NOVEL SYSTEM AND METHOD FOR MONITORING AND MANAGEMENT OF FLUID INFRASTRUCTURE NETWORKS - Patent application

Abstract

A fluid infrastructure management system and method are described. An exemplary method includes: 1 obtaining, inside a server and a remote telemetry unit, a digital twin model providing one or more estimates of flow attributes inside a fluid infrastructure item; 2 receiving a prediction of environmental conditions around an area surrounding a sensor; 3 estimating the flow attributes at the server using one of the predictions and the digital twin model to generate one or more server estimated flow attribute values; 4 using a sensor to obtain sensor measurements of flow attributes at a remote telemetry unit, and if the server does not receive the sensor measurements from the remote telemetry unit, reporting one or more server estimated flow attribute values ​​as being the sensor measurements or as being estimates of the sensor measurements.

Description

[Technical field]

[0001] This patent application claims priority to U.S. Provisional Patent Application No. 63 / 254,175, filed October 11, 2021, which is incorporated by reference herein for all purposes.

[0002] The present teachings and arrangements relate generally to novel systems and methods for management of fluid infrastructure networks, and more particularly, to using digital twins to enable management by monitoring flow attributes within a fluid infrastructure network. [Background technology]

[0003] Fluid infrastructure networks include transport vehicles, pumps, bypasses, storage tanks and wells, and treatment units to deliver or treat fluids or water. Examples of fluid networks are wastewater systems, groundwater well systems, aquifers, waterway networks, and stormwater harvesting systems. These networks are designed to collect and deliver fluid streams and sometimes treat the delivered fluid streams. As an example, infrastructure networks dedicated to collecting and delivering treated water for drinking and agricultural and industrial uses are integral to human life, so they must be properly managed. Unfortunately, whether or not fluid treatment is required, managing these networks is time-consuming, laborious, and a financial burden.

[0004] Therefore, novel systems and processes for the management of fluid infrastructure network systems are needed. Summary of the Invention

[0005] To achieve the above, the present compositions and teachings provide novel systems and methods that use digital twins to enable management by monitoring flow attributes within fluid infrastructure networks.

[0006] In one aspect, the present teachings provide a method for managing a fluid infrastructure network. One such exemplary management method includes: (1) obtaining a copy of a digital twin model at a server; (2) receiving a prediction of environmental conditions around an area surrounding a sensor during an instance in time or during a time period; (3) estimating flow attributes inside the fluid infrastructure item during the instance in time or during the time period using the prediction of the environmental conditions and the copy of the digital twin model at the server to generate one or more estimated flow attribute values ​​of the server; (4) receiving or obtaining sensor measurements of the flow attributes inside the fluid infrastructure item from a sensor inside a remote telemetry unit; and (5) reporting one or more of the estimated flow attribute values ​​of the server as being sensor measurements or estimates of sensor measurements during the instance in time or during the time period if the server does not receive sensor measurements from the remote telemetry unit after the instance in time or after the time period. In a preferred embodiment of the present teachings, the above-mentioned reporting element is executed when the remote telemetry unit determines that the absolute value of the difference between the sensor measurement and one of the remote telemetry unit's estimated flow attribute values ​​is equal to or less than a predefined threshold tolerance value, or when the sensor measurement during an instance in time or during a time period falls within a predefined confidence interval of multiple estimated flow attribute values ​​of the remote telemetry unit of the fluid infrastructure item.

[0007] In another aspect, the present teachings provide a method for managing another type of fluid infrastructure network, where a remote telemetry unit and a sensor are located remotely relative to a server, and the remote telemetry unit is communicatively coupled to a sensor that measures a flow attribute inside a fluid infrastructure item, whereby the remote telemetry unit receives sensor measurements from the sensor.

[0008] Further, an exemplary method of this aspect includes (1) obtaining, inside each of the servers and inside the remote telemetry units, a digital twin model that provides one or more estimates of flow attributes inside the fluid infrastructure item during the instance in time or during the time period; (2) receiving a prediction of environmental conditions around an area surrounding the sensor during the instance in time or during the time period; (3) estimating, at the remote telemetry units, flow attributes inside the fluid infrastructure item during the instance in time or during the time period using the prediction of the environmental conditions and one of the digital twin models to generate estimated flow attribute values ​​for the one or more remote telemetry units; and (4) estimating, at the servers, flow attributes during the instance in time or during the time period using the prediction of the environmental conditions and one of the digital twin models to generate estimated flow attribute values ​​for the one or more servers. (5) obtaining, at the remote telemetry unit, a sensor measurement of the flow attribute inside the fluid infrastructure item using a sensor; (6) comparing, at the remote telemetry unit, the sensor measurement with one or more of the estimated flow attribute values ​​of the remote telemetry unit; and (7) reporting, during the instance in time or during the time period, one or more of the estimated flow attribute values ​​to the server as being a sensor measurement or being an estimate of the sensor measurement if an absolute value of a difference between the sensor measurement and one of the estimated flow attribute values ​​of the remote telemetry unit is equal to or less than a predefined threshold tolerance value or if the sensor measurement falls within a predefined confidence interval of the plurality of estimated flow attribute values ​​of the remote telemetry unit during the instance in time or during the time period.

[0009] However, if the absolute value of the difference between the sensor measurement value and one of the estimated flow attribute values ​​of the remote telemetry unit is greater than a predefined threshold tolerance value, or if the sensor measurement value falls outside a predefined confidence interval of the multiple estimated flow attribute values ​​of the remote telemetry unit during the instance in time or during the time period, the above-mentioned management method of the present teachings preferably further includes transmitting the sensor measurement value during the instance in time or during the time period from the remote telemetry unit to a server.

[0010] The method of this aspect then preferably includes an element of receiving, at the server, after the passage of the instance in time or the passage of the time period, information regarding realized environmental conditions realized in the area surrounding the sensor during the instance in time or the time period.

[0011] Based on information about the realized environmental conditions and the digital twin model residing at the server, the management method of the present teachings determines new estimated flow attribute values ​​for one or more servers during an instance in time or for a time period, which enables a comparison of sensor measurements at the server to one or more of the server's new estimated flow attribute values.

[0012] The management method of the present teachings then proceeds to an element of maintaining a digital twin model at the server if the absolute value of the difference between the sensor measurement and one or more of the server's new estimated flow attribute values ​​is equal to or greater than a check predefined threshold tolerance value, or if the sensor measurement falls outside a check predefined confidence interval of the server's multiple new estimated flow attribute values ​​during an instance in time or during a time period.

[0013] However, if the absolute value of the difference between the sensor measurement and one of the server's new estimated flow attributes is less than a predefined threshold tolerance value, or if the sensor measurement falls within a predefined confidence interval of the server's new estimated flow attribute values ​​during an instance in time or during a time period, the method of the present teachings preferably proceeds to obtain a new digital twin model different from the digital twin model within each of the server and the remote telemetry unit. In other words, the original digital twin model is updated, recalibrated, or reconfigured.

[0014] In yet another aspect, the present teachings provide yet another type of fluid infrastructure network management method. In this aspect, a remote telemetry unit is communicatively coupled to a sensor that measures a flow attribute inside a fluid infrastructure item, whereby the remote telemetry unit receives sensor measurements of the flow attribute from the sensor. One such exemplary method of this aspect includes: (1) obtaining, inside the remote telemetry unit, a digital twin model that provides one or more estimates of a flow attribute inside the fluid infrastructure item during an instance in time or during a time period; (2) receiving a forecast of precipitation expected to be experienced in an area surrounding the sensor at the instance in time or during the time period; (3) estimating, at the remote telemetry unit, one or more remote telemetry unit estimated flow attribute values ​​for the fluid infrastructure item based on the forecast of precipitation and the digital twin model during the instance in time or during the time period; and (4) receiving, at the remote telemetry unit, sensor measurements during the instance in time or during the time period and estimating the sensor during the instance in time or during the time period. (5) receiving an amount of realized precipitation realized in the enclosing area; (6) comparing the sensor measurement value with one or more of the estimated flow attribute values ​​of the remote telemetry unit; and (7) determining, at the remote telemetry unit, one or more new remote telemetry unit estimated flow attribute values ​​for the instance in time or for the time period using the digital twin and the amount of realized precipitation if an absolute value of a difference between the sensor measurement value and one of the estimated flow attribute values ​​of the remote telemetry unit is greater than a predefined threshold tolerance value or if the sensor measurement value falls outside a predefined confidence interval of the multiple estimated flow attribute values ​​of the remote telemetry unit during the instance in time or during the time period; and (7) comparing, at the remote telemetry unit, the sensor measurement value with one or more of the estimated flow attribute values ​​of the new remote telemetry unit.(7) maintaining the digital twin model in the server when an absolute value of a difference between the sensor measurement value and one of the estimated flow attribute values ​​of the new remote telemetry unit is equal to or greater than a check predefined threshold tolerance value, or when the sensor measurement value falls outside a check predefined confidence interval of the estimated flow attribute values ​​of the multiple new remote telemetry units during an instance in time or during a time period;

[0015] In yet another aspect, the present composition provides a fluid infrastructure network management system.One such exemplary management system includes: (1) a plurality of sensors, each sensor designed to measure a flow attribute inside a fluid infrastructure item that is part of a fluid infrastructure network; and (2) a plurality of remote telemetry units, each remote telemetry unit communicatively coupled to at least one of the sensors, the remote telemetry unit including: (i) a battery for providing power; (ii) a sensor module communicatively coupled to the sensors and for receiving sensor measurements; (iii) a memory for storing the sensor measurements and one of the digital twin models providing estimated flow attribute values ​​of the one or more remote telemetry units for the infrastructure item during an instance in time or during a time period; (iv) a communication module connected to the network and designed to transmit information from the remote telemetry unit to the network; and (v) a microprocessor programmed to implement instructions, the instructions including: (a) determining whether or not the remote telemetry unit is configured to measure a flow attribute inside a fluid infrastructure item that is part of a fluid infrastructure network; (b) receiving, during the instance in time or during the time period, a prediction of environmental conditions around an area surrounding the sensor using the communications module; (b) estimating, during the instance in time or during the time period, estimated flow attribute values ​​of one or more remote telemetry units inside the fluid infrastructure item using the prediction of the environmental conditions and the digital twin model; (c) obtaining sensor measurements inside the fluid infrastructure item using the sensor; (d) comparing the sensor measurements to one or more of the estimated flow attribute values ​​of the remote telemetry unit; and (e) communicating the sensor measurements between the instance in time or during the time period from the remote telemetry unit to the network if an absolute value of a difference between the sensor measurement and one of the estimated flow attribute values ​​of the remote telemetry unit is greater than a predefined threshold tolerance value or if the sensor measurement falls outside a predefined confidence interval of the plurality of estimated flow attribute values ​​of the remote telemetry unit during the instance in time or during the time period.

[0016] An exemplary management system also includes a server communicatively coupled to a plurality of remote telemetry units over a network and having a plurality of digital twin models stored therein, each digital twin model being a copy of the digital twin model stored in each of the remote telemetry units, each of the remote telemetry units and each of the sensors being remote with respect to the server, the server being programmed to implement instructions, the instructions including: (a) receiving, over the network from the remote telemetry units, sensor measurements during an instance in time or during a time period; and (b) after the instance in time or after the time period, transmitting, at the server, information regarding realized environmental conditions that were realized in an area surrounding the sensor during the instance in time or during the time period. (c) determining new estimated flow attribute values ​​of one or more servers inside the fluid infrastructure item during the instance in time or during the time period based on the information about the realized environmental conditions and the digital twin model residing in the server; (d) comparing, in the server, the sensor measurement value with one or more of the server's new estimated flow attribute values; and (e) maintaining, in the server, the digital twin model if an absolute value of a difference between the sensor measurement value and one of the server's new estimated flow attribute values ​​is equal to or greater than a check predefined threshold tolerance value or if the sensor measurement value falls outside a check predefined confidence interval of the server's multiple new estimated flow attribute values ​​during the instance in time or during the time period.

[0017] In yet another aspect, the present configuration provides another type of fluid infrastructure network management system. One such exemplary system includes: (1) a plurality of sensors, each sensor designed to generate sensor measurements inside a fluid infrastructure item; (2) a plurality of remote telemetry units, each remote telemetry unit communicatively coupled to at least one of the sensors and having one of the digital twin models stored therein; (3) a server communicatively coupled to the plurality of remote telemetry units and having the plurality of digital twin models stored therein, such that one digital twin model on the server is substantially similar to the digital twin model stored on the remote telemetry unit; and (4) a network through which the plurality of remote telemetry units are communicatively coupled to the server, for (a) receiving a prediction of environmental conditions around an area surrounding at least one of the sensors during an instance in time or during a period of time. (b) estimating, at the server, flow attributes inside the fluid infrastructure item between instances in time or for a time period using one of the predictions of the environmental conditions and the digital twin model to generate estimated flow attribute values ​​for the one or more servers; (c) receiving or obtaining, at a remote telemetry unit, sensor measurements of the flow attributes inside the fluid infrastructure item from the sensor; and (d) reporting, after a lapse of the instance in time or after a lapse of the time period, one or more of the estimated flow attribute values ​​of the server as being a sensor measurement between the instances in time or for a time period or being an estimate of the sensor measurement if the server does not receive a sensor measurement from the remote telemetry unit.

[0018] The systems and methods of operation and advantageous construction resulting from the present teachings and arrangements, however, together with additional objects and advantages thereof, will best be understood from the following description of specific embodiments when read in conjunction with the accompanying drawings. [Brief description of the drawings]

[0019] [Figure 1] A diagram illustrating a fluid infrastructure network management system according to one embodiment of the present configuration, including a server communicatively coupled to a remote telemetry unit, the remote telemetry unit communicatively coupled to a sensor. [Diagram 2] FIG. 2 illustrates the fluid infrastructure network management system of FIG. 1 including a server communicatively coupled to a plurality of remote telemetry units, and for ease of illustration, illustrates the configuration of the server having stored therein a digital twin model that estimates flow attributes as a function of time inside the fluid infrastructure items monitored by the remote telemetry units. [Diagram 3] FIG. 13 illustrates another fluid infrastructure network management system according to another embodiment of the present configuration, including one copy of the digital twin model stored on a server and another copy of the digital twin model stored on a remote telemetry unit. [Figure 4] FIG. 13 illustrates another fluid infrastructure network management system according to another embodiment of the present configuration that uses a digital twin model of a server to generate one or more estimated flow attribute values ​​of a remotely located fluid infrastructure item to facilitate management of the item. [Diagram 5] FIG. 13 illustrates another fluid infrastructure network management system in accordance with another embodiment of the present configuration that uses a digital twin model, one copy at a server and another copy at a remote telemetry unit, to determine whether a prediction of an environmental condition used by the digital twin model in generating an estimated flow attribute value is an outlier and should be maintained, or whether the digital twin model no longer provides reliable estimates and should be updated, recalibrated, or reconfigured. [Figure 6] 1 is a flowchart describing an infrastructure management method according to one embodiment of the present teachings that does not require consuming energy and transmitting sensor measurements of flow attributes of remotely located fluid infrastructure items to a server. [Figure 7] 11 is a flowchart describing another infrastructure management method according to another embodiment, in which a digital twin model reliably estimates attribute flow values ​​inside a remotely located fluid infrastructure item and establishes whether the digital twin model should be maintained in a server and a remote telemetry unit. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present teachings and configurations. However, it will be apparent to one of ordinary skill in the art that the present teachings and configurations may be practiced without being limited to some or all of these specific details. In other instances, well-known process steps have not been described in detail in order not to unnecessarily obscure the present teachings and configurations.

[0021] Effective monitoring systems and methods for the management of fluid infrastructure networks are important for implementing real-time decision support systems. To this end, the present configuration and teachings recognize that while advances in sensing, battery technology, and cellular communications have made the installation and operation of such monitoring systems more cost-effective and robust, little has been done to extend their battery life. In particular, when the transmission of real-time sensor measurements is considered paramount, conventional monitoring systems suffer from the constant consumption of energy to collect and transmit real-time sensor data, which makes it undesirable to replace batteries that power communication modules such as cellular modems.

[0022] The present configuration and teachings solve this "always changing the battery" problem by using synchronous machine learning models, alternatively referred to as digital twin models, in a preferred embodiment. In one configuration, one of the digital twin models runs on a remote telemetry unit ("RTU") and another of the digital twin models runs on a server that is communicatively coupled to a number of RTUs distributed in a fluid infrastructure network and aggregates data from those RTUs. In this configuration, at least one RTU is connected to a sensor, and the RTU and sensor are disposed at a monitoring location of the fluid infrastructure network that is remote with respect to the server. Thus, the server is a central processor that aggregates the remotely collected data and can be considered a digital twin for management of the fluid network. Exemplary fluid infrastructure items according to the present configuration and teachings include conduits, reservoirs, orifices, weirs, pumps, and wastewater treatment plants.

[0023] According to the present teachings, the digital twin model is used to generate estimates of sensor measurements at the RTU and the server. In one embodiment of the present teachings, the RTU compares the actual sensor measurements with estimates obtained from the digital twin model (running at the RTU). If the actual sensor measurements and the estimates from the digital twin model are within an acceptable range, the actual sensor measurements are not communicated from the RTU (where they reside) to the server. In this situation, the server relies on the digital twin model stored thereon and assumes that the estimates of the sensor measurements obtained from the digital twin model (at the server) are the actual sensor measurements. Eliminating the need to transmit actual sensor measurements from the RTU to the server represents an energy-saving feature of the present teachings that is not realized by conventional monitoring systems and methods. Furthermore, the energy-saving feature enables the present configuration and teachings to effectively monitor a fluid infrastructure network for several years without replacing the batteries powering the communication modules.

[0024] Furthermore, the server's ability to provide acceptable estimates of actual sensor measurements by relying on the digital twin model enables the present configuration and system to provide cost-effective real-time decision support for effectively managing fluid infrastructure networks.

[0025] To realize these advantages offered by the digital twin model (e.g., energy savings and / or real-time response measures), it is important that the digital twin model present at the RTU and the digital twin model present at the server are synchronized. For this purpose, if the actual sensor measurements and the estimates from the digital twin model are not within an acceptable range, the actual sensor measurements are communicated from the RTU to the server and a reconciliation process is initiated. At the server, information about the realized environmental conditions (e.g., precipitation) is received and used to obtain estimates of the sensor measurements from the digital twin model. If the difference between the actual sensor measurements at the server and the estimates from the digital twin model is outside an acceptable range, the realized environmental conditions are considered outliers. In this situation, the present teachings consider the digital twin models at the RTU and the digital twin models at the server to be synchronized, and therefore maintain the digital twin model at the server.

[0026] However, if the difference between the actual sensor measurements and the estimates of the sensor measurements obtained from the digital twin model at the server is within an acceptable range, the digital twin model at the RTU and the digital twin model at the server are considered to be out of sync according to the present teachings, and therefore the digital twin model at the server is updated or reconfigured so that it is synchronized with the digital twin model residing in the RTU.

[0027] The present teachings not only recognize the need for energy conservation and real-time response measures, but also provide architectural provisions and process features for synchronizing the underlying digital twin models at the RTU and the server that enable the benefits of the energy conservation and real-time response measures to be continuous and uninterrupted, as described in more detail below.

[0028] 1 illustrates a fluid infrastructure network management system 100 designed to monitor a fluid infrastructure according to one embodiment of the present configuration. The fluid infrastructure monitoring system 100 includes an RTU 102 and a server 104 communicatively coupled through a communication network 110 (e.g., the Internet or satellite). Additionally, the RTU 102 is powered by a battery 132 to perform a wide variety of functions.

[0029] The RTU 102 includes an RTU microprocessor 128 that is communicatively coupled to an RTU communication module 134, which is communicatively coupled to the network 110. Within the RTU 102, an RTU memory module is bidirectionally coupled to the RTU microprocessor 128. The sensor module 136 receives data from sensors 112 configured to measure flow attributes of remote fluid infrastructure items located remotely relative to the server 104.

[0030] 1, sensor measurements taken by sensors 112 are communicated from sensor module 136 through a communications coupling to RTU microprocessor 128. Either after processing or without processing, calculated values ​​that rely on the sensor measurements or the sensor measurements themselves are communicated from RTU microprocessor 128 to RTU communications module 134.

[0031] In an alternative embodiment of the present fluid infrastructure monitoring system, the sensor module 136 may be directly coupled to the RTU communications module 134, and sensor measurements taken at the sensors 112 may not have to pass through the RTU microprocessor 128 to reach the RTU communications module 134. In this alternative configuration, the sensor measurements taken at the sensors 112 are communicated directly from the sensor module 136 to the RTU communications module 134. Regardless of whether the RTU microprocessor 128 is involved in communicating the sensor measurements taken at the sensors 112, the sensor measurements are communicated from the RTU communications module 134 over the network to the server 104.

[0032] The server 104 includes a server communications module 138 that is bidirectionally coupled to a server processor 140, which is bidirectionally coupled to a database 142. The server communications module 138, like the RTU communications module 134, is communicatively coupled to the network 110. With respect to sensor measurements taken at the sensors 112, the server communications module 138 receives the sensor measurements from the network 110 and forwards them to the server processor 140, which uses them, for example, in comparisons described below.

[0033] Although Fig. 1 illustrates server 104 being communicatively coupled to a single RTU 102 through network 110, the present configuration is not so limited. To this end, Fig. 2 illustrates a fluid infrastructure network management system 200 according to another embodiment of the present configuration, including a server 204 communicatively coupled to a plurality of RTUs, e.g., RTU 202 and other RTUs 214A, 214B, and 214C, through network 210. In this embodiment, RTU 202, network 210, and server 204 are substantially similar to their counterparts, i.e., RTU 102, network 110, and server 104, respectively, illustrated in Fig. 1.

[0034] FIG. 2 illustrates that the server 204 of the fluid infrastructure network management system 200 has a machine learning engine 206 and an RTU digital twin model 208 operating therein. One or more sensor measurements taken from a sensor 212 are communicated from the RTU 202 through a network 210 to the server 204. The sensor measurements are not the only information so communicated. Sensor parameter information relating to a sensor 212 of FIG. 2, which is substantially similar to the sensor 112 of FIG. 1, is communicated from the RTU 202 through a network 210 (e.g., the Internet or a satellite) to the server 204. Similarly, other sensor measurements taken using one or more different (i.e., different from the sensor 212) sensors are communicated from one or more RTUs associated with one or more of those different sensors (e.g., RTU 214A, RTU 214B, and RTU 214C, if the fluid infrastructure items associated with each of the RTUs 214 provide an input or input stream to the RTU 202) through the network 210 to the server 204.

[0035] In one embodiment of this configuration, inside the server 204, the machine learning engine 206 uses sensor parameter information of the fluid infrastructure network, sensor measurements of inputs into and outputs from the fluid infrastructure item of interest, i.e., the fluid infrastructure item for which the digital twin model is being built, and probability distribution functions of environmental information surrounding the sensor (e.g., sensor 212 in FIG. 2) to build a digital twin model of the RTU 202. The digital twin model according to a preferred embodiment of the present teachings is a probabilistic predictive model that predicts measurements of one or more of the sensors in the fluid infrastructure network.

[0036] Representative sensor measurements for the fluid inside the fluid infrastructure item include flow rate data, flow level data, static pressure data, flow velocity data, concentration data for one or more chemical species of interest, pH, turbidity, total suspended solids, residual chlorine, total chlorine, total organic carbon, dissolved oxygen, ammonia, nitrates, and phosphorus. The sensor measurements taken using the sensors are communicated to server 204 over network 210 using its associated RTU (e.g., RTU 214A, RTU 214B, and RTU 214C, if the fluid infrastructure item associated with each of the RTUs provides an input or input stream into RTU 202 or an output or output stream out of RTU 202).

[0037] The input information used to construct the digital twin model 208 relating to an input into a fluid infrastructure item of interest (e.g., an infrastructure item having a flow attribute measured by a sensor 212) includes at least one information selected from the group consisting of precipitation information surrounding the sensor (e.g., whose sensor 212 measurements are estimated by the digital twin model 20), liquid levels of one or more different inputs into the fluid infrastructure item, and flow rates of one or more different inputs into the fluid infrastructure item. As an example, if RTU 214A is associated with a conduit that conveys fluid to RTU 202 from a storage tank, which serves as a fluid infrastructure item and is monitored by RTU 214B, the fluid flow sensor measurements of the conduit are communicated from RTU 214A through network 210 to server 204 and / or the liquid level sensor measurements of the storage tank are communicated from RTU 214B through network 210 to server 204. Further, these sensor measurements of input into the RTU 202 are communicated over the network to a server for the construction of a digital twin model 208, one of which ultimately resides in the RTU 202 as well.

[0038] As another example, server 204 receives precipitation data 216 from the National Weather Service that forecasts the amount and duration of precipitation in an area surrounding the location of a remote sensor (e.g., sensor 212) for construction of a digital twin model (e.g., digital twin model 208) of the RTU (e.g., RTU 202) using machine learning engine 206. In this manner, a digital twin model of each RTU (e.g., RTU 202, RTU 214A, RTU 214B, or RTU 214C) is constructed and preferably stored inside the server (e.g., server processor 140 of FIG. 1).

[0039] Similarly, output information used to construct the digital twin model 208 relating to an output from a fluid infrastructure item of interest (e.g., an infrastructure item having a flow attribute measured by a sensor 212) includes at least one information selected from the group consisting of evaporation information around the sensor (e.g., whose sensor 212 measurements are estimated by the digital twin model 20), liquid levels of one or more different outputs from the fluid infrastructure item, and flow rates of one or more different outputs from the fluid infrastructure item. Fluid flow rate measurements and / or liquid level measurements of the output or output streams from the fluid infrastructure item of interest (as monitored by the RTUs 202) are communicated from their respective RTUs over the network to the server. Such sensor measurements of the outputs from the RTUs 202 are used in constructing the digital twin model 208.

[0040] Examples of sensor parameters include the location of a sensor (e.g., sensor 212), the type of sensor, critical high / low transmission rates of a sensor and / or an RTU (e.g., RTU 202, RTU 214A, RTU 214B, or RTU 214C), battery level of a sensor, etc. Some or all of these parameters may be obtained from inputs at the sensor, the RTU, or the server.

[0041] In one embodiment of the present teachings, a digital twin model (e.g., digital twin model 208 of FIG. 2) is first created using a microprocessor (e.g., microprocessor 128 of FIG. 1) in an RTU (e.g., RTU 102 of FIG. 1) using the information described above, and then a copy of the digital twin model is communicated to a server (e.g., server 204 of FIG. 2). However, the server may enjoy greater data processing capabilities than the RTU, and therefore, in a preferred embodiment of the present teachings, digital twin model 208 is first constructed inside server 204, and then a copy of the digital twin model is communicated to RTU 202 through network 210 such that a copy of digital twin model 208 exists in server 204 and in RTU 202.

[0042] In this context, Figure 3 illustrates a fluid infrastructure network management system 200 in accordance with a preferred embodiment of the present arrangements and teachings, in which an RTU 302 is communicatively coupled to a server 304 through a network 310, which is also communicatively coupled to a number of RTUs, namely, RTU 314A, RTU 314B, and RTU 314C. Additionally, the server 304 has resident therein a machine learning engine 306 and a digital twin model 308 that are used to predict or forecast measurements of sensors (e.g., sensor 212 of Figure 2) monitored by the RTU 302. In this configuration, the RTU 302, network 310, server 304, RTU 314A, RTU 314B, and RTU 314C, machine learning engine 306, and digital twin model 308 are substantially counterparts, i.e., RTU 202, network 210, server 204, RTU 214A, RTU 214B, and RTU 214C, machine learning engine 206, and digital twin model 208, as shown in FIG. 2. Furthermore, in this configuration, a copy of the digital twin model is preferably communicated from server 304 to RTU 302 through network 310. As a result, there are two copies of the digital twin model, one at server 304 and denoted by reference numeral "308" and another at RTU 302 and denoted by reference numeral "318."

[0043] The present teachings thus recognize that digital twin models for other infrastructure items in the fluid infrastructure network monitored by other RTUs, e.g., RTU314A, RTU314B, and RTU314C, are preferably created either in server 304 or in RTU302, and then copies of the digital twin models are communicated to the other, i.e., preferably either RTU302 or server 304. Regardless of the manner in which the initial digital twin model is created, the server, e.g., server 304, has stored thereon copies of the digital twin models stored inside each of the RTUs monitoring different fluid infrastructure items in the fluid infrastructure network. More importantly, the present teachings recognize that the server preferably incorporates infrastructure connections between the RTUs (e.g., RTU314B monitors a tank located downstream from RTU314A, which monitors a conduit connected to a fluid infrastructure item monitored by RTU302) to automatically determine a "system-wide digital twin model." To this end, the present teachings also recognize that such a "system-wide digital twin model" may enable several scheduled transmissions of actual sensor measurements from one or more RTUs to a server to obtain more accurate estimates of the sensor measurements being monitored by one or more of the RTUs without dramatically increasing the energy requirements associated with transmitting the sensor measurements from the RTUs to the server. Rather than having to create a "system-wide digital twin model," the present teachings recognize that by simply constructing different digital twin models at the server, where a copy of each digital twin model is also stored on the RTUs, the present configuration provides the benefits of the RTUs without relying on monitoring the activity of the RTUs.

[0044] 4 illustrates a fluid infrastructure network management system 400 according to a preferred embodiment of the present arrangement. As described below, the management system 400 relies on a server's estimation of flow attributes inside the fluid infrastructure items as being sensor measurements of the flow attributes inside the fluid infrastructure items or as being estimates of such sensor measurements. In this embodiment of the present teachings, the management system 400 preferably does not receive sensor measurements from RTUs that monitor and periodically receive sensor measurements of the flow attributes inside the fluid infrastructure items.

[0045] In the embodiment shown in FIG. 4, the RTU 402 preferably receives predictions of precipitation data 416 (e.g., time and amount of precipitation, which may be in the form of a probability distribution function) from time to time over the network 410. Using the predictions of the precipitation data, the RTU digital twin model 418 preferably instructs a machine learning engine 420 (located at the RTU) to generate a probability distribution function that estimates flow attributes that may be measured by a sensor during a future instance in time or during a future period in time. As another input, the RTU 402 also receives sensor measurements in a sensor module 422 from a sensor 412 that measures flow attributes inside a fluid infrastructure item. The RTU 402 is thus well positioned to compare the sensor measurements to one or more of the RTU's estimated flow attribute values. In a preferred embodiment of the present teachings, the RTU 402 electronically compares the sensor measurements to the probability distribution functions of the estimated flow attribute values, a graphical representation of which is indicated by reference numeral "424."

[0046] If the sensor measurements fall within a confidence interval of one or more RTU's estimated flow attribute values ​​(e.g., values ​​ranging from approximately ±10% of the mean of the probability distribution function to approximately ±15% of the mean of the probability distribution function), then the present teachings recognize that, as an energy conservation measure, the RTU need not communicate the actual sensor measurements to the server.

[0047] The server 404 also receives predictions of precipitation data 416 (e.g., time and amount of precipitation, which may be in the form of a probability distribution function) over the network 410. The server 404 uses the server's digital twin model 408 to use the precipitation data 416 and generate a probability distribution function that estimates a flow attribute value for the infrastructure item. In a preferred embodiment of the present configuration, the digital twin model 408 instructs the machine learning engine 406 to use the precipitation data 416 and generate a probability distribution function that estimates the flow attribute value. Regardless of the manner in which the flow attribute sensor measurements are estimated, the server 404 assumes that the sensor measurements are the mean value of the probability distribution function that estimates the flow attribute value and reports the mean value of the server's probability distribution function as being the sensor measurement or as being an estimate of the sensor measurement in the server's user interface.

[0048] Such reporting may be performed periodically, e.g., every 5 minutes, and is not necessarily performed after the passage of a future instance in time, which forms the basis for an estimate, e.g., a probability distribution function, of the sensor measurement. In a preferred embodiment of the present teachings, after the passage of a future instance in time or after the passage of a future period in time, and when the server 404 does not receive a sensor measurement from the RTU 402, the server reports the average value of the server's probability distribution function to the user as being the sensor measurement or a reliable estimate of the sensor measurement.

[0049] Moreover, if the actual sensor measurement falls within the confidence interval, the RTU 402 may still communicate the actual sensor measurement over the network 410 to the server 404. Under such conditions, and to conserve energy, the RTU 402 may so communicate the actual sensor measurement relatively infrequently, for example, once a day or once a week. In some embodiments of the present teachings, when the server 408 does not receive a sensor measurement from the RTU 402 at a predefined time, at or after the predefined time, the server 408 relies on precipitation data to calculate its estimated flow attribute values, e.g., probability distribution functions, and reports the average of these estimated flow attribute values ​​as being the sensor measurement or as being a reliable estimate of the sensor measurement.

[0050] The present teachings also recognize the importance of the possibility that a sensor measurement may fall outside the confidence interval of one or more estimated flow attribute values ​​of the RTU (e.g., the value may fall outside the range of about ±10% of the mean of the probability distribution function to about ±15% of the mean of the probability distribution function), and the present teachings then recognize that an additional condition must be tested, namely, whether the actual precipitation achieved is an outlier or within the expected precipitation (as known from precipitation data 516).

[0051] In the embodiment shown in FIG. 5, the fluid network management system 500 includes an RTU 502 communicatively coupled to a server 504 through a network 510 . Additionally, the RTU 502, server 504, server machine learning engine 506, server digital twin model 508, network 510, sensors 512, RTU digital twin 518, RTU machine learning 520, sensor module 522, graphical representation of sensor data for the RTU's inferred sensor data 524, and graphical representation of sensor data for the server's inferred sensor data 526 are substantially similar to their counterparts shown in FIG. 4 , i.e., the RTU 402, server 404, server machine learning engine 406, server digital twin model 408, network 410, sensors 412, RTU digital twin 418, RTU machine learning 420, sensor module 422, graphical representation of sensor data for the RTU's inferred sensor data 424, and graphical representation of sensor data for the server's inferred sensor data 426.

[0052] Precipitation data 516, e.g., a prediction of environmental conditions in an area surrounding one or more sensors, such as sensor 512, is communicated, preferably over network 510, to both the RTU 502 and server 504. In this configuration, the RTU 502 uses the machine learning engine 520 and the prediction of environmental conditions obtained from the precipitation data 516 to determine new estimated flow attribute values ​​for the RTU inside the fluid infrastructure item during an instance in time or period of time.

[0053] If the absolute value of the difference between the sensor measurements obtained from the sensors 512 and the sensor modules 522 and the new estimated flow attribute values ​​of the RTU 502 is greater than a predefined threshold tolerance difference value, or if the sensor measurements fall outside of the predefined confidence intervals of the estimated flow attribute values ​​of the RTU 502 during an instance in time or during a time period, the underlying sensor measurements used to calculate the estimates are communicated from the RTU 502 to the server 504 over the network 510. In the server 502, the precipitation data 516, e.g., actual precipitation achieved around an area surrounding one or more sensors, such as the sensor 512, is used to calculate one or more new estimated flow attribute values ​​of the server 504 (see, e.g., the curve in the graphical representation 526). This estimate is an estimate of the flow attribute during an instance in time or during a time period inside the fluid infrastructure item monitored by the sensor 512. The server 504 compares the sensor measurements taken from the sensors 512 with one or more of the server's 504 new estimated flow attribute values ​​(e.g., a probability distribution function including multiple estimated values), as shown in the graphical representation 526.

[0054] The digital twin model 508 is maintained in the server 504 if the absolute value of the difference between the sensor measurement and one of the server's new estimated flow attribute values ​​is equal to or greater than a check predefined threshold tolerance value, or if the sensor measurement falls outside a check predefined confidence interval of the server's new estimated flow attribute values ​​during an instance in time or during a time period. In this scenario, the present teachings recognize that the actual precipitation realized in the area surrounding the sensor 512 is an outlier to the predicted precipitation in that area. As a result, the present teachings do not consider the digital twin model to be inaccurate in estimating the flow attributes inside the remotely located fluid infrastructure item, but rather understand that the prediction did not accurately predict the precipitation.

[0055] However, if the absolute value of the difference between the sensor measurement and one of the server's new estimated flow attribute values ​​is equal to or less than a check predefined threshold tolerance value, or if the sensor measurement falls within a check predefined confidence interval of the server's new estimated flow attribute values ​​during an instance in time or during a time period, the digital twin model 508 is updated, recalibrated, or reconfigured at the server 504. A copy of this updated, recalibrated, or reconfigured digital model is communicated to the RTU 502 so that the "twin" nature of the digital model is re-established. Because the present teachings recognize that the estimated flow attribute values ​​of the previous digital twin model were unreliable or otherwise unacceptable, the digital twin model is updated, recalibrated, or reconfigured. As a result, the present teachings suggest updating, recalibrating, or reconfiguring as necessary so that the estimated flow attribute values ​​of the new digital twin model are reliable and acceptable.

[0056] The present teachings also provide novel methods of infrastructure management. Although implementation of these novel methods does not require any particular system configuration, the novel methods implemented using the structural details described in Figures 1-5 represent preferred embodiments of the present configuration.

[0057] FIG. 6 illustrates an infrastructure management method 600 according to one embodiment of the present teachings. The infrastructure management method 600 preferably begins with element 602, which involves obtaining a digital twin model inside each of the servers (e.g., server 104 of FIG. 1 ) and inside the RTU (e.g., RTU 102 of FIG. 1 ). As explained above, the digital twin model provides one or more estimates of flow attributes inside the fluid infrastructure item during an instance in time or during a time period. The method 600 then proceeds to element 604, which involves receiving a prediction of environmental conditions around an area surrounding a sensor (e.g., sensor 112 of FIG. 1 ) during an instance in time or during a time period. Element 606 is then executed at the RTU during the instance in time or during a time period, using the prediction of the environmental conditions (obtained in element 604) and the digital twin model (obtained in element 602) to generate estimated flow attributes of the RTU inside the fluid infrastructure item.

[0058] Element 606 may also include performing an estimation using the prediction of environmental conditions and the digital twin model at the server during an instance in time or during a time period to generate an estimated flow attribute value of one or more servers for the fluid infrastructure item. In another embodiment, another element different from element 606 is implemented in method 600 to arrive at an estimated flow attribute value of one or more servers for the fluid infrastructure item. As a result, generating these two estimates, i.e., an estimated flow attribute value of the RTU of the fluid infrastructure item and an estimated flow attribute value of the server of the fluid infrastructure item, may be considered as two separate elements of method 600 or as a single element.

[0059] Regardless of whether one or two elements are implemented to obtain these estimated flow attribute values, the method 600 proceeds to element 608 which involves obtaining, at the RTU, a sensor measurement of a flow attribute inside the fluid infrastructure item using a sensor (e.g., sensor 112 of FIG. 1). The method 600 then proceeds to element 610 which compares the sensor measurement at the RTU to the estimated flow attribute values ​​of the RTU. At element 610, the sensor measurement of the flow attribute inside the fluid infrastructure item at the RTU is compared to one or more of the estimated flow attribute values ​​of the RTU. By way of example, FIG. 4 illustrates graphical representation 424 as an exemplary illustration of the "comparison" performed under element 610.

[0060] Preferably, element 612 is performed after element 610 is completed. According to the embodiment shown in Figure 6, element 612 of method 600 involves reporting the estimated flow attribute value of the server as being a sensor measurement of a flow attribute inside the fluid infrastructure item or being an estimate of a sensor measurement of a flow attribute inside the fluid infrastructure item during an instance in time or during a time period if the absolute value of the difference between the sensor measurement and the estimated flow attribute value of the RTU is equal to or less than a predefined threshold tolerance difference value or if the sensor measurement falls within a predefined confidence interval of the estimated flow attribute values ​​of the RTU during the instance in time or during the time period. By way of example, Figure 4 shows graphical representation 426 as an exemplary illustration of the "reporting" performed under element 612.

[0061] In another preferred embodiment of the present teachings, element 612 involves reporting the server's estimated flow attributes to be provided in place of the sensor measurements of the fluid infrastructure item after an instance in time or after a time period if no sensor measurements of the flow attributes inside the fluid infrastructure item are received at the server. In one implementation of this embodiment, if the RTU determines that the absolute value of the difference between the sensor measurements and the RTU's estimated flow attribute values ​​is equal to or less than a predefined threshold tolerance value or if the RTU determines that the sensor measurements fall within a predefined confidence interval of the RTU's estimated flow attribute values ​​of the fluid infrastructure item during the instance in time or during the time period, the RTU does not communicate the sensor measurements of the fluid infrastructure item to the server.

[0062] Prior to implementation of element 610 and / or element 612, method 600 more preferably does not include communicating sensor measurements of flow attributes inside the fluid infrastructure item from the RTU to a server.

[0063] Figure 7 illustrates a fluid infrastructure network management method 700 according to another embodiment of the present teachings. The management method 700 preferably includes performing elements 602, 604, 606, 608, 610, and 612 as described above with respect to Figure 6. Instead of performing element 612, Figure 7 provides element 702 as it provides a context in which an RTU communicates sensor measurements of flow attributes inside a fluid infrastructure item to a server.

[0064] Thus, element 702 involves communicating a sensor measurement between instances in time or for a time period from the RTU to the server if the absolute value of the difference between the sensor measurement and one of the RTU's estimated flow attribute values ​​is greater than a predefined threshold tolerance difference value, or if the sensor measurement falls outside a predefined confidence interval of the RTU's estimated flow attribute values ​​during the instance in time or for the time period. By way of example, FIG. 5 illustrates graphical representation 524 as an illustrative view of the underlying calculations prior to the "communication" described in element 702.

[0065] Method 700 includes an element 704 for receiving, at a server, information regarding realized environmental conditions, i.e., realized environmental conditions around an area surrounding the fluid infrastructure item and the sensor during an instance in time or period of time, after the instance in time or period of time has elapsed. However, in one embodiment of method 700, element 704 of the present teachings may be performed before the instance in time or period of time.

[0066] Next, method 700 involves executing element 706 which involves determining new estimated flow attribute values ​​of the server inside the fluid infrastructure item for the instance in time or period of time based on information about the realized environmental conditions (obtained from element 704) and the digital twin model residing on the server.

[0067] The method 700 then includes an element 708 of comparing, at the server, the sensor measurement (obtained from element 702) with one of the server's new estimated flow attribute values ​​(obtained from element 706).

[0068] The method 700 then proceeds to element 710 which involves maintaining the digital twin model at the server if the absolute value of the difference between the sensor measurement and one of the server's new estimated flow attribute values ​​is equal to or greater than a check predefined threshold tolerance value, or if the sensor measurement falls outside a check predefined confidence interval of a plurality of the server's new estimated flow attribute values ​​during an instance in time or during a time period. By way of example, FIG. 5 shows a graphical representation 526 to illustrate the underlying calculations performed prior to "maintaining", as described under maintaining. 1. The above-mentioned elements of methods 600 and 700 may be implemented in a wide variety of different embodiments. As an example, element 602 includes building a digital twin model inside a server or RTU that provides one or more estimated flow attribute values ​​of a fluid infrastructure item during an instance in time or during a time period. In a preferred embodiment, element 602 of the present teachings includes building a probabilistic predictive model using sensor parameter information, probability distribution functions of environmental information in an area surrounding the sensor, and input and output information about the fluid infrastructure item. To this end, FIG. 2 illustrates an exemplary digital twin model 208 formed in server 204.

[0069] The sensor parameter information includes at least one item selected from the group consisting of a location of the sensor, a type of the sensor, a critical high flow attribute detectable by the sensor, a critical low flow attribute detectable by the sensor, a transmission rate of the sensor data, and a battery level of the sensor.

[0070] The input information includes at least one information selected from the group consisting of precipitation information in an area surrounding a sensor associated with the RTU, liquid levels of one or more different inputs into the fluid infrastructure item, and flow rates of one or more different inputs into the fluid infrastructure item.

[0071] The output information includes at least one information selected from the group consisting of evaporation information around a sensor associated with the RTU, liquid levels at one or more distinct outputs exiting the fluid infrastructure item, and flow rates at one or more distinct outputs exiting the fluid infrastructure item.

[0072] In addition to liquid level and / or fluid flow rate, the flow attributes of the input and output information may also include at least one attribute selected from the group consisting of hydraulic head, groundwater flux, groundwater flux rate, pH, conductivity, total organic compound concentration, total suspended solids, and turbidity.

[0073] The construction contemplated under element 602 preferably includes using a machine learning module to generate a probability distribution function of estimated sensor measurements of flow attributes inside the fluid infrastructure item.

[0074] Regardless of the manner in which the digital twin model is derived, element 602 may include modifying the digital twin model using the RTUs and sensors (e.g., sensor measurements provide insight into the reliability of the digital twin model's estimates) until the digital twin model reliably estimates flow attributes inside the fluid infrastructure item. By way of example, modifying the digital twin model may include changing values ​​of one or more parameters associated with the fluid infrastructure item used to build the initial version of the digital twin model until one or more of the estimates of the flow attributes derived from the digital twin model are equal to or close to the sensor measurements (e.g., sensor measurements derived from sensor 212 in FIG. 2).

[0075] In a preferred embodiment of the present teachings, modifying the digital twin model includes determining that a Nash-Sutcliff coefficient of the digital twin model is greater than a predefined tolerance value or determining that a root mean square deviation between an estimate of the flow attribute and a sensor measurement of the flow attribute obtained during different instances in time or during different time periods is less than or equal to a predefined error value.

[0076] The present teachings recognize that the construction and modification, when implemented, preferably provide at least one of the digital twin models (e.g., digital twin model 208 of FIG. 2 ) inside a server (e.g., server 204 of FIG. 2 ). Additionally, the server (e.g., server 204 of FIG. 2 ) has installed therein or is communicatively coupled to a machine learning engine (e.g., machine learning engine 206 of FIG. 2 ) that constructs the digital twin model and verifies that the digital twin model is an acceptable model for reliably estimating flow attributes inside the fluid infrastructure item.

[0077] According to some preferred embodiments of the present teachings, once an acceptable digital twin model is present on the server, method 600 includes an element of communicating or transmitting a copy of the digital twin model from the server to a remote telemetry unit (e.g., as shown in FIG. 3 , server 304 communicates a copy of digital twin model 308 over network 310 to RTU 302 such that the copy is stored therein as digital twin model 318).

[0078] After an acceptable digital twin model is obtained at the server, the method 600 of FIG. 6 preferably further includes storing the digital twin model on a memory (e.g., database 142) associated with the server (e.g., server 104 of FIG. 1) and on a remote memory (e.g., memory 130 of the RTU of FIG. 1) that is remote to the memory (e.g., database 142) or the server (e.g., server 104 of FIG. 1).

[0079] 6, the method 600 includes receiving a probability distribution function that predicts the amount of precipitation that will be realized around the area surrounding the sensor during an instance in time or during a period of time. The probability distribution function preferably provides multiple estimates for the amount of precipitation that will be realized around the area surrounding the sensor.

[0080] According to a preferred implementation of the present teachings, when implementing element 606 of method 600 shown in FIG. 6, the neural network calculates an estimated flow attribute value of the server between instances in time or between time periods. Furthermore, a machine learning engine disposed inside the RTU calculates an estimated flow attribute value of the RTU between instances in time or between time periods. Furthermore, each of the estimated flow attribute value of the server and the estimated flow attribute value of the RTU is preferably a probability distribution function. As an example, in FIG. 4, each of the graphical representation 426 of the server 404 and the graphical representation 424 of the RTU 402 shows a probability distribution function estimating a flow attribute value inside a fluid infrastructure item.

[0081] Continuing with this example, and with respect to element 608 shown in FIG. 6, the method 600 preferably includes comparing, at the RTU, a sensor measurement of the flow attribute to a probability distribution function of the RTU that estimates a flow attribute value inside the fluid infrastructure item during an instance in time or during a time period.

[0082] Element 612 preferably includes reporting the mean value of the server's probability distribution function as being the sensor measurement or the estimate of the flow attribute inside the fluid infrastructure item during the instance in time or during the time period if the sensor measurement during the instance in time or during the time period falls within a predefined confidence interval of the mean value of the RTU's probability distribution function. In one embodiment, the present teachings recognize that the confidence interval spans a predefined range of the estimated flow attribute values ​​that deviate from the mean value of the RTU's probability distribution function.

[0083] In one preferred implementation of method 600, element 604 includes receiving a prediction of an environmental condition around an area surrounding the sensor at the periodic rate. In one implementation of this example, element 606 of the present teachings includes estimating a flow attribute inside the fluid infrastructure item at the periodic rate or a rate less than the periodic rate and at the RTU to generate an estimated flow attribute value for one or more RTUs. Continuing with this example, element 608 of the present teachings includes obtaining sensor measurements at a sampling rate of the sensor. In this example, the sampling rate is greater than the periodic rate.

[0084] Furthermore, if the absolute value of the difference between the sensor measurement and one of the RTU's estimated flow attribute values ​​is greater than a predefined threshold tolerance value, or if the sensor measurement falls outside a predefined confidence interval of the RTU's estimated flow attribute values ​​during the instance in time or during the time period, the management method of the present teachings preferably further includes an element (e.g., element 702) that communicates the sensor measurement taken during the instance in time or during the time period from the RTU to the server at at least the sampling rate or at a rate less than the sampling rate.

[0085] This preferred embodiment further includes an element (e.g., substantially similar to element 704 of FIG. 7 ) that receives, at the server, after the instance in time or after the time period, information about realized environmental conditions (e.g., precipitation) realized in the area surrounding the sensor during the instance in time or during the time period. Then, as in element 706 of FIG. 7 , based on the information about the realized environmental conditions and the digital twin model present at the server, the management method of the present teachings provides an element that determines a new estimated flow attribute value of the server during the instance in time or during the time period. In one embodiment of the determining element, the new estimated flow attribute value of the server is a probability distribution function (e.g., the probability distribution function shown in the graphical representation 526 of FIG. 5 ).

[0086] At the server, the management method of the present teachings then implements a comparing element (e.g., substantially similar to element 708 of FIG. 7 ) that involves comparing the sensor measurement to the server's new estimated flow attribute values. The management method of the present teachings then proceeds to maintain the digital twin model at the server if an absolute value of a difference between the sensor measurement and one of the server's new estimated flow attribute values ​​is equal to or greater than a check predefined threshold tolerance value, or if the sensor measurement falls outside a check predefined confidence interval of the server's multiple new estimated flow attribute values ​​during an instance in time or during a time period. If the digital twin model was maintained at the server, the management method of the present teachings includes maintaining a copy of the digital twin model at the RTU.

[0087] In a preferred embodiment, the management method of the present teachings includes obtaining a new digital twin model within each of the servers and within the RTU that differs from the (original) digital twin model if the absolute value of the difference between the sensor measurement value and one of the server's new estimated flow attribute values ​​is less than a predefined threshold tolerance value or if the sensor measurement value falls within a predefined confidence interval of a number of the server's new estimated flow attribute values ​​during an instance in time or during a time period.

[0088] With respect to method 700, the present teachings provide additional different embodiments for effective management of a fluid network. In one such embodiment, method 700 includes obtaining a new digital twin model inside the RTU if an absolute value of a difference between the sensor measurement value and one of the new estimated flow attribute values ​​of the remote telemetry unit is less than a check predefined threshold tolerance value or if the sensor measurement value falls within a plurality of check predefined confidence intervals of the new estimated flow attribute value of the remote telemetry unit during an instance in time or during a time period.

[0089] Continuing with this embodiment, the method 700 preferably further includes an element that communicates the new digital twin model formed in the RTU to a server. In this embodiment, the communicating element is performed using a communication coupling between the RTU and the server. In this configuration, the server is communicatively coupled to a plurality of RTUs and has stored thereon one copy of the digital twin model that is also stored on one of the RTUs. Additionally, each RTU is coupled to a sensor that measures a particular type of flow attribute inside the fluid infrastructure item. The digital twin in the server enables the server to arrive at a new estimated flow attribute value for the server and to operate independently of the RTU in some instances, thereby using information regarding the amount of precipitation realized after the lapse of an instance in time or after the lapse of a time period to conserve energy.

[0090] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present teachings and configurations. However, it will be apparent to one of ordinary skill in the art that the present teachings and configurations may be practiced without being limited to some or all of these specific details. In other instances, well-known process steps have not been described in detail in order not to unnecessarily obscure the present teachings and configurations.

Claims

1. 1. A method of infrastructure management, the method comprising: obtaining, inside each of the servers and inside a remote telemetry unit, a digital twin model that provides one or more estimates of flow attributes inside the fluid infrastructure item during an instance in time or during a time period, the remote telemetry unit and a sensor located remotely with respect to the servers, the remote telemetry unit being communicatively coupled to a sensor that measures a flow attribute inside the fluid infrastructure item, whereby the remote telemetry unit receives sensor measurements from the sensor; receiving a prediction of environmental conditions around an area surrounding the sensor during the instance in time or during the time period; estimating, at one or more remote telemetry units, flow attributes inside the fluid infrastructure item during the instance in time or during the time period using the predictions of the environmental conditions and one of the digital twin models to generate estimated flow attribute values ​​for the remote telemetry units; estimating, at one or more servers, flow attributes inside the fluid infrastructure item during the instance in time or during the time period using the predictions of the environmental conditions and one of the digital twin models to generate estimated flow attribute values ​​for the one or more servers; obtaining, at the remote telemetry unit, the sensor measurements of the flow attributes inside the fluid infrastructure item using the sensor; comparing, at the remote telemetry unit, the sensor measurements to one or more of the remote telemetry unit's estimated flow attribute values; reporting one or more of the server's estimated flow attribute values ​​as being said sensor measurement or as being an estimate of said sensor measurement during said instance in time or during said time period if an absolute value of a difference between said sensor measurement value and one of said remote telemetry unit's estimated flow attribute values ​​is equal to or less than a predefined threshold tolerance difference value or if said sensor measurement value falls within a predefined confidence interval of a plurality of said remote telemetry unit's estimated flow attribute values ​​during said instance in time or during said time period; (a) a methodology for managing infrastructure, including:

2. 2. The method of infrastructure management of claim 1, wherein the fluid infrastructure item is at least one item selected from the group consisting of a conduit, a reservoir, an orifice, a weir, a pump, and a wastewater treatment plant.

3. 3. The method of infrastructure management of claim 2, wherein the flow attributes include at least one attribute selected from the group consisting of liquid level, fluid flow rate, hydraulic head, groundwater flux, groundwater flux rate, pH, conductivity, total organic compound concentration, total suspended solids, and turbidity.

4. The step of obtaining building, inside the server or the remote telemetry unit, the digital twin model giving one or more of the estimated flow attribute values ​​of the remote telemetry unit or one or more of the estimated flow attribute values ​​of the server during the instance in time or during the time period; modifying the digital twin model using the remote telemetry unit or the sensors until the digital twin model reliably estimates the flow attributes inside the fluid infrastructure item; and implementing the digital twin model in the server or in the remote telemetry unit. The method of infrastructure management of claim 1 , comprising:

5. 5. The method of infrastructure management of claim 4, wherein the constructing step comprises constructing a probabilistic predictive model using sensor parameter information, a probability distribution function of environmental information about the area surrounding the sensor, and input and output information regarding the fluid infrastructure item.

6. the sensor parameter information includes at least one item selected from the group consisting of a location of the sensor, a type of sensor, a critical high flow attribute detectable by the sensor, a critical low flow attribute detectable by the sensor, a data transmission rate of the sensor, and a battery level of the sensor; the input information comprises at least one information selected from the group consisting of: precipitation information around the area surrounding the sensor associated with the remote telemetry unit; liquid levels of one or more of the different inputs to the fluid infrastructure item; and flow rates of one or more of the different inputs to the fluid infrastructure item; 6. The method of infrastructure management according to claim 5, wherein the output information comprises at least one information selected from the group consisting of precipitation information around the sensor associated with the remote telemetry unit, liquid levels of one or more different outputs exiting the fluid infrastructure item, and flow rates of one or more different outputs exiting the fluid infrastructure item.

7. 6. The method of infrastructure management according to claim 5, wherein the constructing step comprises using a machine learning module to generate a probability distribution function of the sensor measurements of the flow attribute inside the fluid infrastructure item.

8. 6. The method of infrastructure management of claim 5, wherein the modifying step comprises changing one or more parameter values ​​associated with the fluid infrastructure item and used in the constructing of the digital twin model until one or more of the estimates of the flow attributes obtained from the digital twin model are close to or equal to the sensor measurements.

9. 9. The method of infrastructure management of claim 8, wherein the modifying step comprises determining that a Nash-Sutcliffe coefficient of the digital twin model is greater than a predefined tolerance value or determining that a root mean square deviation between the estimate of the flow attribute and the sensor measurements of the flow attribute obtained during different instances in time or during different time periods is less than or equal to a predefined error value.

10. The method of infrastructure management of claim 5 , further comprising deploying one of the digital twin models at the server.

11. 11. The method of infrastructure management of claim 10, further comprising the step of communicating a copy of the digital twin model to the remote telemetry unit from the server, the server having installed thereon or communicatively coupled to a machine learning engine that performs the building and verifying steps.

12. 12. The method of infrastructure management of claim 11, further comprising storing the digital twin model in a memory associated with the server and in a remote memory remote to the memory.

13. 2. The method of infrastructure management of claim 1, wherein the receiving step comprises receiving a probability distribution function forecasting an amount of precipitation in the area surrounding the sensor during the instance in time or during the time period.

14. 2. The method of infrastructure management of claim 1, wherein the estimating at the server during the instance in time or during the time period comprises: using a neural network to calculate one or more of the estimated flow attribute values ​​of the server; and using a machine learning engine disposed inside the remote telemetry unit to calculate one or more of the estimated flow attribute values ​​of the remote telemetry unit.

15. 2. The method of infrastructure management according to claim 1, wherein each of the estimated flow attribute values ​​of the remote telemetry unit and the estimated flow attribute value of the server is a probability distribution function estimating the flow attribute inside the fluid infrastructure item.

16. 16. The method of infrastructure management of claim 15, wherein the comparing step is performed at the remote telemetry unit by comparing the sensor measurements of the flow attributes to a probability distribution function of the remote telemetry unit.

17. 16. The method of infrastructure management of claim 15, wherein the reporting step comprises the step of reporting a mean value of the server's probability distribution function during the instance in time or during the time period as being the sensor measurement or as an estimate of the sensor measurement if the sensor measurement falls within a predefined confidence interval of the mean value of the remote telemetry unit's probability distribution function during the instance in time or during the time period.

18. 20. The method of infrastructure management of claim 17, wherein the predefined range of estimated flow attribute values ​​ranges from approximately ±10% of the mean of the probability distribution function of the remote telemetry units to approximately ±15% of the mean of the probability distribution function of the remote telemetry units.

19. 2. The method of infrastructure management of claim 1, wherein the comparing and reporting steps are performed without communicating the sensor measurements of the flow attributes from the remote telemetry unit to the server.

20. the receiving step includes receiving, at a periodic rate, the prediction of the environmental condition about the area surrounding the sensor; the estimating step includes estimating an estimated flow attribute value of the remote telemetry unit at or below the periodic rate; the obtaining step includes obtaining the prediction of the environmental condition about the area surrounding the sensor at a sampling rate of the sensor; 2. The method of infrastructure management according to claim 1, wherein said sampling rate is greater than said periodic rate.

21. If the absolute value of the difference between the sensor measurement and one of the estimated flow attribute values ​​of the remote telemetry unit is greater than the predefined threshold tolerance value, or if the sensor measurement falls outside the predefined confidence interval of a plurality of estimated flow attribute values ​​of the remote telemetry unit during the instance in time or during the time period, the method comprises: transmitting the sensor measurements during the instance in time or during the time period from the remote telemetry unit to the server; receiving, at the server, at said instance in time or after said time period, information regarding realized environmental conditions realized in said area surrounding said sensor during said instance in time or during said time period; determining new estimated flow attribute values ​​of one or more servers for the fluid infrastructure item and during the instance in time or during the time period based on the information about the realized environmental conditions and the digital twin model residing on the server; comparing, at the server, the sensor measurements to one or more of the server's new estimated flow attribute values; maintaining the digital twin model at the server if the absolute value of the difference between the sensor measurement value and one or more of the server's new estimated flow attribute values ​​is equal to or greater than a check predefined threshold tolerance value, or if the sensor measurement value falls outside a check predefined confidence interval of a plurality of the server's new estimated flow attribute values ​​during the instance in time or during the time period; The method of infrastructure management of claim 1 , further comprising:

22. The method of infrastructure management of claim 1 , wherein the information regarding the realized environmental conditions includes an amount of realized precipitation during the instance in time or during the time period.

23. 23. The method of infrastructure management of claim 22, wherein in said determining step, the new estimated flow attribute of the server is a probability distribution function determined using the amount of realized precipitation.

24. 2. The method of infrastructure management of claim 1, further comprising the step of maintaining the digital twin model at the remote telemetry unit if the maintaining of the digital twin model at the server is performed.

25. 2. The method of infrastructure management of claim 1, further comprising obtaining a new digital twin model different from the digital twin model inside each of the servers and inside the remote telemetry unit if the absolute value of the difference between the sensor measurement and one of the server's new estimated flow attributes is less than a predefined threshold tolerance value or if the sensor measurement falls within a predefined confidence interval of a plurality of new estimated flow attribute values ​​of the server during the instance in time or during the time period.

26. 1. A method of infrastructure management, the method comprising: obtaining, inside a remote telemetry unit, a digital twin model providing one or more estimates of flow attributes inside a fluid infrastructure item during an instance in time or during a time period, the remote telemetry unit being communicatively coupled to sensors measuring the flow attributes inside the fluid infrastructure item, whereby the remote telemetry unit receives sensor measurements of the flow attributes from the sensors; receiving a forecast of the amount of precipitation expected to be experienced in an area surrounding said sensor at said instance in time or during said time period; at the remote telemetry unit, estimating one or more remote telemetry unit estimated flow attribute values ​​for the fluid infrastructure item based on the forecast of the precipitation and the digital twin model during the instance in time or during the time period; receiving at the remote telemetry unit the sensor measurements during the instance in time or during the time period; and receiving an amount of realized precipitation that was realized during the instance in time or during the time period in an area surrounding the sensor; comparing the sensor measurements to one or more of the estimated flow attribute values ​​of the remote telemetry unit; determining, at the remote telemetry unit, one or more new remote telemetry unit estimated flow attribute values ​​during said instance in time or during said time period using the digital twin and the amount of realized precipitation if the absolute value of the difference between the sensor measurement and one of the remote telemetry unit estimated flow attribute values ​​is greater than a predefined threshold tolerance value or if the sensor measurement falls outside a predefined confidence interval of a plurality of remote telemetry unit estimated flow attribute values ​​during said instance in time or during said time period; comparing, at the remote telemetry unit, the sensor measurements with one or more of the estimated flow attribute values ​​of the new remote telemetry unit; maintaining the digital twin model at the server if the absolute value of the difference between the sensor measurement value and one of the estimated flow attribute values ​​of the new remote telemetry unit is equal to or greater than a check predefined threshold tolerance value or if the sensor measurement value falls outside a check predefined confidence interval of a plurality of estimated flow attribute values ​​of the new remote telemetry unit during the instance in time or during the time period; (a) a methodology for managing infrastructure, including:

27. 27. The method of infrastructure management of claim 26, further comprising the step of deriving a new digital twin model inside the remote telemetry unit that is different from the digital twin model if the absolute value of the difference between the sensor measurement value and one of the estimated flow attributes of the new remote telemetry unit is less than a check predefined threshold tolerance value or if the sensor measurement value falls inside the check predefined confidence interval of the estimated flow attribute value of the new remote telemetry unit during the instance in time or multiple times during the time period.

28. transmitting the new digital twin model to a server; 28. The method of infrastructure management of claim 27, wherein the communicating step is performed using a communicative coupling between the remote telemetry unit and the server, the server being communicatively coupled to a plurality of remote telemetry units and storing at least one copy of the digital twin model thereon, the at least one copy of the digital twin model being stored in each of the remote telemetry units coupled to sensors measuring one or more different types of flow attributes inside one or more different types of fluid infrastructure items.

29. 28. The method of infrastructure management of claim 27, wherein said determining step comprises using information regarding the amount of precipitation achieved after said instance in time or after said period of time.

30. 28. The method of infrastructure management of claim 27, wherein the receiving step is performed prior to the instance in time or prior to the period of time.

31. 1. A method of fluid infrastructure network management, the method comprising: - obtaining inside a server a digital twin model providing one or more estimates of flow attributes inside a fluid infrastructure item during an instance in time or during a time period, wherein a remote telemetry unit and a sensor are located remotely with respect to the server, the remote telemetry unit being communicatively coupled to a sensor measuring a flow attribute inside the fluid infrastructure item, whereby the remote telemetry unit receives sensor measurements from the sensor, and the fluid infrastructure item is part of a fluid infrastructure network; receiving a prediction of environmental conditions around an area surrounding the sensor during the instance in time or during the time period; estimating, at one or more servers, flow attributes inside the fluid infrastructure item during the instance in time or during the time period using the predictions of the environmental conditions and one of the digital twin models to generate estimated flow attribute values ​​for the one or more servers; obtaining, at the remote telemetry unit, the sensor measurements of the flow attributes inside the fluid infrastructure item using the sensor; reporting one or more of said server's estimated flow attribute values ​​as being said sensor measurements or as being estimates of said sensor measurements during said instance in time or during said time period if said server does not receive said sensor measurements from said remote telemetry unit; A method for fluid infrastructure network management comprising:

32. 32. The method of fluid infrastructure network management of claim 31, wherein the reporting is performed if the server does not receive the sensor measurements from the remote telemetry unit after the instance in time or after the period of time has elapsed.

33. 32. The method of infrastructure management of claim 31, wherein the fluid infrastructure network management manages a fluid infrastructure network that includes one element selected from the group consisting of a groundwater system, a storm water system, and a wastewater system.

34. a plurality of sensors, each sensor designed to measure a flow attribute inside a fluid infrastructure item that is part of a fluid infrastructure network; a plurality of remote telemetry units, each remote telemetry unit communicatively coupled to at least one of the sensors; A fluid infrastructure network management system comprising: The remote telemetry unit comprises: A battery for providing power; a sensor module communicatively coupled to the sensor for receiving the sensor measurements; a memory for storing sensor measurements and one of a digital twin model providing estimated flow attribute values ​​of one or more remote telemetry units for said infrastructure item during an instance in time or during a time period; a communication module connected to a network and designed to transmit information from said remote telemetry unit to said network; A microprocessor programmed to implement instructions, the instructions comprising: receiving, using the communications module, a prediction of environmental conditions around an area surrounding the sensor during the instance in time or during the time period; using the predictions of the environmental conditions and the digital twin model to estimate estimated flow attribute values ​​of one or more remote telemetry units inside the fluid infrastructure item during the instance in time or during the time period; obtaining sensor measurements inside the fluid infrastructure item using the sensor; comparing the sensor measurements to one or more of the estimated flow attribute values ​​of the remote telemetry unit; communicating the sensor measurement value during said instance in time or during said time period from the remote telemetry unit to a network if the absolute value of the difference between the sensor measurement value and one of the remote telemetry unit's estimated flow attribute values ​​is greater than a predefined threshold tolerance value or if the sensor measurement value falls outside a predefined confidence interval of the remote telemetry unit's estimated flow attribute values ​​during said instance in time or during said time period; a microprocessor for performing a server communicatively coupled to a plurality of the remote telemetry units through the network and having a plurality of digital twin models stored thereon, each digital twin model being a copy of a digital twin model stored in each of the remote telemetry units, each of the remote telemetry units and each of the sensors being remote with respect to the server, the server being programmed to implement instructions; wherein the instructions include: receiving the sensor measurements during the instance in time or during the time period from the remote telemetry unit over the network; receiving, at the server after the instance in time or after the time period, information regarding realized environmental conditions realized in the area surrounding the sensor during the instance in time or during the time period; determining new estimated flow attribute values ​​of one or more servers inside the fluid infrastructure item during the instance in time or during the time period based on the information about the realized environmental conditions and the digital twin model residing on the server; comparing, at the server, the sensor measurements to one or more of the server's new estimated flow attribute values; maintaining the digital twin model at the server if an absolute value of the difference between the sensor measurement and one of the server's new estimated flow attribute values ​​is equal to or greater than a check predefined threshold tolerance value, or if the sensor measurement falls outside a check predefined confidence interval of a plurality of the server's new estimated flow attribute values ​​during the instance in time or during the time period; A fluid infrastructure network management system that:

35. 35. The fluid infrastructure network management system of claim 34, wherein the remote telemetry unit further comprises a direct connection between the battery and the communications module without passing through the microprocessor, the microprocessor programmed to instruct the battery to electrically activate dormant communications modules and to instruct electrically active communications modules to transmit the sensor measurements to the server.

36. 35. The fluid infrastructure network management system of claim 34, wherein the remote telemetry unit further comprises an indirect connection between the battery and the communications module through the microprocessor, the microprocessor programmed to instruct the battery to electrically activate a dormant communications module, convert it to an electrically active communications module, and transfer the sensor measurements to the electrically active communications module.

37. 35. The fluid infrastructure network management system of claim 34, wherein the fluid infrastructure network management manages a fluid infrastructure network including at least one water system selected from the group consisting of a groundwater system, a storm water system, and a wastewater system.

38. 38. The fluid infrastructure network management system of claim 37, wherein the sewer system includes one sewer system item including at least one element selected from the group consisting of a conduit, a reservoir, an orifice, a weir, a pump, and a wastewater treatment plant.

39. 38. The fluid infrastructure network management system of claim 37, wherein the groundwater system includes one groundwater system item including at least one element selected from the group consisting of an aquifer, a lake, a river, a pump, and a groundwater recharge system.

40. 40. The method of infrastructure management of claim 39, wherein the flow attributes include at least one attribute selected from the group consisting of hydraulic head, groundwater flux, groundwater flux rate, pH, conductivity, total organic compound concentration, total suspended solids, and turbidity.

41. a plurality of sensors, each sensor configured to generate a sensor measurement of a flow attribute inside the fluid infrastructure item; a plurality of remote telemetry units, each remote telemetry unit communicatively coupled to at least one of the sensors and having one of the digital twin models stored therein; a server communicatively coupled to a plurality of the remote telemetry units and having a plurality of digital twin models stored therein, such that a digital twin model on the server is substantially similar to the digital twin model stored on the remote telemetry units; a network through which the plurality of remote telemetry units are communicatively coupled to the server; A fluid infrastructure network management system comprising: The server, receiving a prediction of environmental conditions around an area surrounding at least one of the sensors during the instance in time or during the time period; estimating flow attributes inside the fluid infrastructure item during the instance in time or during the time period using the predictions of the environmental conditions and one of the digital twin models at one or more servers to generate estimated flow attribute values ​​for the servers; receiving or obtaining the sensor measurements of the flow attributes inside the fluid infrastructure item from the sensors to the remote telemetry unit; reporting one or more of the server's estimated flow attribute values ​​as being the sensor measurement during said instance in time or during said time period or as being an estimate of the sensor measurement during said instance in time or during said time period if said server does not receive said sensor measurement from said remote telemetry unit; 1. A fluid infrastructure network management system comprising: a microprocessor programmed with instructions to perform:

Images