TW202008280A - Deployment decision apparatus and method thereof for sensing elements in fluid distribution pipeline - Google Patents

Abstract

A deployment decision apparatus and a method thereof for sensing elements in fluid distribution pipe are provided. In the method, a fluid pipeline network model is established, and the fluid pipeline network model includes at least one fluid pipeline. At least one sensing element is deployed in the fluid pipeline. At least one status caused by leakage, obstruction, replacement, deformation or deterioration in the fluid pipeline are simulated, so as to obtain simulated value of each sensing element. A real deployment configuration of the sensing element is evaluated based on the simulated value through a convergence mechanism of machine learning technology. Accordingly, it is easier to find the best deployment place, type and number of the sensing element, and the method is adapted for leakage, source and obstruction evaluations, so as to position the factors faster.

Description

Arrangement decision device and method of sensing element in fluid transmission and distribution pipeline network

The invention relates to a machine learning technology and a problem positioning estimation and inspection technology of a pipeline network, and in particular to a deployment decision device and method of a sensing element in a fluid transmission and distribution pipeline network.

The planning and construction of water, oil, gas and other fluid pipeline networks is a symbol of the progress of modern civilization and an important foundation for urban development. These water, oil, gas and other energy sources not only promote the prosperity and progress of industry and commerce, but also are closely related to the daily food, clothing, housing, transportation and entertainment of the people. Therefore, the supply, transmission, distribution, inspection, and management of these fluid substances have always been very important issues for the government or industry. These liquid or gaseous fluid substances will be transmitted and distributed among the various supply sources, relay points, transfer points, storage stations, or users through complex pipeline networks. Generally speaking, the industry invests a huge amount of money every year to build many precision and expensive gauges, pressure gauges, flow meters, mass gauges and other sensing elements in the pipeline network, so as to monitor supply and distribution For management purposes. In addition to hoping to achieve the most efficient allocation, utilization, and effective management with limited resources, the industry also hopes that when problems occur in the pipeline, it can be dealt with and eliminated as soon as possible to reduce the risk of poor construction, pipe aging, and pipe Various reasons, such as abnormal pressure, cause leakage, blockage, alteration or deterioration of the pipe network, resulting in huge economic losses and inconvenience in life.

However, traditionally, there are no clear principles or reference basis for the deployment of these sensing elements, and manufacturers often build them at the source or terminal of the pipeline network based on experience. At the same time, when testing whether the sensing element is suitable for installation, the manufacturer not only has to carry out construction, which consumes time and money, but may also cause damage to the existing pipeline network and adversely affect it. In addition, the built-in sensing elements are often unable to measure the proper flow state in the pipeline network due to insufficient quantity. Or, the manufacturer built more unnecessary equipment, which caused waste, and there is no guarantee that the additional equipment will bring better measurement results. It can be seen that there are still many deficiencies in the above-mentioned idiomatic methods, which is not a good design and needs to be improved urgently.

In view of this, the present invention provides a deployment decision-making device and method for sensing elements in a fluid transmission and distribution pipeline network, which judges whether each pipeline is converged based on a machine learning convergence mechanism to serve as a sense in the actual pipeline network The basis for the deployment of test components.

The deployment decision method of the sensing element in the fluid transmission and distribution pipeline network of the present invention includes the following steps. The fluid pipe network model is established, and the fluid pipe network model includes at least one fluid pipeline. At least one sensing element is provided in these fluid lines. Simulate at least one state of the fluid line and obtain analog readings of each sensing element accordingly. Based on the analog readings of these sensing elements, the deployment settings of the sensing elements are determined.

On the other hand, the deployment decision device of the sensing element in the fluid transmission and distribution pipeline network of the present invention includes a storage and a processor. The memory records multiple modules. The processor is coupled to the storage, and accesses and loads the modules recorded in the storage. And those modules include model building module, sensing element management module, state management module and decision module. The model building module establishes a fluid pipe network model, and the fluid pipe network model includes at least one fluid pipeline. The sensing element management module sets at least one sensing element in these fluid lines. The state management module simulates at least one state of these fluid lines, and accordingly obtains simulated readings of each sensing element. The decision module determines the deployment setting of the sensing element based on the analog reading of the sensing element.

Based on the above, the decision-making device and method for the deployment of sensing elements in the fluid transmission and distribution pipeline network according to the embodiments of the present invention use the pipeline network simulation analysis software to establish the fluid pipeline network model, and then separately in each section of the fluid pipeline network model The interval positions of the pipelines are alternately added with models or parameters such as leaks or blockages that can change the status of the pipeline network, and then be simulated and calculated. Next, in the simulation process in which various pipeline factors occur in various sections of pipelines and at different positions, the embodiments of the present invention obtain analog readings of sensing elements such as pre-built pressure gauges and flow meters. These simulated readings will be imported into the machine learning algorithm in stages, so as to train the variable estimation model, and estimate the location of the status change of each pipeline separately. The convergence of these estimates to the expected value can determine whether to complete the learning and training. After completing the deployment setting of the learning training (the convergence result reaches the convergence standard), it can be used as the basis for the actual pipeline network in the deployment of the sensing element, so as to obtain the optimal deployment position.

In order to make the above-mentioned features and advantages of the present invention more obvious and understandable, the embodiments are specifically described below in conjunction with the accompanying drawings for detailed description as follows.

FIG. 1 is a block diagram of components of a deployment decision device 1 according to an embodiment of the invention. Referring to FIG. 1, the deployment decision device 1 includes at least but not limited to the storage 11 and the processor 13. The deployment decision device 1 may be a computing device such as a smart phone, tablet computer, desktop computer, notebook computer, server, or the like.

The storage 11 may be any type of fixed or removable random access memory (RAM), read only memory (Read Only Memory, ROM), flash memory (flash memory), or traditional hard disk (Hard Disk Drive, HDD), Solid-State Drive (SSD) or similar components, and used to record code, software modules (for example, model building module 111, state management module 113, sensing components Management module 115, decision module 117, etc.), fluid network model, simulated readings, machine learning calculation software, deployment settings (including setting location, quantity and type, etc.) and other data or files, the details of which will be followed Detailed description of the examples.

The processor 13 is coupled to the memory 11, and the processor 13 may be a central processing unit (Central Processing Unit, CPU), a microcontroller, a programmable controller, an application specific integrated circuit, a chip or other similar components or the above components The combination. In this embodiment, the processor 13 executes all operations of the deployment decision device 1, and the processor 13 can access and load those software modules recorded in the storage 11.

In order to facilitate understanding of the operation process of the embodiment of the present invention, a number of embodiments will be described in detail below to describe the deployment decision process of the sensing element in the fluid transmission and distribution pipeline network in the embodiment of the present invention. Hereinafter, the methods described in the embodiments of the present invention will be described with various devices, components, and modules in the deployment decision device 1. The various processes of the method can be adjusted according to the implementation situation, and it is not limited to this.

2 is a flowchart of a deployment decision method according to an embodiment of the invention. Please refer to FIG. 2. First, the model building module 111 creates a fluid pipe network model based on the numerical information of the actual fluid pipeline network status (step S210 ), and the fluid pipe network model includes one or more fluid pipes. Specifically, due to the development of theory and the advancement of technology, today's fluid pipeline network analysis and management can use pipe network simulation analysis software to first establish a pipe network model on a computer system. For example, the United States Environmental Protection Agency (EPA) assists utilities and consultants in the maintenance and management of water supply systems and the improvement of water quality. The EPANET software for water supply network hydraulic and water quality simulation analysis has been developed. This tool is often used as a calculation engine for water conservancy network analysis. For the analysis of the gas pipeline network, various commercial pipeline network analysis software such as PIPEFLOW can be used. In other words, the model building module 111 of this embodiment generates a fluid pipe network model by software simulation. The fluid pipe network model may be the same as the actual pipeline network or may be changed according to actual needs.

For example, FIG. 3 is an example of a fluid pipe network model. Referring to FIG. 3, a plurality of water supply and distribution pipelines 301 are used to connect from a plurality of water sources 303, through a water supply connection point or output node 302, a water supply pressure control device 304 and a water supply flow rate or flow control device 305, and finally distributed to Water storage facility 306.

Next, in the training data generation phase (step S230), the state management module 113 alternately adds pipeline problems such as leaks, blockages, replacements, displacements, deformations, or deteriorations at intervals between fluid pipelines to change the status to create or generate pipelines Variables to simulate at least one state of the fluid pipeline (step S231) (each state may include single or multiple types of pipeline variables, and the parameters of each pipeline variable (eg, location, size, strength, type, etc.) Can be adjusted), and obtain such as water meter, water gauge, water pressure gauge, differential pressure gauge, flow meter, flow meter, water quality meter, thermometer, gas meter, oil meter or other sensing for measuring the flow state of gas or liquid Analog reading of the element (step S233). Taking FIG. 3 as an example, the state management module 113 can add a leakage point model 311 to the water supply pipeline interval, and after the calculation by the pipe network simulation analysis software, it can extract a plurality of built-in ones that have been built on the water supply pipe network. Analog readings of the water flow sensing element 207 and the water pressure sensing element 208 that have been built.

Next, the decision module 117 performs machine learning training on each section of the fluid pipeline (step S250). Machine learning is a technique of artificial intelligence. It is a computer calculation method that uses mathematical optimization methods and high-speed computer calculations to automatically obtain laws from existing experience and predict unknown data. The training process of machine learning is mainly a process of using the calculation method to automatically change the parameter values in the machine learning model to make the estimated value approach the expected value. In addition, through the error convergence value (ie, the convergence result) between the estimated value and the expected value of the model, it can be judged whether the learning training has been completed (the convergence result is completed if it is not less than the convergence criterion; the convergence result is less than the convergence criterion is not carry out).

The purpose of the sensing element is to effectively detect the state and change of the fluid pipeline network. Therefore, when any section of pipeline is under the condition of the predetermined sensing element deployment, the actual position of the pipeline variable is used as the expected output, and the analog reading of the corresponding sensing element will be input to the machine learning for the variable position When the model is used to estimate the location of the variable but cannot meet the set convergence standard, it means that the deployment setting (or configuration) of the sensing element set in this section of pipeline is not good under the predetermined sensing element deployment conditions. The monitoring data cannot be used effectively. When the learning training of any section of pipelines still fails to meet the convergence standard under the specified number of learning trainings or time (the time or number of estimates of the estimated cause of each pipeline exceeds the threshold), the sensing element can be changed Position or deploy new sensing elements. In other words, the sensing element management module 115 can adjust the deployment position of the sensing elements in these pipe network models, add new sensing elements, or change the type (ie, change or adjust the deployment settings of the sensing elements) (Step S270). Each section of the pipeline (using node 302 as the intermediate point to distinguish the adjacent section of pipeline 301 as shown in FIG. 3) will be re-added into the pipeline cause and re-simulated calculation to obtain new simulated readings after the sensing element changes. Next, the decision module 117 will re-import these new simulated readings into the machine learning calculation software for learning training and adjust the deployment settings of the sensing elements accordingly, until the estimated position of all fluid pipelines has reached the machine learning training Up to the convergence criterion (for example, the convergence result is greater than the convergence criterion, and the convergence result is determined based on the difference between the actual position of the pipeline cause and the estimated cause) (step S270). At this time, the position of the sensing element of the pipe network model can be used as a decision basis for the construction and deployment of the sensing element in the actual fluid transmission and distribution pipeline network (step S290). In other words, the decision module 117 determines the deployment settings of the sensing elements based on the simulated readings of those sensing elements through the convergence mechanism of machine learning.

Taking FIG. 3 as an example, the decision module 117 uses the location data of each leak point model 311 and the corresponding sensing elements 307, 308 as the input information for the next stage of machine learning training and variable position estimation. The decision module 117 then adds or changes a plurality of proposed sensing elements (for example, the proposed water flow sensing element 209 and the proposed establishment based on the learning and training convergence value (or convergence result) of the machine learning (The water pressure sensing element 210 is to adjust the setting of the sensing element). Based on the adjusted deployment settings of the sensing element, it is determined that the module 117 will recalculate through the pipeline network simulation analysis software to obtain new simulated readings of the sensing element in the water supply pipeline network after re-deployment of the sensing element, And continue the follow-up learning and training process until the machine learning training value estimated by the position of the leakage point of each section of the pipeline (the location of the pipeline variable) reaches the convergence standard, then the type, number, and location of the sensing elements at this time can be (I.e., deployment setting) as the basis for decision-making in deployment.

Furthermore, the mean square error (MSE) convergence value of a certain fluid pipeline in the pipeline network during machine learning to estimate the location of the leak point (ie, the estimated variable location) is about 10^ -1 is not easy to converge, so it fails to meet the training convergence standard set less than 10^-3 (can be adjusted according to actual needs). The foregoing result is because the use of existing pressure sensors and flowmeters and other sensing elements is not sufficient to sense the location of the leakage point in this section of pipeline. Therefore, the sensing element management 115 can newly deploy a flow meter at the end of this pipeline, and re-import the new and existing sensing element measurement values into the machine learning software to learn and estimate the leakage point. training. If all pipelines at this time can meet the convergence standard of the MSE of less than 10^-3, it means that the current deployment status of the sensing element can effectively sense the status change of the pipeline network and is suitable as the actual pipeline The actual deployment location of the sensing element in the network.

Figure 4 shows a schematic diagram of one type of neural network algorithm in machine learning algorithms. Please refer to Figure 3 and Figure 4 at the same time. The decision module 117 first inputs the analog readings of the plurality of built-in sensing elements 306 read in the previous stage and the analog readings of the plurality of proposed sensing elements 307 to the input layer nerve of the neural network-like Element 401, then goes through a weighted value link 404. The decision module 117 can transfer the value of the simulated reading 407 of the proposed sensor element to the hidden layer neuron 402 and the output layer neuron 403 after weighting, and the neuron can produce the deviation value through the partial weight 405. Neural network-like learning and training is to optimize the value of the weighted value link 404 and the partial weight value 405 through repeated iterations (the convergence result is not converged, that is, adjusted according to the convergence result) to reduce the estimated value output 408 and approximate the expectation The output value corresponds to the input variable location. The decision module 117 may use the estimated and expected output mean squared error (MSE, Mean squared error) less than a specified value as the convergence criterion for learning and training.

In summary, the embodiments of the present invention use the convergence mechanism of pipe network simulation and machine learning, so that the optimal deployment type, number and position of the sensing elements in the fluid pipeline network can be found intelligently, quickly and effectively . In addition, the embodiments of the present invention can effectively reduce the number of deployment of sensing elements such as pressure gauges, flow meters, etc., so as to reduce the construction cost of physical equipment. The embodiments of the present invention will not cause damage or any adverse effects on the actual fluid transmission and distribution pipeline network, and can also be used to eliminate problems such as leak detection, traceability or blockage detection of the pipeline network. The embodiments of the present invention can be used to eliminate problems such as leak detection, traceability or blockage detection of water, oil, gas and other fluid pipeline networks to achieve the benefits of timeliness, saving, tracking, maintenance, and management.

Although the present invention has been disclosed as above with examples, it is not intended to limit the present invention. Any person with ordinary knowledge in the technical field can make some changes and modifications without departing from the spirit and scope of the present invention. The scope of protection of the present invention shall be subject to the scope defined in the appended patent application.

1‧‧‧Building decision device 11‧‧‧Storage 111‧‧‧Model building module 113‧‧‧ State management module 115‧‧‧ Sensing element management module 117‧‧‧Decision module 13‧‧ ‧Processor S210~S290‧‧‧Step 301‧‧‧Pipe 302‧‧‧Node 303‧‧‧Reservoir, water source 304‧‧‧Water pump 305‧‧‧Water valve 306‧‧‧Water tower, water tank 307‧‧‧ Flowmeter 308 has been built ‧‧‧Hydraulic pressure meter 309 has been built ‧‧‧ Flowmeter 310 is planned to be built ‧‧‧ Flowmeter 311 is planned to be built ‧‧‧ Leakage point model 401‧‧‧ Input layer neuron 402‧ ‧‧ Neuron in hidden layer 403‧‧‧ Neuron in input layer 404‧‧‧ Weighted value link 405‧‧‧Partial weight 406‧‧‧Sensor element analog reading input 407‧‧‧Proposed Sensing element analog reading input 408‧‧‧ estimated value output

FIG. 1 is a block diagram of components of a deployment decision device according to an embodiment of the invention. 2 is a flowchart of a deployment decision method according to an embodiment of the invention. Fig. 3 is an example illustrating a fluid pipe network model. 4 is a schematic diagram of a machine learning-like neural network algorithm according to an embodiment of the invention.

S210~S290‧‧‧Step

Claims (10)

A deployment decision method for sensing elements in a fluid transmission and distribution pipeline network includes: establishing a fluid pipe network model, wherein the fluid pipe network model includes at least one fluid pipeline; and at least one fluid pipeline is disposed in the at least one fluid pipeline A sensing element; simulating at least one state of the at least one fluid line, and accordingly obtaining an analog reading of the at least one sensing element; and determining the at least one sensing element based on the analog reading of the at least one sensing element Provisioning settings. The deployment decision method of the sensing element in the fluid transmission and distribution pipeline network as described in item 1 of the patent application scope, wherein the states are derived from at least one pipeline cause and are based on the at least one sensing element The step of determining the deployment setting of the at least one sensing element by the analog reading includes: using the actual position of the at least one pipeline variable as a desired output, and correspondingly using the analog reading of the at least one sensing element as an input, and Estimate the estimated variable position of the at least one pipeline variable through a variable estimation model of the machine learning algorithm; determine based on the actual position of the at least one pipeline variable and the at least one estimated variable position A convergence result; and determining whether the convergence result has converged to adjust the deployment setting of the at least one sensing element. As described in item 2 of the scope of the patent application, the deployment decision method of the sensing element in the fluid transmission and distribution pipeline network, wherein the at least one fluid pipeline includes a plurality of the fluid pipelines, and judges whether the convergence result is converged to adjust the After the step of the variable estimation model, the method further includes: if the convergence results of each of the fluid pipelines are converged, then the current deployment setting of the at least one sensing element is used as the deployment basis of the actual pipeline network, wherein the Provisioning settings include setting location, quantity, and type. As described in item 2 of the scope of the patent application, the deployment decision method of the sensing element in the fluid transmission and distribution pipeline network, wherein after the step of adjusting the deployment setting of the at least one sensing element, it further includes: if the at least one is estimated If the time or the number of times the estimated cause location exceeds the threshold value but still cannot converge, the deployment setting of the at least one sensing element is changed. The method for deciding the deployment of sensing elements in a fluid transmission and distribution pipeline network as described in item 2 of the patent application scope, wherein the step of simulating the at least one state of the at least one fluid pipeline includes: At least one such cause of leakage, blockage, displacement, displacement, deformation or deterioration is added to the at least one fluid line. A decision-making device for deployment of sensing elements in a fluid transmission and distribution pipeline network includes: a storage, recording a plurality of modules; and a processor, coupled to the storage, and accessed and loaded into the storage The recorded modules, and the modules include: a model building module to create a fluid pipe network model, wherein the fluid pipe network model includes at least one fluid pipeline; a sensing element management module, in the At least one sensing element is disposed in at least one fluid line; a state management module simulates at least one state of the at least one fluid line and obtains the simulated reading of the at least one sensing element accordingly; and a decision mode Group, the deployment setting of the at least one sensing element is determined based on the analog reading of the at least one sensing element. The decision-making device for the deployment of sensing elements in the fluid transmission and distribution pipeline network as described in item 6 of the patent scope, wherein the states are derived from at least one pipeline cause, and the decision module uses the at least one The actual position of the pipeline variable is used as the expected output, and the simulated reading of the at least one sensing element is correspondingly input, and the at least one pipeline is estimated through a variable estimation model trained based on the machine learning algorithm The estimated location of the variable, the decision module determines the convergence result based on the actual location of the at least one pipeline variable and the at least one estimated variable location, and the decision module determines whether the convergence result has converged to adjust The deployment setting of the at least one sensing element. The deployment decision device of the sensing element in the fluid transmission and distribution pipeline network as described in item 7 of the patent scope, wherein the at least one fluid pipeline includes a plurality of the fluid pipelines, and if each of the fluid pipelines If the convergence results are all converged, the decision module uses the current deployment settings of the at least one sensing element as the deployment basis of the actual pipeline network, where the deployment settings include the location, number, and type. The deployment decision device of the sensing element in the fluid transmission and distribution pipeline network as described in item 7 of the patent application scope, wherein if the time or number of times that the at least one estimated variable location is estimated exceeds the threshold but still cannot converge, Then, the sensing element management module changes the deployment setting of the at least one sensing element. As described in Item 7 of the patent application scope, the deployment decision device of the sensing element in the fluid transmission and distribution pipeline network, wherein the model building module adds leakage, blockage, At least one of the cause of substitution, displacement, deformation or deterioration.

Images