Multi-stage pressure regulation steam conveying network and optimization method thereof

Abstract

The invention provides a multi-stage pressure regulation steam conveying network and an optimization method thereof, and the method comprises the steps: judging the steam grade according to a preset steam grade division standard, and obtaining a steam grade distribution diagram in a pipe network; according to a steam demand order submitted by an enterprise, key steam parameters are extracted, the key steam parameters are matched with the pipe network steam level distribution diagram, a feasible conveying path from a steam source to a target enterprise is searched, and a target path is selected as a conveying scheme of the order; real-time parameters of a target node in the pipe network are continuously monitored, the real-time parameters are compared with the prediction parameters, when the deviation value exceeds a preset threshold value, it is judged that the conveying state is abnormal, and corresponding adjustment measures are taken; a pressure regulation optimization model is constructed, and a preset multi-stage pressure regulation method is regulated; when enterprises in the park change, the topological structure and the conveying path of a local pipe network are re-planned and are brought into a conveying scheduling system for unified management.

Description

Technical Field

[0001] The invention relates to the field of information technology, and in particular to a steam transmission network with multi-stage pressure regulation and an optimization method thereof. Background Art

[0002] The park's steam pipeline network faces a complex and ever-changing supply and demand relationship and operating environment. How to achieve efficient and accurate steam scheduling has become a major challenge. The traditional static scheduling method is difficult to adapt to the dynamic changes in the company's steam demand, resulting in resource waste and insufficient supply. At the same time, the pressure loss and parameter fluctuations during long-distance transportation make it difficult for the end of the pipeline network to stably meet user needs. In addition, changes in park enterprises have also brought uncertainty to the layout and scheduling of the pipeline network. These factors together lead to problems such as low efficiency of steam resource utilization, high energy consumption, and poor reliability. How to achieve real-time perception, precise control, and dynamic optimization of the overall pipeline network in a complex and changing environment is a core technical problem that needs to be solved urgently. In addition, enterprises may be added or cancelled in the park, and the steam demand of enterprises may also change. How to dynamically optimize and adjust the transportation strategy during the operation of the steam pipeline network to ensure the stability and efficiency of steam transportation is also a complex optimization problem. Summary of the invention

[0003] The present invention provides a steam delivery network with multi-stage pressure regulation and an optimization method thereof, which mainly includes:

[0004] Obtain the steam monitoring value of the target node of the park steam pipe network;

[0005] Determine the steam grade according to the preset steam grade classification standard and obtain the steam grade distribution map in the pipe network;

[0006] For the steam demand order submitted by the enterprise, extract the key steam parameters, match the key steam parameters with the steam grade distribution map of the pipeline network, search for feasible transportation paths from the steam source to the target enterprise, and select the target path as the transportation plan for the order;

[0007] Predict the steam parameters at the end of the pipeline according to the transportation plan, compare the prediction results with the steam demand of the target enterprise, and optimize the steam supply parameters until the target demand is met;

[0008] Continuously monitor the real-time parameters of the target nodes in the pipeline network, compare the real-time parameters with the predicted parameters, and when the deviation value exceeds the preset threshold, it is judged that the transportation status is abnormal and corresponding adjustment measures are taken;

[0009] Obtain steam pressure change data based on regulation measures, build a pressure regulation optimization model, and adjust the preset multi-stage pressure regulation method;

[0010] When there are changes in the enterprises within the park, the topological structure and transportation routes of the local pipeline network will be replanned and incorporated into the transportation scheduling system for unified management.

[0011] Furthermore, the obtaining of the steam parameters of the target node of the park steam network includes:

[0012] Obtain the pressure data, temperature data, and flow data of the target node; calculate the average value and standard deviation of each sensor in the recent time period based on the data; if the data point deviates from the average value by more than a preset multiple of the standard deviation, it is determined to be abnormal data; use the interpolation method to correct the abnormal data to obtain the corrected data.

[0013] Furthermore, judging the steam grade according to the preset steam grade classification standard to obtain the steam grade distribution map in the pipe network includes:

[0014] The steam level of each target node is determined according to the corrected data and the preset steam level classification standard; the target node is spatially interpolated by using the inverse distance weighted method to obtain an interpolation result; the park is gridded using the interpolation result to determine the steam level value of each grid unit; and a park steam level distribution map is generated according to the steam level value.

[0015] Furthermore, the steam parameters at the end of the pipeline are predicted according to the transportation scheme, and the prediction results are compared with the steam demand of the target enterprise to optimize the steam supply parameters until the target demand is met, including:

[0016] The pipeline length, diameter, insulation material parameters and pipeline roughness are obtained according to the steam transmission path; the pressure loss and heat loss are calculated; the predicted steam parameters at the end of the pipeline are obtained according to the pressure loss and heat loss calculation results; the predicted steam parameters are compared with the steam demand of the target enterprise, and if the pressure error or the temperature error is greater than a threshold, the gradient descent method is started to optimize the steam supply parameters.

[0017] Furthermore, the real-time parameters of the target nodes in the pipeline network are continuously monitored, and the real-time parameters are compared with the predicted parameters. When the deviation value exceeds a preset threshold, it is determined that the transportation state is abnormal, and corresponding adjustment measures are taken, including:

[0018] Collect target node pressure data, temperature data, and flow data; perform lossless compression on the data to obtain compressed data; compare the compressed data with pre-stored prediction parameters, and use a sliding window method to calculate the average deviation value to obtain pressure deviation, temperature deviation, and flow deviation; if at least one of the pressure deviation, temperature deviation, and flow deviation exceeds a preset threshold, trigger an early warning mechanism.

[0019] Furthermore, the steam pressure change data is obtained according to the regulation measures, a pressure regulation optimization model is constructed, and the preset multi-stage pressure regulation method is adjusted, including:

[0020] Acquire real-time pressure data of each node in the steam network; perform data cleaning and normalization processing based on the real-time pressure data to obtain a standardized pressure change data set; use dynamic time warping distance as a similarity measure to perform hierarchical clustering on the standardized pressure change data set, divide the pressure change data into multiple levels, and obtain a multi-level pressure change data set; based on the multi-level pressure change data set, construct a long short-term memory network as a pressure regulation optimization model.

[0021] Furthermore, when there are changes in enterprises within the park, the topological structure and transportation path of the local pipeline network are replanned and incorporated into the transportation scheduling system for unified management, including:

[0022] Obtain geographic information data of the park's steam pipeline network; establish a digital twin model based on the geographic information data; use the Dijkstra algorithm to calculate the optimal transmission path and pressure distribution plan; receive access signals from new enterprises; use the Steiner tree algorithm to replan the local pipeline network structure based on the access signals from the new enterprises; obtain steam demand data from the new enterprises, and use the ARIMA model to predict future steam consumption.

[0023] The technical solution provided by the embodiment of the present invention may have the following beneficial effects:

[0024] The present invention discloses a multi-stage pressure-regulated steam transportation network and an optimization method thereof. The method collects data in real time by installing sensors at key nodes of the pipeline network, establishes a steam grade distribution map, and matches the optimal transportation path according to the enterprise order requirements. After determining the path, the present invention calculates the loss during the transportation process, predicts the parameters at the end of the pipeline, and adjusts the upstream parameters to meet the demand when necessary. During the transportation process, the present invention continuously monitors the real-time parameters, triggers an early warning and takes adjustment measures when the deviation exceeds the threshold. By constructing a pressure regulation optimization model, the present invention can continuously improve the regulation technology. In addition, the present invention can also flexibly adjust the transportation scheduling system according to changes in park enterprises to achieve dynamic optimization of the pipeline network. This method significantly improves the operating efficiency and reliability of the park steam pipeline network, and realizes accurate scheduling of steam resources and energy conservation and emission reduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The present invention is a flow chart of a steam transport network with multi-stage pressure regulation and an optimization method thereof.

[0026] Figure 2 A schematic diagram of a steam delivery network with multi-stage pressure regulation and an optimization method thereof according to the present invention. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on the embodiments in this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of this specification.

[0028] like Figure 1-2 The steam delivery network with multi-stage pressure regulation and the optimization method thereof in this embodiment may specifically include:

[0029] S101. Install pressure, temperature and flow sensors at the target nodes of the park steam network, collect steam monitoring values ​​of the target nodes in the network in real time, transmit the collected data to the data processing center, establish steam grade classification standards in advance, determine the steam grade of each point according to the steam grade classification standards, and obtain the steam grade distribution map in the network.

[0030] Obtain the pressure data, temperature data, and flow data of the target node, the data being collected by sensors in the steam network; calculate the average value and standard deviation of each sensor in a recent time period based on the data, and determine that a data point is abnormal if it deviates from the average value by more than a preset multiple of the standard deviation; use an interpolation algorithm to correct the abnormal data to obtain corrected data; determine the steam grade of each target node based on the corrected data and a preset steam grade classification standard; use an inverse distance weighted method to perform spatial interpolation on the target node to obtain an interpolation result; use the interpolation result to grid the park to determine the steam grade value of each grid unit; generate a park steam grade distribution map based on the steam grade value, and the distribution map uses different colors to represent different steam grades.

[0031] Exemplarily, the pressure, temperature, and flow data of the target node of the park steam network are collected, and the collected real-time data is transmitted to the data processing center through the data transmission network. The data of each node is classified and processed according to the pre-established steam grade classification standard to generate a steam grade distribution map in the network. For the data collected by each sensor, the average value and standard deviation in the last 10 minutes are calculated. If a data point deviates from the average value by more than 3 times the standard deviation, it is marked as abnormal data. For abnormal data points, linear interpolation is used to correct them to ensure the continuity and reliability of the data. According to the preset steam grade classification standards, such as pressure range, temperature range and flow range, the steam grade of each node is judged. If the pressure is between 0.6-0.8MPa, the temperature is between 150-180℃, and the flow rate is between 50-100t / h, it is judged as medium-pressure steam. For high-pressure and low-pressure steam, the corresponding parameter range is set for judgment. Based on the geographical location information of the target node and the steam grade judgment result, the inverse distance weighted method is used for spatial interpolation. For locations where sensors are not installed in the network, the steam grade of the location is calculated based on the steam grade and distance of the surrounding known nodes. The closer the known node is, the greater its influence weight is. Using the interpolation results, the entire park is gridded, and each grid unit is assigned a corresponding steam grade value. Different colors are used to represent different steam grades, and a heat map of the steam grade distribution of the entire park is generated. The high-pressure steam area is represented by red, the medium pressure is represented by yellow, and the low pressure is represented by green. The continuous change of steam grade is displayed by color gradient, and the steam quality status of different areas in the pipe network is intuitively displayed. Pressure, temperature, and flow sensors are installed at the target nodes of the park steam pipe network, and the sensor sampling frequency is set to once every 10 seconds. The collected data is transmitted to the data processing center via industrial Ethernet with a transmission rate of 100Mbps. After receiving the real-time data, the data processing center first performs data preprocessing. The average value and standard deviation of each sensor are calculated with a time window of 10 minutes. For example, a pressure sensor collects 60 data points in 10 minutes, and the average pressure is calculated to be 0.7MPa and the standard deviation is 0.05MPa. If the pressure value of a data point exceeds the range of 0.7±0.15MPa, it is marked as abnormal data. For abnormal data, linear interpolation is used for correction. The corrected data enters the next step of processing together with normal data. According to the preset steam grade classification standard, the steam grade of each node is determined. High-pressure steam is defined as pressure greater than 0.8MPa, temperature higher than 180℃, and flow greater than 100t / h; medium-pressure steam is defined as pressure between 0.6-0.8MPa, temperature between 150-180℃, and flow between 50-100t / h; low-pressure steam is defined as pressure less than 0.6MPa, temperature less than 150℃, and flow less than 50t / h. For locations in the pipeline network where sensors are not installed, the inverse distance weighted method is used for spatial interpolation.Suppose there is an unknown point P, surrounded by three known points A, B, and C, with distances of 10m, 15m, and 20m respectively. The steam levels of points A, B, and C are high pressure, medium pressure, and low pressure, respectively. When calculating the steam level of point P, the weight of point A is 1 / 10, point B is 1 / 15, and point C is 1 / 20. The steam level of point P is calculated based on the weight. The entire park is gridded, and the grid size is set to 5m×5m. Each grid cell is assigned a corresponding steam level value based on the interpolation result. Finally, a steam level distribution heat map is generated, using red, yellow, and green to represent high, medium, and low pressure steam areas, respectively. The color gradient area represents the transition area of ​​the steam level, which intuitively displays the steam quality distribution status in different areas of the pipeline network.

[0032] S102. For the steam demand order submitted by the enterprise, extract the key steam parameters of the target in the order, which include grade, pressure, and temperature. Match the key steam parameters with the steam grade distribution map of the pipeline network, search for feasible transportation paths from the steam source to the target enterprise, calculate the transportation efficiency and energy consumption of each path, and select the target path as the transportation plan for the order.

[0033] A steam demand order is obtained, and key steam parameters are extracted from the steam demand order, wherein the key steam parameters include steam grade, pressure and temperature requirements; a matching search is performed in a distribution map of steam grade in a pipeline network according to the key steam parameters, and the pipeline network nodes are organized by geographical location using an R-tree spatial index structure; if the query range meets the preset conditions, the steam source nodes and the location nodes of the target enterprise in the pipeline network that meet the requirements are determined; according to the steam source nodes and the target enterprise nodes, a plurality of feasible transportation paths are searched using the Dijkstra algorithm, and the weights of the feasible transportation paths are calculated by the pipeline length, the pressure loss rate and the temperature decay rate; for the plurality of feasible transportation paths, the transportation efficiency and energy consumption of each path are estimated, and the thermodynamic calculation model includes pipeline friction loss calculation and heat loss calculation; according to the transportation efficiency and energy consumption, a comprehensive score of each feasible transportation path is calculated; it is determined whether the comprehensive score is the highest, and if it is the highest, the corresponding feasible transportation path is determined as the preliminary transportation plan for the order.

[0034] For example, key steam parameters, including steam grade, pressure and temperature requirements, are extracted from steam demand orders submitted by enterprises, and regular expressions are used to match the values ​​and units in the order text, such as "pressure: 0.8MPa", "temperature: 180℃", etc. The extracted parameters are stored as structured data in key-value pair format. According to the extracted key steam parameters, a matching search is performed in the steam grade distribution map of the pipeline network, and the pipeline network nodes are organized by geographical location using the R-tree spatial index structure. According to the extracted parameters, the query range is set, such as pressure ±0.1MPa, temperature ±10℃, and the steam source nodes that meet the requirements are quickly located, and the location nodes of the target enterprise in the pipeline network are determined at the same time. Using the shortest path algorithm in graph theory, such as the Dijkstra algorithm, multiple feasible transmission paths from the steam source node to the target enterprise node are searched, taking into account factors such as pipeline length, pressure loss and temperature attenuation. When applying the Dijkstra algorithm, the pipeline length, pressure loss and temperature decay are converted into edge weights. The weight calculation formula is weight = pipeline length * (1 + pressure loss rate + temperature decay rate). The comprehensive score of each path is calculated by comprehensively considering multiple factors. For the multiple transportation paths searched, the thermodynamic calculation model is used to estimate the transportation efficiency and energy consumption of each path. The thermodynamic calculation model includes: pipeline friction loss calculation, using the Darcy-Weisbach equation; heat loss calculation, considering the pipeline insulation layer, using the heat conduction formula; transportation efficiency calculation, output steam energy divided by input steam energy; energy consumption calculation, considering the power consumption and heat loss of the pump. Taking into account factors such as pipeline friction and heat loss, the path with the highest comprehensive score is selected as the final transportation plan for the order. After the enterprise submits the steam demand order, the regular expression "(\d+(.\d+)?)([MPa|℃])" is used to match the numerical value and unit in the order text. For example, from the text "Demand: medium-pressure steam, pressure: 0.8MPa, temperature: 180℃", the pressure of 0.8MPa and the temperature of 180℃ are extracted. The extracted parameters are stored as key-value pairs {"pressure":0.8,"temperature":180}. Next, a matching search is performed in the steam grade distribution map of the pipeline network. The pipeline network nodes are organized using the R-tree spatial index structure, and each node contains location coordinates and steam parameters. The query range is set to pressure ±0.1MPa and temperature ±10℃, that is, 0.7-0.9MPa and 170-190℃. The R-tree quickly locates the steam source nodes that meet the conditions and determines the target enterprise nodes. Then, the Dijkstra algorithm is applied to search for feasible transmission paths. The pipeline length, pressure loss, and temperature attenuation are converted into edge weights. Assuming that a pipeline is 100 meters long, the pressure loss rate is 0.01 MPa / 100m, and the temperature decay rate is 1°C / 100m, the weight of the edge is calculated as 100*(1+0.01+0.01)=102. The algorithm traverses all possible paths, calculates the total weight, and finds multiple paths with the smallest weight. Finally, the searched paths are thermodynamically calculated.The Darcy-Weisbach equation is used to calculate the friction loss of the pipeline. Assuming the pipe diameter is 100mm, the flow rate is 20m / s, and the friction coefficient is 0.02, the pressure loss of a 100-meter pipeline is approximately 0.04MPa. The heat loss calculation takes into account the pipeline insulation layer. Assuming the insulation layer thickness is 50mm, the thermal conductivity is 0.04W / (m·K), and the ambient temperature is 20℃, the temperature loss of a 100-meter pipeline is approximately 2℃. The transportation efficiency is calculated as the output steam energy divided by the input steam energy, considering pressure and temperature losses, and assuming an efficiency of 95%. The energy consumption calculation includes pump power consumption and heat loss. Assuming the pump power is 5kW, the heat loss is 2kW, and the total energy consumption is 7kW. The comprehensive scores of each path are comprehensively evaluated, and the path with the highest score is selected as the final transportation plan.

[0035] Among them, on the basis of the preliminary transportation plan, a set of enterprises with overlapping paths is identified, and for the overlapping path segments, it is determined whether the pressure and temperature requirements of the enterprises involved match. If they match, the original plan is maintained, and if they do not match, alternative paths are re-searched through a dynamic programming algorithm until the needs of all enterprises are met to form a final transportation plan; which includes: according to the preliminary transportation plan, the steam demand parameters of each enterprise are obtained, including pressure, temperature and flow requirements; the paths in the preliminary transportation plan are traversed using a graph theory algorithm to identify a set of enterprises with overlapping paths; for the overlapping path segments, the pressure and temperature demand parameters of the enterprises involved are extracted, and whether the needs of each enterprise match is determined by similarity calculation; if the matching degree of the demand parameters of the enterprises involved in the overlapping path segments is greater than the parameter matching degree threshold, the original transportation plan is maintained unchanged; If the matching degree of the demand parameters of the enterprises involved in the overlapping path segment is less than the parameter matching degree threshold, the path segment is marked as needing to be replanned; for the overlapping path segments that need to be replanned, the alternative paths are searched through the dynamic programming algorithm, and factors such as pipeline length, pipe diameter, and flow constraints are considered to obtain multiple alternative paths; the alternative paths are evaluated and scored, and the pipeline laying cost, transportation efficiency, and energy consumption are comprehensively considered. The path with the best score is selected as an alternative to the overlapping path segment; the replanned path replaces the overlapping path segment in the original transportation plan to form a new transportation plan; it is determined whether the new transportation plan meets the steam demand of all enterprises. If it meets the requirements, it is determined as the final transportation plan. If not, it is iteratively executed. The similarity calculation is performed to form a new transportation plan step until the requirements of all enterprises are met or the upper limit of the number of iterations is reached. Exemplarily, the steam demand parameters of each enterprise are first obtained according to the preliminary transportation plan. For example, enterprise A needs steam with a pressure of 2MPa, a temperature of 300℃, and a flow rate of 5t / h, and enterprise B needs steam with a pressure of 0MPa, a temperature of 280℃, and a flow rate of 3t / h. Then, the Dijkstra algorithm was used to traverse the paths in the preliminary transportation plan, and it was identified that there was a 20km overlap in the paths of the three companies A, B, and C. For this overlapping path segment, the pressure and temperature demand parameters of the three companies were extracted, and the matching degree of each company's needs was determined by cosine similarity calculation. If the matching degree is greater than the threshold of 9, the original plan is maintained, and if it is less than 9, the path segment is marked as needing to be replanned. Then, the dynamic programming algorithm was used to search for alternative paths, and two alternative paths were obtained under the constraints of pipeline length not exceeding 25km, pipe diameter DN200, and flow rate 10t / h. The alternative paths were evaluated and scored, and the path with a total score of 85 was selected as an alternative plan based on comprehensive indicators such as pipeline laying cost, transportation efficiency, and pump power and energy consumption. The replanned path replaced the overlapping section in the original plan to form a new transportation plan. Finally, it was determined whether the new plan met the needs of all companies. If it did, it was determined as the final plan. If it did not, the above steps were iteratively executed until the needs were met or the upper limit of 10 iterations was reached, forming a complete multi-enterprise steam transportation path optimization plan.

[0036] S103. After determining the steam transmission path, calculate the pressure loss and heat loss during the transmission process according to the length, diameter and insulation material parameters of the pipeline, and predict the steam parameters at the end of the pipeline. If the predicted parameters meet the steam demand of the target enterprise, directly execute the transmission plan; if the predicted parameters do not meet the demand, adjust the steam supply parameters of the upstream steam source, recalculate and optimize the transmission plan until the target demand is met.

[0037] According to the steam transmission path, the pipeline length, diameter, insulation material parameters and pipeline roughness are obtained; the pressure loss and heat loss are calculated; and the predicted steam parameters at the end of the pipeline are obtained according to the pressure loss and heat loss calculation results. The predicted steam parameters are compared with the steam demand of the target enterprise. If the pressure error between the predicted parameters and the steam demand is greater than a threshold or the temperature error is greater than a threshold, the gradient descent method is started to optimize the steam supply parameters. In the gradient descent method, the objective function is defined as the sum of the square differences between the predicted parameters and the demand parameters, and the partial derivatives of the objective function with respect to each steam supply parameter are calculated, and the steam supply parameters of the upstream steam source are adjusted along the gradient direction. The adjusted steam supply parameters are substituted into the Darcy-Weisbach equation and the heat conduction formula, and the pipeline end parameters are recalculated. If the objective function value is less than the preset value or the number of iterations reaches the preset number, the optimal steam supply parameters are output.

[0038] For example, according to the determined steam transport path, the length, diameter, insulation material parameters and pipeline roughness of the pipeline are obtained from the database, and the pressure loss is calculated using the Darcy-Weisbach equation.

[0039]

[0040] In the calculation formula, ΔP represents the pressure loss in the pipeline, f is the friction factor, L is the pipe length, D is the pipe diameter, ρ is the steam density, and v is the flow rate. The heat loss is calculated using the heat conduction formula

[0041]

[0042] , Q represents the heat loss per unit time, Δt represents the time interval, k represents the rafting coefficient, A represents the surface area of ​​the pipeline, T1 represents the steam temperature, T2 represents the ambient temperature, and d represents the thickness of the insulation layer. The predicted steam parameters at the end of the pipeline are obtained according to the calculation results. The predicted steam parameters at the end of the pipeline are compared with the steam demand of the target enterprise, and the parameter matching threshold is set. The pressure error is ±0.05MPa and the temperature error is ±5℃. If the predicted parameters are within the threshold range, it is determined to meet the demand, and the transportation plan is marked as executable and stored in the plan database. If the predicted parameters do not meet the demand, the gradient descent method is started to optimize the steam supply parameters. The objective function is defined as the sum of the square differences between the predicted parameters and the demand parameters, and the partial derivatives of the objective function with respect to each steam supply parameter are calculated. The steam supply parameters of the upstream steam source are automatically adjusted along the gradient direction, including pressure, temperature and flow. The iterative calculation method is adopted, and the maximum number of iterations is set to 100 times, and the convergence threshold is that the objective function value is less than 0.01. Substitute the adjusted steam supply parameters into the Darcy-Weisbach equation and heat conduction formula, recalculate the pipeline end parameters, and compare them with the target requirements. After each iteration, check whether the termination conditions are met. If satisfied, output the optimal steam supply parameters; if not satisfied, continue to iterate until the optimal steam supply parameter combination that meets the requirements is found or the maximum number of iterations is reached. After determining the steam transmission path, the pipeline parameters extracted from the database include a length of 1000 meters, a diameter of 200 mm, a thermal conductivity of the insulation material of 0.04W / (m·K), an insulation layer thickness of 50 mm, and a pipeline roughness of 0.045 mm. Use the Darcy-Weisbach equation to calculate the pressure loss, assuming that the steam density is 5kg / m 3 , the flow rate is 20m / s, the friction factor f is calculated to be 0.015, and the final pressure loss is 0.15MPa. The heat loss is calculated using the heat conduction formula. Assuming that the initial steam temperature is 200℃ and the ambient temperature is 20℃, the heat loss is calculated to be 50kW. Based on these calculation results, the predicted steam parameters at the end of the pipeline are pressure 0.85MPa and temperature 185℃. The steam demand of the target enterprise is pressure 0.8±0.05MPa and temperature 180±5℃. Comparing the predicted parameters with the demand parameters, it is found that the pressure is within the threshold range, but the temperature is slightly higher. Therefore, the gradient descent method is started to optimize the steam supply parameters. The objective function is defined as (0.85-0.8) 2 +(185-180) 2=25.0025, calculate the partial derivatives of the objective function with respect to pressure and temperature, and obtain the adjustment direction. The first iteration adjusted the steam supply parameters to reduce the pressure by 0.02MPa and the temperature by 2°C. Substitute the new parameters into the calculation model to obtain new terminal prediction parameters: pressure 0.83MPa, temperature 183°C. The objective function value is reduced to 9.49, which is less than the set convergence threshold of 0.01, but still not optimal. Continue iterative calculations, fine-tune the steam supply parameters each time and re-evaluate the terminal situation. After 15 iterations, the optimal steam supply parameters are finally determined to be 0.95MPa pressure and 188°C temperature. At this time, the predicted terminal parameters fully meet the needs of the target enterprise. This set of optimal parameters is stored in the database as the final transportation plan.

[0043] S104. During the steam transportation process, the real-time parameters of the target nodes in the pipeline network are continuously monitored, the real-time parameters are compared with the predicted parameters, and the deviation value is calculated. When the deviation value exceeds the preset threshold, it is judged that the transportation state is abnormal, and an early warning is automatically triggered. According to the deviation size and abnormal location, corresponding adjustment measures are taken. The adjustment measures include adjusting the valve opening and pressure difference distribution, and increasing or decreasing the steam supply from the steam source.

[0044] The pressure data, temperature data and flow data of the target node in the distributed sensor network are collected, and the data are obtained by real-time collection; the data are losslessly compressed to obtain compressed data; the compressed data are compared with the pre-stored prediction parameters, and the average deviation value is calculated by a sliding window algorithm to obtain the pressure deviation, temperature deviation and flow deviation; it is determined whether the pressure deviation, temperature deviation and flow deviation exceed the preset threshold value, and if at least one of the pressure deviation, temperature deviation and flow deviation exceeds the preset threshold value, an early warning mechanism is triggered; an abnormal classification result is generated according to the early warning mechanism, and the abnormal classification result includes the abnormal type and severity; an adjustment instruction is automatically generated according to the abnormal classification result, and the adjustment instruction is used to adjust the opening of the specified valve; after executing the adjustment instruction, the changes in the pressure, temperature and flow data are continuously monitored.

[0045] Exemplarily, the pressure, temperature and flow data of the target node in the pipe network are collected in real time through a distributed sensor network, and the collected data are losslessly compressed using Huffman coding, and the compressed data is transmitted to the central control system. After receiving the real-time data, the central control system compares it with the pre-stored prediction parameters, and uses a sliding window algorithm to calculate the average deviation value in the last hour to obtain the pressure deviation, temperature deviation and flow deviation. When calculating the deviation value, the relative error formula is used: deviation rate = (measured value-predicted value) / predicted value*100%, and the deviation rate is calculated for pressure, temperature and flow respectively. The abnormal state is judged according to the preset threshold. If the deviation value exceeds the threshold, the early warning mechanism is triggered. The early warning mechanism uses a decision tree to classify the abnormal type. The root node of the decision tree is the deviation type (pressure / temperature / flow), the second layer is the deviation size (high / medium / low), the third layer is the abnormal duration (short / medium / long), and the leaf node is the specific abnormal type and severity to determine the abnormal location and severity. Based on the abnormal classification results, the adjustment instructions are automatically generated, including adjusting the opening of the specified valve, redistributing the pressure difference of each pipe section, and sending control signals to the steam source to increase or decrease the steam supply. According to the type and severity of the abnormality, the corresponding adjustment scheme is selected from the preset adjustment rule library. For example, for the case of moderately high pressure and medium duration, an instruction to reduce the upstream valve opening by 5% is generated. After executing the adjustment instruction, the parameter changes are continuously monitored for 30 minutes. If the deviation does not improve, the secondary adjustment is triggered to achieve automatic adjustment and optimization of the steam transportation process. During the steam transportation process, the distributed sensor network collects the pressure, temperature, and flow data of the target node in the pipeline network every 5 seconds. Taking a certain node as an example, the collected raw data is pressure 0.82MPa, temperature 178℃, and flow 50t / h. This set of data is compressed using Huffman coding, with a compression rate of 40%, and the compressed data is transmitted to the central control system. After receiving the data, it is decompressed and compared with the predicted parameters, including pressure 0.80MPa, temperature 180℃, and flow 48t / h. A 60-minute sliding window is used to calculate the average deviation rate, with a pressure deviation of 2.5%, a temperature deviation of -1.1%, and a flow deviation of 4.2%. The preset abnormal thresholds are pressure ±5%, temperature ±3%, and flow ±10%. At this time, the temperature and flow are within the normal range, but the pressure is close to the abnormal threshold. The decision tree analysis results show that the pressure deviation type is "moderately high" and the duration is "medium", which is 30-60 minutes. According to the regulation rule base, the regulation instruction is automatically generated: reduce the opening of the upstream valve A by 3%. After executing the instruction, continue monitoring for 30 minutes, and find that the pressure drops to 0.81MPa and the deviation rate drops to 1.25%, which determines that the regulation is effective. If the pressure still exceeds 0.83MPa after adjustment, it will trigger secondary regulation to further reduce the valve opening or reduce the steam supply from the steam source.

[0046] S105. According to the regulation measures, analyze the adjustment of valve opening and pressure difference distribution, obtain steam pressure change data, divide the pressure change data into multiple levels, build a pressure regulation optimization model, use the multi-level pressure change data as model input, obtain multi-level pressure regulation optimization results, and adjust the preset multi-level pressure regulation method.

[0047] The real-time pressure data of each node of the steam network is obtained, and the real-time pressure data is collected by a data collector; data cleaning and normalization are performed according to the real-time pressure data to obtain a standardized pressure change data set, wherein the data cleaning adopts the 3σ principle to identify abnormal values, and the normalization adopts the Min-Max method to scale the data to the [0, 1] interval; the dynamic time warping distance is used as a similarity measure to perform hierarchical clustering on the standardized pressure change data set, and the pressure change data is divided into five levels: very low, low, medium, high, and very high, to obtain a multi-level pressure change data set; based on the multi-level pressure change data set, a long short-term memory network is constructed as a pressure regulation optimization model, and the long short-term memory network uses the past pressure change data as input features and the future pressure prediction value as the output target; according to the output result of the pressure regulation optimization model, the preset multi-level pressure regulation method is improved, and if the predicted pressure exceeds the preset threshold, the relevant valve opening is adjusted.

[0048] For example, according to the adjustment records of valve opening and pressure difference distribution, a data collector is used to obtain the real-time pressure data of each node in the steam network, and a standardized pressure change data set is generated through data cleaning and normalization. The data cleaning process uses the 3σ principle to identify and process outliers, and the normalization uses the Min-Max method to scale the data to the [0,1] interval. Using the dynamic time warping distance as a similarity measure, combined with the hierarchical clustering method, the pressure change data is divided into five levels: very low, low, medium, high, and very high, and a multi-level pressure change data set is obtained. Based on the multi-level pressure change data set, a long short-term memory network is constructed as a pressure regulation optimization model, and the pressure change data of the past 24 hours is used as the input feature, and the pressure prediction value of the next hour is used as the output target. The model parameters are trained and optimized by the cross-validation method. The trained long short-term memory network is applied to the new pressure change data to obtain the multi-level pressure regulation optimization result. A rule-based decision system is designed based on the optimization results. If the predicted pressure will exceed the preset threshold, the relevant valve opening is adjusted in advance. At the same time, a feedback mechanism is established to compare the error between the predicted value and the actual value, and the model is retrained every week to improve accuracy. According to the output of the decision system, the preset multi-level pressure regulation technology is improved, the valve opening control strategy and the pressure difference distribution plan are adjusted, and the regulation technology parameter library is updated. During the steam network optimization process, the data collector collects pressure data from 100 nodes every 5 minutes. The original data range is 0.5-1.2MPa, and 0.2% of the outliers are eliminated after processing by the 3σ principle. The data is mapped to the [0,1] interval using Min-Max normalization, such as converting 0.8MPa to 0.5. For the processed 24-hour data, a total of 2880 data points are applied to calculate the similarity using the dynamic time warping distance, and the maximum window width is set to 60 to obtain the similarity matrix. The hierarchical clustering method is used, and the number of clusters is set to 5, and the pressure changes are divided into five levels: extremely low 0-0.2, low 0.2-0.4, medium 0.4-0.6, high 0.6-0.8, and extremely high 0.8-1.0. Construct a long short-term memory network, with 24 neurons in the input layer, corresponding to 24 hours of data, 128 LSTM units in the hidden layer, and 1 neuron in the output layer, which is used to predict the pressure in the next hour. Use the data from the past 30 days for training, set the batch size to 64, the learning rate to 0.001, and train for 100 epochs. Apply the trained model, input the data of the last 24 hours, such as 0.75, 0.78, 0.82..., and predict that the pressure in the next hour is 0.85, which is the normalized value. Based on the prediction results, the decision system determines that the pressure is about to exceed the threshold of 0.8, and automatically generates instructions: reduce the opening of the upstream valve A by 5%, and increase the pressure difference distribution of branch B by 2%. After the adjustment, the actual pressure measured after 1 hour is 0.83, and the error with the predicted value of 0.85 is 2.35%.The prediction error was recorded and after accumulating one week of data, the model was retrained using the newly added 10,080 data points. The model prediction accuracy increased from the original 94.5% to 95.8%.

[0049] S106. When there are changes in enterprises within the park, a transportation and dispatching system is established. When a new enterprise is connected to the steam pipeline network, the topological structure and transportation route of the local pipeline network are replanned according to its geographical location and steam demand, and it is included in the entire transportation and dispatching system for unified management. When an enterprise exits the pipeline network, it is removed from the dispatching system to release the transportation resources it occupies, and the operation strategy of the surrounding pipeline network is optimized and adjusted.

[0050] Obtain geographic information data of the park steam pipeline network, the geographic information data including pipeline network topology, enterprise distribution and steam demand data; establish a digital twin model based on the geographic information data, the digital twin model updates the pipeline network topology, enterprise distribution and steam demand data in real time; use the Dijkstra algorithm to calculate the optimal transmission path and pressure distribution plan; receive a new enterprise access signal, the new enterprise access signal includes the geographic coordinates and steam usage parameters of the new enterprise; according to the new enterprise access signal, use the Steiner tree algorithm to re-plan the local pipeline network structure, the Steiner tree algorithm calculates the optimal pipeline connection plan from the nearest steam source to the new enterprise; obtain the steam demand data of the new enterprise, and use the ARIMA model to predict future steam consumption; according to the prediction results of the ARIMA model, use the weighted average method to integrate the prediction results of the new enterprise with the existing load prediction; receive the enterprise exit pipeline network signal, the enterprise exit pipeline network signal includes the identification information of the exiting enterprise; according to the enterprise exit pipeline network signal, use the genetic algorithm to re-optimize the pressure distribution and flow control parameters of the surrounding pipeline network, the genetic algorithm uses real number coding to represent the pressure and flow of each pipe section.

[0051] For example, a digital twin model of the park steam pipeline network based on the geographic information system is constructed to update the pipeline network topology, enterprise distribution and steam demand data in real time, and the optimal transmission path and pressure distribution plan are calculated by the Dijkstra algorithm, taking into account the pipeline length, diameter and pressure loss factors. When the access signal of a new enterprise is detected, its geographic coordinates and steam parameters are automatically obtained, and the local pipeline network structure is replanned using the Steiner tree algorithm considering capacity constraints, and the optimal pipeline connection plan from the nearest steam source to the new enterprise is calculated. At the same time, the pipeline network pressure balance calculation is performed to ensure that the new connection does not affect the stability of the existing system. The steam demand data of the new enterprise is input into the load forecasting model, and the ARIMA model is used to predict the steam consumption in the next week, and the weighted average method is used to integrate the forecast results of the new enterprise with the existing load forecast. According to the forecast results, the adaptive control algorithm is used to dynamically adjust the valve opening and pump speed of the relevant pipe section to optimize the transmission capacity and pressure distribution. If the enterprise exits the pipeline network signal, the resource release program is started, the enterprise is removed from the scheduling database, and the genetic algorithm is used to re-optimize the pressure distribution and flow control parameters of the surrounding pipeline network. The specific implementation of the genetic algorithm includes the use of real number coding to represent the pressure and flow of each pipe section, the fitness function considers the total energy consumption and pressure balance, the selection operation uses tournament selection, the crossover operation uses arithmetic crossover, and the mutation operation uses Gaussian mutation. Set the population size and the number of iterations to obtain the optimization results and realize the dynamic adjustment of the pipeline operation strategy. In the dynamic scheduling process of the park steam pipeline network, the digital twin model updates the data of 100 nodes in real time, including the steam demand of 50 enterprises and the pressure and flow information of 50 key points of the pipeline network. The Dijkstra algorithm recalculates the optimal path at intervals of 5 minutes, considering factors including the pipeline length, which is 50 kilometers in total, the diameter, which is 200-500 mm and the pressure loss, which is 0.01-0.05 MPa per kilometer. When the access signal of the new enterprise A is detected, its geographic coordinates, X:120.5, Y:30.8 and steam parameters, including pressure 0.8MPa and flow 5 tons / hour, are automatically obtained. The Steiner tree algorithm calculates the optimal connection plan, and the new pipeline length is 1.2 kilometers and the diameter is 300 mm. The calculation of the pressure balance of the pipeline network shows that the pressure at the connection point has dropped from 0.85MPa to 0.82MPa, which is still within the safe range. The ARIMA model (parameters p=1, d=1, q=1) predicts that the average daily steam consumption of enterprise A in the next seven days will be 4.8, 5.2, 5.1, 4.9, 5.0, 5.3, and 5.1 tons / hour. The weighted average method integrates this result with the existing load forecast, with a weight ratio of 1:9. Based on this, the adaptive control algorithm adjusts the opening of the three regulating valves of the relevant pipe sections to 75%, 80%, and 85%, respectively, and the speed of the two variable frequency pumps is adjusted to 45Hz and 48Hz. When enterprise B exits the pipeline network, the genetic algorithm optimizes the parameters of the surrounding pipeline network, the population size is set to 100, and iterates 500 times.The final optimization plan closed the upstream valve of the original enterprise B, reset the pressure of the three adjacent nodes to 0.79MPa, 0.81MPa and 0.78MPa, and adjusted the flow rate to decrease by 2%, increase by 3% and remain unchanged respectively.

[0052] The above is only a specific implementation method of this specification. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working process of the system, module and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. It should be understood that the protection scope of this specification is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in this specification, and these modifications or replacements should be covered within the protection scope of this specification.

Claims

1. A steam delivery network with multi-stage pressure regulation and an optimization method thereof, characterized in that: The method comprises: Obtain the steam monitoring value of the target node of the park steam pipe network; Determine the steam grade according to the preset steam grade classification standard and obtain the steam grade distribution map in the pipe network; For the steam demand order submitted by the enterprise, extract the key steam parameters, match the key steam parameters with the steam grade distribution map of the pipeline network, search for feasible transportation paths from the steam source to the target enterprise, and select the target path as the transportation plan for the order; Predict the steam parameters at the end of the pipeline according to the transportation plan, compare the prediction results with the steam demand of the target enterprise, and optimize the steam supply parameters until the target demand is met; Continuously monitor the real-time parameters of the target nodes in the pipeline network, compare the real-time parameters with the predicted parameters, and when the deviation value exceeds the preset threshold, it is judged that the transportation status is abnormal and corresponding adjustment measures are taken; Obtain steam pressure change data based on regulation measures, build a pressure regulation optimization model, and adjust the preset multi-stage pressure regulation method; When there are changes in the enterprises within the park, the topological structure and transportation routes of the local pipeline network will be replanned and incorporated into the transportation scheduling system for unified management.

2. The multi-stage pressure-regulated steam delivery network and optimization method thereof according to claim 1, characterized in that: The step of obtaining the steam monitoring value of the target node of the park steam pipe network includes: Obtain pressure data, temperature data, and flow data of the target node; Calculate the average value and standard deviation of each sensor in the most recent time period based on the data, and if the data point deviates from the average value by more than a preset multiple of the standard deviation, it is determined to be abnormal data; The interpolation method is used to correct abnormal data and obtain the corrected data.

3. The multi-stage pressure-regulated steam delivery network and optimization method thereof according to claim 1, characterized in that: The steam grade is determined according to the preset steam grade classification standard to obtain the steam grade distribution map in the pipe network, including: Determine the steam level of each target node according to the corrected data and the preset steam level classification standard; Using an inverse distance weighted method to perform spatial interpolation on the target node to obtain an interpolation result; Gridding the park using the interpolation result to determine the steam level value of each grid unit; A park steam level distribution map is generated according to the steam level value.

4. The multi-stage pressure-regulated steam delivery network and optimization method thereof according to claim 1, characterized in that: Predict the steam parameters at the end of the pipeline according to the transportation plan, and compare the prediction results with the steam demand of the target enterprise to optimize the steam supply parameters until the target demand is met, including: Obtaining the pipeline length, diameter, insulation material parameters and pipeline roughness according to the steam delivery path; Calculate pressure loss and heat loss; Obtaining predicted steam parameters at the end of the pipeline according to the pressure loss and heat loss calculation results; The predicted steam parameters are compared with the steam demand of the target enterprise. If the pressure error or the temperature error is greater than a threshold, the gradient descent method is started to optimize the steam supply parameters.

5. The multi-stage pressure-regulated steam delivery network and optimization method thereof according to claim 1, characterized in that: The real-time parameters of the target nodes in the pipeline network are continuously monitored, and the real-time parameters are compared with the predicted parameters. When the deviation value exceeds the preset threshold, it is judged that the transportation state is abnormal, and corresponding adjustment measures are taken, including: Collect target node pressure data, temperature data, and flow data; Losslessly compressing the data to obtain compressed data; Comparing the compressed data with the pre-stored prediction parameters, calculating the average deviation value using a sliding window method, and obtaining the pressure deviation, temperature deviation and flow deviation; If at least one of the pressure deviation, temperature deviation and flow deviation exceeds a preset threshold, an early warning mechanism is triggered.

6. The multi-stage pressure-regulated steam delivery network and optimization method thereof according to claim 1, characterized in that: The steam pressure change data is obtained according to the regulation measures, a pressure regulation optimization model is constructed, and the preset multi-stage pressure regulation method is adjusted, including: Obtain real-time pressure data of each node in the steam network; Performing data cleaning and normalization processing according to the real-time pressure data to obtain a standardized pressure change data set; Using dynamic time warping distance as a similarity metric, hierarchical clustering is performed on the standardized pressure change data set, and the pressure change data set is divided into multiple levels to obtain a multi-level pressure change data set; Based on the multi-level pressure change data set, a long short-term memory network is constructed as a pressure regulation optimization model.

7. The multi-stage pressure-regulated steam delivery network and optimization method thereof according to claim 1, characterized in that: When there are changes in enterprises within the park, the topological structure and transportation routes of the local pipeline network are replanned and incorporated into the transportation scheduling system for unified management, including: Obtain geographic information data of the park’s steam pipeline network; Establishing a digital twin model based on the geographic information data; The Dijkstra algorithm is used to calculate the optimal delivery path and pressure distribution plan; Receive access signals from new enterprises; According to the newly added enterprise access signal, the local pipe network structure is replanned using the Steiner tree algorithm; Obtain steam demand data for newly added enterprises and use the ARIMA model to predict future steam consumption.

Images