In one implementation, the water pipe network and related data are used to train a water pipe network digital twin model, monitor the communication status between the water pipe network digital twin model and the outside, and then construct a communication coefficient Txs based on the monitoring data. If the communication coefficient Txs is lower than expected, communication enhancement measures are taken, including:
[0019] Extract some data from the water service data set as sample data, build an initial model using a machine learning algorithm, train the initial model using the sample data, and obtain a trained digital twin model of the water service network;
[0020] After receiving the data input, the communication status of the digital twin model of the water pipe network is monitored to obtain the corresponding monitoring data, including the signal-to-noise ratio and data delay; and the acquired communication data is aggregated to construct a communication data set of the communication model;
[0021] The communication coefficient Txs of the model is constructed from the model communication data set;
[0022] After obtaining the communication coefficient Txs, the information update capability of the digital twin model is verified and evaluated using the communication coefficient Txs. The communication threshold is set in advance based on the expectations for model communication management. If the communication coefficient Txs is lower than expected, it means that the model’s current communication capability with the outside world is poor. At this time, the current data communication capability should be strengthened.
[0023] In one embodiment, the communication coefficient Txs of the model is constructed from the model communication data set in the following manner: linear normalization is performed on the signal-to-noise ratio Hp and the data delay Bp, and the corresponding data values are mapped to the interval [0, 1] in the following manner:
[0024]
[0025] Among them, k2 and k1 are weight coefficients, 0≤k1≤1, 0≤k2≤1, and k1+k2=1, Bp i The ith data delay, Hp i is the i-th signal-to-noise ratio, i=1, 2,…n, and n is the number of monitoring nodes.
[0026] In one implementation, the simulation analysis of the water pipe network is performed, and the stability Wo of the water pipe network is constructed based on the test data. If the stability Wo does not exceed the stability threshold, a maintenance instruction is issued to the outside, including:
[0027] The digital twin model of the water supply network is used to simulate and analyze the water supply network, and the corresponding operation data is obtained. The load degree Ho on each analysis node is then constructed based on the operation data. The load degree Ho is generated as follows: the water flow Sn and the operation noise Zn are linearly normalized, and the corresponding data values are mapped to the interval [0, 1] in the following manner:
[0028]
[0029] Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1, i=1, 2, …, k, k is the number of monitoring locations; Sn i is the water flow at the ith monitoring location, Sn avg is the mean water flow rate, As its qualified standard value, Zn i is the operating noise at the ith monitoring position, Zn avg is the mean of the operating noise, is its qualified standard value;
[0030] Summarize the acquired load degree Ho to construct a water service operation data set;
[0031] The stability Wo of the water supply network is constructed from the water supply operation data set. If the obtained stability Wo exceeds the stability threshold, a water quality monitoring instruction is issued to the outside; otherwise, a maintenance instruction is issued to the outside.
[0032] In one embodiment, providing an optimization solution for the water pipe network by using the water pipe network optimization knowledge graph includes:
[0033] After receiving the maintenance instruction, the digital twin model of the water pipe network predicts the operation status of the water pipe network and obtains the corresponding operation prediction data; identifies the operation prediction data and obtains the corresponding optimization features;
[0034] Taking water pipe network as the target word, a water pipe network optimization knowledge graph is pre-constructed. According to the correspondence between the optimization scheme and optimization features of the water pipe network, the water pipe network optimization knowledge graph provides an optimization scheme for the water pipe network.
[0035] In one embodiment, the output optimization scheme is screened based on the optimization degree Yo constructed by a plurality of stability degrees Wo, and the water pipe network is optimized by the screened optimization scheme, including:
[0036] After obtaining the output optimization plan, the water pipe network digital twin model is used to test the optimization plan, and the corresponding test data is obtained at each test node. The stability Wo of the water pipe network is reconstructed based on the test data;
[0037] The optimization degree Yo is constructed by several stabilities Wo. If the optimization degree Yo is lower than the optimization degree threshold, the optimization algorithm constructs a solution optimization model, and the solution optimization model is used to optimize the optimization solution to obtain the optimized solution, which is used as the target solution. If the optimization degree is not lower than the optimization degree threshold, the corresponding optimization solution is used as the target solution.
[0038] In one implementation, the optimization degree Yo is as follows: after obtaining the stability degree Wo on each prediction node, the stability degrees Wo before and after the test optimization scheme are aligned one by one, according to the following formula:
[0039]
[0040] Among them, Tz i is the median stability value at the i-th test node, Tz avg is its mean, i is the test node number, i = 1, 2, ..., n, n is the number of test nodes, AWo i and BWo i are the stability of the i-th node before and after optimization, AWo avg and BWo avg is the corresponding mean.
[0041] In one embodiment, the water quality data of water bodies related to the water pipe network is collected, and the pollution degree Rxs of the water body is constructed from the water quality data. If the pollution degree Rxs exceeds the pollution threshold, a water quality adjustment instruction is issued to the outside, including:
[0042] After the target plan is implemented to optimize the water pipe network, the pollution status of the water body is monitored during the monitoring period to obtain the corresponding monitoring data; a water body monitoring data set is constructed after aggregation; the water body pollution degree Rxs is constructed from the monitoring data in the water body monitoring data set, wherein the turbidity Uo is normalized according to the following formula:
[0043]
[0044] Where i = 1, 2, ..., m, m is the number of monitoring nodes, Uo i is the turbidity of the water body at the i-th monitoring node, Uo avg is the corresponding mean, is the qualified reference value of turbidity; weight coefficient: 0≤F1≤1, 0≤F2≤1, and F1+F2=1; the pollution threshold is set in advance based on historical data and expectations for water quality management; if the obtained pollution degree Rxs exceeds the pollution threshold, it means that the current water quality is abnormal, and a water quality adjustment instruction is issued to the outside.
[0045] The present invention also provides a smart water affairs dynamic supervision platform based on digital twins, including:
[0046] The load monitoring unit monitors the operation status of the water supply network and constructs the load degree Ho of the water supply network from the data obtained by monitoring. If the load degree of the water supply network exceeds expectations, the relevant data of the water supply network is collected and summarized to construct a water supply data set;
[0047] The communication condition evaluation unit is used to obtain the digital twin model of the water pipe network through training of the water pipe network and related data, monitor the communication status between the digital twin model of the water pipe network and the outside, and then construct the communication coefficient Txs from the monitoring data. If the communication coefficient Txs is lower than expected, communication enhancement measures are taken;
[0048] The stability analysis unit is used to simulate and analyze the water pipe network and construct the stability Wo of the water pipe network based on the test data. If the stability Wo does not exceed the stability threshold, a maintenance instruction is issued to the outside.
[0049] The scheme optimization unit is used to provide an optimization scheme for the water pipe network based on the water pipe network optimization knowledge graph, screen the output optimization scheme according to the optimization degree Yo constructed by a number of stability degrees Wo, and optimize the water pipe network with the screened optimization scheme;
[0050] The pollution warning unit is used to collect water quality data related to the water supply network, and construct the pollution degree Rxs of the water body based on the water quality data. If the pollution degree exceeds the pollution threshold, a water quality adjustment instruction is issued to the outside.
[0051] Beneficial effects of the present invention:
[0052] 1. Construct the corresponding load degree Ho from the monitoring data, and judge whether the current water pipe network is large or within the acceptable range based on the load degree Ho. If it is already in an overloaded state, it can be dealt with in a timely manner.
[0053] 2. Through the digital twin model of the water supply network, the operating status of the water supply network can be monitored and predicted, so that the operating status of the water supply network can be understood and perceived in advance. If there are any problems, they can be dealt with in a timely manner.
[0054] 3. When the digital twin model of the water pipe network is in operation, the data on its communication status with the outside world is collected and analyzed, and then the communication coefficient Txs is constructed. The communication coefficient Txs can be used to comprehensively evaluate and analyze the model update and operation status. If it is judged that the current communication status of the model is poor, the communication capabilities of the model and corresponding equipment can be guaranteed in time.
[0055] 4. Evaluate the stability of the water supply network based on the constructed stability Ho. If the water supply network is often operated under high load conditions, it means that the water supply network's carrying capacity for work tasks is limited, and the operational stability and safety of the water supply network may be affected to a certain extent. Timely inspection and maintenance of the water supply network can ensure the safe operation of the water supply network.
[0056] 5. Predict and obtain the operating status data of the water supply network, and then use the pre-built water supply network optimization knowledge graph to provide corresponding maintenance plans for the water supply network, so that the water supply network can be quickly optimized when necessary; combined with the simulation of the digital twin model of the water supply network, and based on the optimization degree Yo, each optimization plan can be screened. When optimizing the water supply network, the optimization effect can be guaranteed, so that the water supply network can always be in a better operating state.
[0057] 6. Monitor the status of water bodies corresponding to the water supply network, and construct the corresponding pollution degree based on the monitoring data. Evaluate whether the water quality is usable based on the water pollution degree, so that when the water quality is poor, it can be treated in time to ensure the safety of the water body. BRIEF DESCRIPTION OF THE DRAWINGS
[0058] Figure 1 A schematic diagram of a process flow of a smart water affairs dynamic supervision method based on digital twins provided in one embodiment of the present invention;
[0059] Figure 2 This is a schematic diagram of the structure of the smart water dynamic supervision platform based on digital twins of the present invention. DETAILED DESCRIPTION
[0060] The following will be combined with 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 of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0061] like Figure 1 As shown, in a specific embodiment, the present invention provides a method for dynamic supervision of smart water affairs based on digital twins, the method comprising the following steps:
[0062] Step 100: Monitor the operating status of the water supply network, and construct the load degree Ho of the water supply network during operation based on the data obtained from the monitoring. If the load degree of the water supply network exceeds expectations, collect relevant data of the water supply network and summarize and construct a water supply data set.
[0063] In the embodiment of the present application, step 100 specifically includes the following steps:
[0064] Step 101: When the water supply network is in operation, select several monitoring positions in the water supply network, and monitor the operation status of the water supply network at each monitoring position; obtain the following monitoring data including: obtain the liquid flow velocity or amount of the water pipe, channel or pump station to obtain the water flow Sn, obtain the operation noise Zn from its noise level, and after aggregating the obtained monitoring data, construct a water supply network operation data set.
[0065] Step 102: construct the load degree Ho of the water pipe network during operation from the water pipe network operation data set. If the load degree of the water pipe network exceeds expectations, send a data collection instruction to the outside.
[0066] In the embodiment of the present application, step 102 specifically includes:
[0067] Perform linear normalization on the water flow Sn and the operating noise Zn, and map the corresponding data values to the interval [0, 1] in the following way:
[0068]
[0069] Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1, i=1, 2, …, k, k is the number of monitoring locations; Sn i is the water flow at the ith monitoring location, Sn avg is the mean water flow rate, As its qualified standard value, Zn i is the operating noise at the ith monitoring position, Zn avg is the mean of the operating noise, is its qualified standard value;
[0070] Pre-set load thresholds based on load management expectations and historical data for the water supply network;
[0071] Based on the relationship between the load degree Ho and the load threshold, it is determined whether the load degree of the water pipe network exceeds expectations. If it exceeds expectations, a data collection instruction is issued to the outside.
[0072] Step 103: After receiving the data collection instruction, a sensor network and a data collection system are built for the water pipe network; various data in the water pipe network are collected, including: water volume, water pressure, water quality, pipe network status and other information, and data from the geographic information system, water management system and other related systems, as well as specification data and operation status data of the water pipe network are integrated; at this time, the collected data are summarized to construct a water data set.
[0073] When in use, combined with the contents of steps 101 to 103, when the water pipe network is in operation, the operation status of the water pipe network is monitored, and the monitoring data is compared with the pre-set standard value to determine whether there is an operating load in the water pipe network. The corresponding load degree Ho is constructed from the monitoring data, and the load degree Ho is used to determine whether the current water pipe network is large and whether it is within an acceptable range. If it is already in an overloaded operating state, it can be dealt with in a timely manner.
[0074] In combination with the contents in the above application and the contents in the prior art, after the water pipe network is installed and set up and enters the operating state, if it is under a large operating load for a long time, the water pipe network may have certain safety hazards. In order to avoid such safety hazards or failures, the existing smart water dynamic supervision methods usually collect current data of the water pipe network and analyze the collected data to determine whether there are currently any faults or hidden dangers. However, this method can only judge and analyze the current problem and cannot effectively cover the next situation. At the same time, when the water pipe network has already failed or its performance has declined, it is difficult to quickly provide a response plan and it is difficult to achieve rapid processing.
[0075] Step 200: Obtain a digital twin model of the water pipe network through training with the water pipe network and related data, monitor the communication status between the digital twin model of the water pipe network and the outside, and then construct a communication coefficient Txs based on the monitoring data. If the communication coefficient Txs is lower than expected, take communication enhancement measures.
[0076] In the embodiment of the present application, step 200 specifically includes the following steps:
[0077] Step 201: extract some data from the water data set as sample data, build an initial model by a machine learning algorithm, train the initial model with the sample data, and after testing and verification, obtain the trained digital twin model of the water pipe network, so as to simulate the structure, operation status and behavior of the water pipe network.
[0078] Step 202: After receiving the data input, the digital twin model of the water supply network will update the data. At this time, the communication status of the digital twin model of the water supply network is monitored at each monitoring node, the corresponding monitoring data is obtained, and the obtained communication data is summarized to construct a communication model communication data set.
[0079] Further, obtaining corresponding monitoring data includes:
[0080] Signal-to-noise ratio: The signal-to-noise ratio describes the ratio of signal to noise power, reflecting the strength and clarity of the signal. The higher the signal-to-noise ratio, the less interference the signal is subject to during transmission, and the better the communication quality.
[0081] Data delay: Data delay is the time required from the sender sending data to the receiver receiving the data. It includes transmission delay, propagation delay, and processing delay. The smaller the delay, the better the real-time performance of the communication.
[0082] Step 203: Construct the communication coefficients Txs of the model from the model communication data set.
[0083] Furthermore, the specific steps of constructing the communication coefficient Txs of the model from the model communication data set include:
[0084] Perform linear normalization on the signal-to-noise ratio Hp and data delay Bp, and map the corresponding data values to the interval [0, 1] as follows:
[0085]
[0086] Among them, k2 and k1 are weight coefficients, 0≤k1≤1, 0≤k2≤1, and k1+k2=1, Bp i The ith data delay, Hp i is the i-th signal-to-noise ratio, i=1, 2,…n, and n is the number of monitoring nodes.
[0087] Step 204: After obtaining the communication coefficient Txs, the information update capability of the digital twin model is verified and evaluated using the communication coefficient Txs, and a communication threshold is pre-set based on expectations for model communication management; if the communication coefficient Txs is lower than expected, that is, the communication coefficient is lower than the communication threshold, it means that the model's current communication capability with the outside world is poor. At this time, the current data communication capability is strengthened, for example, by increasing communication signals, reducing communication delays, or increasing bandwidth.
[0088] When in use, combined with the contents of steps 201 to 204, a digital twin model of the water pipe network is trained and established based on the collected data. Through the digital twin model of the water pipe network, the operating status of the water pipe network can be monitored and predicted, and the operating status of the water pipe network can be understood and perceived in advance. If there are any problems, they can be dealt with in a timely manner.
[0089] When the digital twin model of the water pipe network is in operation, the data on its communication status with the outside world is collected and analyzed, and then the communication coefficient Txs is constructed. At this time, the communication coefficient Txs can be used to comprehensively evaluate and analyze the model update and operation status. If it is judged that the current communication status of the model is poor, the communication capabilities of the model and corresponding equipment can be guaranteed in time.
[0090] Step 300: Perform simulation analysis on the water pipe network, and construct the stability Wo of the water pipe network based on the test data. If the stability Wo does not exceed the stability threshold, a maintenance instruction is issued to the outside.
[0091] In the embodiment of the present application, step 300 specifically includes the following steps:
[0092] Step 301: After the water pipe network is in operation, the water pipe network digital twin model is used to simulate and analyze the water pipe network under the current operating environment, and the corresponding operating data is obtained, and the load degree Ho on each analysis node is constructed based on the operating data.
[0093] Furthermore, the load degree Ho is generated as follows:
[0094] Perform linear normalization on the water flow Sn and the operating noise Zn, and map the corresponding data values to the interval [0, 1] in the following way:
[0095]
[0096] Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1, i=1, 2, …, k, k is the number of monitoring locations; Sn i is the water flow at the ith monitoring location, Sn avg is the mean water flow rate, As its qualified standard value, Zn i is the operating noise at the ith monitoring position, Zn avg is the mean of the operating noise, is its qualified standard value;
[0097] Step 302: Summarize the acquired load degrees Ho to construct a water service operation data set.
[0098] Step 303: construct the stability Wo of the water pipe network from the water operation data set. If the obtained stability Wo exceeds the stability threshold, issue a water quality monitoring instruction to the outside; otherwise, issue a maintenance instruction to the outside.
[0099] Furthermore, the stability of the water supply network Wo is constructed from the water supply operation data set as follows:
[0100]
[0101] Where i = 1, 2, ..., p, p is the number of analysis nodes; Ho i is the load on the i-th analysis node, Ho avg is the mean value of load degree;
[0102] Based on the management expectations of the water supply network based on historical data, a stability threshold is set in advance; if the obtained stability Wo exceeds the stability threshold, it can be judged that the current operation of the water supply network is relatively stable and can still maintain a normal operating state, and its reliability is relatively high. Therefore, the current usage status can continue to be maintained. At this time, a water quality monitoring instruction is issued to the outside; if the opposite is true, it means that the reliability of the water supply network is lower than expected, and the water supply network needs to be maintained, and a maintenance instruction is issued to the outside.
[0103] When in use, combined with the contents in steps 301 and 303, after continuously predicting and obtaining several load degrees Ho, the stability Wo is constructed according to the changes in the load degree Ho, and the stability of the water pipe network is evaluated based on the constructed stability Ho. If the water pipe network is often operated under higher load conditions, it means that the water pipe network has limited carrying capacity for work tasks, and the operating stability and safety of the water pipe network may be affected to a certain extent. At this time, timely inspection and maintenance of the water pipe network can ensure the safe operation of the water pipe network.
[0104] Step 400: After the water pipe network optimization knowledge graph provides an optimization plan for the water pipe network, the output optimization plan is screened according to the optimization degree Yo constructed by a number of stabilities Wo, and the water pipe network is optimized by the screened optimization plan.
[0105] In the embodiment of the present application, step 400 specifically includes the following steps:
[0106] Step 401: After receiving the maintenance instruction, the digital twin model of the water pipe network predicts the operating status of the water pipe network and obtains the corresponding operating prediction data; after setting the optimization standard for each monitoring data of the water pipe network, the operating prediction data is identified to obtain the corresponding optimization features.
[0107] Step 402: Taking water pipe network as the target word, after deep retrieval and building corresponding entity relationships, a water pipe network optimization knowledge graph is pre-constructed, and the trained matching model is used to provide an optimization solution for the water pipe network based on the correspondence between the optimization solution and the optimization features of the water pipe network.
[0108] When in use, if it is predicted that the water pipe network will be running at high load for a long time, the operating status data of the water pipe network will be predicted and obtained, and then the pre-built water pipe network optimization knowledge graph will give a corresponding maintenance plan for the water pipe network, so that the water pipe network can be quickly optimized when necessary.
[0109] Step 403: After obtaining the output optimization plan, the optimization plan is tested using the digital twin model of the water pipe network. A test cycle including several test nodes is set up, and corresponding test data is obtained at each test node. The stability Wo of the water pipe network is reconstructed based on the test data.
[0110] The optimization degree Yo is constructed from several stability degrees Wo. After obtaining the stability degree Wo on each prediction node, the stability degrees Wo before and after the test optimization scheme are aligned one by one according to the following method:
[0111]
[0112] Among them, Tz i is the median stability value at the i-th test node, Tz avg is its mean, i is the test node number, i = 1, 2, ..., n, n is the number of test nodes, AWo i and BWo i are the stability of the i-th node before and after optimization, AWo avg and BWo avg is the corresponding mean;
[0113] Step 404: Based on the expected optimization effect of the optimization scheme and the corresponding historical data, an optimization degree threshold is pre-constructed; if the optimization degree Yo is lower than the optimization degree threshold, it means that the current optimization scheme has failed to achieve the desired effect, and the optimization algorithm constructs a scheme optimization model, and uses the scheme optimization model to optimize the optimization scheme, and obtains the optimized optimization scheme, which is used as the target scheme; if the optimization degree is not lower than the optimization degree threshold, it means that the optimization scheme is feasible, and the corresponding optimization scheme is used as the target scheme.
[0114] When in use, combined with the contents in steps 401 to 404, on the basis of obtaining the optimization plan, combined with the simulation of the digital twin model of the water pipe network, and based on the optimization degree Yo, each optimization plan can be screened. When optimizing the water pipe network, the optimization effect can be guaranteed, thereby enabling the water pipe network to always be in a better operating state.
[0115] Step 500: Collect water quality data related to the water pipe network, and construct the pollution degree Rxs of the water body from the water quality data. If the pollution degree exceeds the pollution threshold, issue a water quality adjustment instruction to the outside.
[0116] In the embodiment of the present application, step 500 specifically includes the following steps:
[0117] Step 501: After the target solution is executed to optimize the water pipe network, a monitoring cycle including several monitoring nodes is set to monitor the pollution status of the water body within the monitoring cycle and obtain corresponding monitoring data, including turbidity, pH value, dissolved oxygen, ammonia nitrogen, etc., to monitor whether the water quality meets the safe drinking standard, and to construct a water body monitoring data set after aggregation;
[0118] Step 502: Construct the pollution level Rxs of the water body from the monitoring data in the water body monitoring data set, wherein the turbidity Uo is normalized according to the following method:
[0119]
[0120] Where i = 1, 2, ..., m, m is the number of monitoring nodes, Uo i is the turbidity of the water body at the i-th monitoring node, Uo avg is the corresponding mean, is the qualified reference value of turbidity; weight coefficient: 0≤F1≤1, 0≤F2≤1, and F1+F2=1; the weight coefficient is obtained by referring to the hierarchical analysis method;
[0121] Step 503: Pre-set the pollution threshold based on historical data and expectations for water quality management; if the obtained pollution level Rxs exceeds the pollution threshold, it means that the current water quality is abnormal and there is a certain amount of water pollution, and the water quality needs to be controlled in time. At this time, a water quality adjustment instruction is issued to the outside; after receiving the water quality adjustment instruction, personnel can be arranged to process the water quality in time to ensure the water quality.
[0122] When in use, combined with the contents of steps 501 and 503, after the water pipe network is optimized and in a better operating state, the state of the water body corresponding to the water pipe network is monitored, and the corresponding pollution degree is constructed from the monitoring data. The usability of the water quality is evaluated based on the water pollution degree, so that when the water quality is poor, it can be treated in time to ensure the safety of the water body.
[0123] See also Figure 2 In the embodiment of the present application, a smart water affairs dynamic supervision platform based on digital twin is provided, including:
[0124] The load monitoring unit is used to monitor the operation status of the water pipe network and construct the load degree Ho of the water pipe network during operation based on the data obtained from the monitoring. If the load degree of the water pipe network exceeds the expectation, the relevant data of the water pipe network is collected and summarized to construct a water data set;
[0125] The communication condition evaluation unit is used to obtain the digital twin model of the water pipe network through training of the water pipe network and related data, monitor the communication status between the digital twin model of the water pipe network and the outside, and then construct the communication coefficient Txs from the monitoring data. If the communication coefficient Txs is lower than expected, communication enhancement measures are taken;
[0126] The stability analysis unit is used to simulate and analyze the water pipe network and construct the stability Wo of the water pipe network based on the test data. If the stability Wo does not exceed the stability threshold, a maintenance instruction is issued to the outside.
[0127] The scheme optimization unit is used to provide an optimization scheme for the water pipe network based on the water pipe network optimization knowledge graph, screen the output optimization scheme according to the optimization degree Yo constructed by a number of stability degrees Wo, and optimize the water pipe network with the screened optimization scheme;
[0128] The pollution warning unit is used to collect water quality data related to the water supply network, and construct the pollution degree Rxs of the water body based on the water quality data. If the pollution degree exceeds the pollution threshold, a water quality adjustment instruction is issued to the outside.
[0129] The Analytic Hierarchy Process (AHP) is a decision-making method that decomposes decision-related elements into levels such as goals, criteria, and plans, and conducts qualitative and quantitative analysis on this basis. It is particularly suitable for dealing with target systems with hierarchical and staggered evaluation indicators, and when the target value is difficult to describe quantitatively, the AHP is an effective decision-making tool.
[0130] The core of the hierarchical analysis method is to decompose the decision problem into multiple levels to form a hierarchical structure, which usually includes the target level, the criterion level, the sub-criterion level and the solution level. By solving the eigenvector of the judgment matrix, the priority weight of each element of each level to an element of the previous level is obtained, and the final weight of each alternative solution to the total goal is recursively merged by the weighted sum method, so as to find the optimal solution.
[0131] The construction method of the water pipe network optimization knowledge graph is a comprehensive process that combines data collection, processing, analysis and knowledge representation. The following is a detailed construction step:
[0132] Data collection and preprocessing:
[0133] Data source determination: First, it is necessary to determine the data source related to the water supply network, which may include network design drawings, historical operation data, real-time monitoring data, maintenance records, etc.
[0134] Data collection: Use crawlers, API interfaces and other technical means to collect the required data from the above data sources.
[0135] Data preprocessing: Clean the collected data, including format conversion, deduplication, outlier processing, missing value filling and other operations to ensure the accuracy and consistency of the data.
[0136] Entity recognition and attribute extraction: Entity recognition: Identify key entities in the water supply network, such as pump stations, valves, pipes, nodes, etc. Attribute extraction: Extract relevant attributes of these entities, such as location, size, material, flow, pressure, etc.
[0137] Relation extraction and knowledge modeling:
[0138] Relationship extraction: Analyze the relationship between entities, such as the connection relationship between pipelines, the control relationship between pump stations and valves, etc. Knowledge modeling: Based on the extracted entities and relationships, use methods such as graph models to build knowledge models to form a knowledge network for water pipe networks.
[0139] Knowledge representation and storage: Knowledge representation: Represent the knowledge model in the form of a graph for easy understanding and visualization. Knowledge storage: Select an appropriate graph database or knowledge graph storage technology to store the knowledge graph for subsequent query and analysis.
[0140] Application of optimization algorithms and strategies: Pipeline network optimization analysis: Based on the constructed knowledge graph, optimization algorithms and strategies are used to optimize and analyze the water pipe network to identify potential problems and bottlenecks. Strategy formulation and implementation: According to the optimization analysis results, corresponding pipe network optimization strategies are formulated, such as adjusting pump station operating parameters, optimizing valve control strategies, etc., and implemented in the actual pipe network.
[0141] Continuous update and maintenance: Data update: With the operation and changes of the water supply network, data needs to be updated regularly to ensure the timeliness and accuracy of the knowledge graph. Knowledge graph maintenance: Regularly maintain and optimize the knowledge graph to maintain its performance and stability.
[0142] By constructing a knowledge graph for water pipe network optimization, we can provide decision support for water management personnel and improve the operation efficiency and management level of water pipe networks. At the same time, we can also provide strong support for the planning, design, and transformation of water pipe networks, and promote the sustainable development of the water industry.
[0143] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0144] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.
[0145] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.
[0146] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.
Claims
1. A dynamic supervision method for smart water services based on digital twins, characterized by: include, Monitor the operation status of the water supply network, and construct the load degree Ho of the water supply network during operation based on the data obtained from the monitoring. If the load degree of the water supply network exceeds expectations, collect the relevant data of the water supply network and summarize and construct a water supply data set; The digital twin model of the water pipe network is obtained by training the water pipe network and related data, the communication status between the digital twin model of the water pipe network and the outside is monitored, and the communication coefficient Txs is then constructed based on the monitoring data. If the communication coefficient Txs is lower than expected, communication enhancement measures are taken. The water pipe network is simulated and analyzed, and the stability Wo of the water pipe network is constructed based on the test data. If the stability Wo does not exceed the stability threshold, a maintenance instruction is issued to the outside. The construction method of the stability Wo is as follows: Where i = 1, 2, ..., p, p is the number of analysis nodes; Ho i is the load on the i-th analysis node, Ho avg is the mean value of load degree; After the water pipe network optimization knowledge graph gives an optimization plan for the water pipe network, the output optimization plan is screened according to the optimization degree Yo constructed by a number of stability degrees Wo, and the water pipe network is optimized by the screened optimization plan; Collect water quality data related to the water pipe network, and construct the pollution degree Rxs of the water body from the water quality data. If the pollution degree Rxs exceeds the pollution threshold, issue a water quality adjustment instruction to the outside. The communication coefficient Txs of the model is constructed from the model communication data set in the following way: linear normalization is performed on the signal-to-noise ratio Hp and the data delay Bp, and the corresponding data values are mapped to the interval [0,1] according to the following method: Among them, k2 and k1 are weight coefficients, 0≤k1≤1, 0≤k2≤1, and k1+k2=1, Bp j The jth data delay, Hp j is the jth signal-to-noise ratio, j = 1, 2, ... n, n is the number of monitoring nodes; The water pipe network is simulated and analyzed, and the stability Wo of the water pipe network is constructed based on the test data. If the stability Ho does not exceed the stability threshold, a maintenance instruction is issued to the outside, including: The digital twin model of the water supply network is used to simulate and analyze the water supply network, and the corresponding operation data is obtained. The load degree Ho on each analysis node is then constructed based on the operation data. The load degree Ho is generated as follows: Perform linear normalization on the water flow Sn and the operating noise Zn, and map the corresponding data values to the interval [0,1] in the following way: Weight coefficient: 0≤β≤1, 0≤α≤1, and α+β=1, r=1, 2, …, k, k is the number of monitoring locations; Sn r is the water flow at the rth monitoring location, Sn avg is the mean water flow rate, As its qualified standard value, Zn r is the operating noise at the rth monitoring position, Zn avg is the mean of the operating noise, is its qualified standard value; Summarize the acquired load degree Ho to construct a water service operation data set; The stability Wo of the water supply network is constructed from the water supply operation data set. If the obtained stability Wo exceeds the stability threshold, a water quality monitoring instruction is issued to the outside; otherwise, a maintenance instruction is issued to the outside.
2. According to a digital twin-based smart water affairs dynamic supervision method according to claim 1, it is characterized by: The monitoring of the operation status of the water pipe network and the construction of the load degree Ho of the water pipe network during operation from the data obtained by the monitoring, if the load degree of the water pipe network exceeds the expectation, the relevant data of the water pipe network are collected and the water data set is constructed in aggregating, including: Select several monitoring locations in the water pipe network, monitor the operation status of the water pipe network at each monitoring location, obtain the water flow Sn and operation noise Zn, and after aggregating the acquired monitoring data, construct a water pipe network operation data set; The load degree Ho of the water pipe network during operation is constructed from the water pipe network operation data set; if the load degree of the water pipe network exceeds expectations, a data collection instruction is issued to the outside; After receiving the data collection instruction, a sensor network and data collection system are built for the water supply network to collect various types of data in the water supply network, and the collected data are summarized to construct a water supply data set.
3. According to a digital twin-based smart water affairs dynamic supervision method according to claim 2, it is characterized by: The water pipe network and related data are used to train the water pipe network digital twin model, monitor the communication status between the water pipe network digital twin model and the outside, and then construct the communication coefficient Txs from the monitoring data. If the communication coefficient Txs is lower than expected, communication enhancement measures are taken, including: Extract some data from the water service data set as sample data, build an initial model using a machine learning algorithm, train the initial model using the sample data, and obtain a trained digital twin model of the water service network; After receiving the data input, the communication status of the digital twin model of the water pipe network is monitored to obtain the corresponding monitoring data, including the signal-to-noise ratio and data delay; and the acquired communication data is aggregated to construct a communication data set of the communication model; The communication coefficient Txs of the model is constructed from the model communication data set; After obtaining the communication coefficient Txs, the information update capability of the digital twin model is verified and evaluated with the communication coefficient Txs. The communication threshold is set in advance based on the expectations for model communication management. If the communication coefficient Txs is lower than expected, it means that the model’s current communication capability with the outside world is poor, and the current data communication capability should be strengthened.
4. According to a digital twin-based smart water affairs dynamic supervision method according to claim 1, it is characterized by: The optimization scheme provided for the water pipe network by the water pipe network optimization knowledge graph includes: After receiving the maintenance instruction, the digital twin model of the water pipe network predicts the operation status of the water pipe network and obtains the corresponding operation prediction data; identifies the operation prediction data and obtains the corresponding optimization features; Taking water pipe network as the target word, a water pipe network optimization knowledge graph is pre-constructed. According to the correspondence between the optimization scheme and optimization features of the water pipe network, the water pipe network optimization knowledge graph provides an optimization scheme for the water pipe network.
5. According to claim 4, a method for dynamic supervision of smart water affairs based on digital twins is characterized in that: The optimization scheme outputted by constructing the optimization degree Yo according to a plurality of stability degrees Wo is screened, and the water pipe network is optimized by the screened optimization scheme, including: After obtaining the output optimization plan, the water pipe network digital twin model is used to test the optimization plan, and the corresponding test data is obtained at each test node. The stability Wo of the water pipe network is reconstructed based on the test data; The optimization degree Yo is constructed by several stabilities Wo. If the optimization degree Yo is lower than the optimization degree threshold, the optimization algorithm constructs a solution optimization model, and the solution optimization model is used to optimize the optimization solution to obtain the optimized solution, which is used as the target solution. If the optimization degree is not lower than the optimization degree threshold, the corresponding optimization solution is used as the target solution.
6. According to a digital twin-based smart water affairs dynamic supervision method according to claim 5, it is characterized by: The optimization method of Yo is as follows: after obtaining the stability Wo on each prediction node, the stability Wo before and after the test optimization scheme is aligned one by one, according to the following formula: Among them, Tz i is the median stability value at the i-th test node, Tz avg is its mean, i is the test node number, i=1, 2,…,n, n is the number of test nodes, AWo i and BWo i are the stability of the i-th node before and after optimization, AWo avg and BWo avg is the corresponding mean.
7. The method for dynamic supervision of smart water affairs based on digital twin according to claim 1 is characterized in that: The water quality data of water bodies related to the water pipe network is collected, and the pollution degree Rxs of the water body is constructed from the water quality data. If the pollution degree Rxs exceeds the pollution threshold, a water quality adjustment instruction is issued to the outside, including: After optimizing the water pipe network by executing the target plan, the pollution status of the water body is monitored during the monitoring period to obtain the corresponding monitoring data; after summarizing, a water body monitoring data set is constructed; The pollution degree Rxs of the water body is constructed from the monitoring data in the water body monitoring data set, wherein the turbidity Uo is normalized according to the following formula: Among them, □=1, 2, …, m, m is the number of monitoring nodes, Uo □ is the turbidity of the water body at the □th monitoring node, Uo avg is the corresponding mean, It is the qualified reference value of turbidity; weight coefficient: 0≤F1≤1, 0≤F2≤1, and F1+F2=1; Based on historical data and expectations for water quality management, a pollution threshold is set in advance; if the obtained pollution level Rxs exceeds the pollution threshold, it means that the current water quality is abnormal, and a water quality adjustment instruction is issued to the outside.
8. A smart water affairs dynamic supervision platform based on digital twins, characterized by: include: The load monitoring unit is used to monitor the operation status of the water pipe network and construct the load degree Ho of the water pipe network during operation based on the data obtained from the monitoring. If the load degree of the water pipe network exceeds the expectation, the relevant data of the water pipe network is collected and summarized to construct a water data set; The communication condition evaluation unit is used to obtain the digital twin model of the water pipe network through training of the water pipe network and related data, monitor the communication status between the digital twin model of the water pipe network and the outside, and then construct the communication coefficient Txs from the monitoring data. If the communication coefficient Txs is lower than expected, communication enhancement measures are taken; The stability analysis unit is used to simulate and analyze the water pipe network and construct the stability Wo of the water pipe network based on the test data. If the stability Wo does not exceed the stability threshold, a maintenance instruction is issued to the outside. The scheme optimization unit is used to provide an optimization scheme for the water pipe network based on the water pipe network optimization knowledge graph, screen the output optimization scheme according to the optimization degree Yo constructed by a number of stability degrees Wo, and optimize the water pipe network with the screened optimization scheme; The pollution warning unit is used to collect water quality data related to the water supply network, and construct the pollution degree Rxs of the water body based on the water quality data. If the pollution degree exceeds the pollution threshold, a water quality adjustment instruction is issued to the outside.
