A method and system for optimizing roughness of a heating pipe network based on hydraulic simulation

Abstract

The application discloses a kind of based on hydraulic simulation roughness optimization method and system of heating pipe network, it is related to heating system pipe network roughness optimization technique field, heating pipe network is divided into W region, constructs hydraulic simulation model, obtains heating data, heating data includes initial cold period data, severe cold period data and last cold period data, heating data is arranged as the input parameter set of hydraulic simulation model, the optimized roughness of output heating pipe network, according to the roughness of the population of heating pipe network and preset population number, constructs, the fitness value of different individual in population is calculated, obtains the population after optimization, respectively obtains the optimal roughness of initial cold period, severe cold period and last cold period of preset region, initial cold period, severe cold period and last cold period are calculated according to preset proportion, obtain the optimal roughness of preset region, obtain the optimal roughness of W region, adjust stem flow distribution and actual working condition consistent, optimize and verify the roughness of overall heating pipe network.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of heating system pipe network roughness optimization, and particularly relates to a heating pipe network roughness optimization method and system based on hydraulic simulation. BACKGROUND

[0002] The heating pipe network is one of the basic components of the heating system, and undertakes the important task of transporting and distributing heat to each end user. The heating pipe network includes complex structures such as branch pipe networks and ring pipe networks, and most of the heating pipe networks have been constructed for a long time. Due to various factors such as water quality and water flow erosion, the designed roughness and the actual pipe roughness of the pipe network deviate to varying degrees, and the roughness deviation of different pipes is different. If the designed roughness is used for hydraulic simulation calculation to guide the stable and reliable operation of the heating system, it will bring serious challenges. Therefore, the actual pipe roughness is calculated according to the actual operation data to avoid the phenomenon that the hydraulic simulation model calculation data deviates far from the actual value, and to ensure the accuracy of the hydraulic simulation and correctly guide the stable and reliable operation of the heating system. The actual pipe roughness of different heating pipe networks changes in a complex manner due to factors such as construction time, corrosion, scaling, pipe wear degree, and actual operation control. It is a very challenging task to accurately calculate the roughness of the heating pipe network. At present, the main methods are manual trial and error and optimization of the hydraulic simulation model using theoretical formulas. However, due to the complex and variable actual situation of different heating pipe networks and the large number of pipes, there are problems of long time consumption and low precision. It is necessary to use an optimization algorithm to optimize and calculate the pipe roughness of the heating pipe network based on the historical data of the heating pipe network.

[0003] At present, the Chinese patent with application number 202211624928.8 discloses a heating pipe network roughness identification method and device. A simulation model is established based on the basic information of the heating pipe network, the pipes in the simulation model are grouped according to the size of the pipe diameter or flow rate, the particle swarm algorithm is used, the absolute value of the difference between the one-side supply and return water pressure difference calculated by the simulation model and the actual pressure difference is taken as the objective function, and the roughness values of the pipes in each group are optimized in turn according to the grouping order. The above-mentioned technology has the following disadvantages: first, the spatial decoupling is not complete, and the calculation convergence is difficult. The optimization space under this method is still large, and the interference between groups is obvious, which leads to the optimization algorithm easily falling into a local optimal solution, or the calculation convergence time is too long, and it is difficult to meet the demand of online real-time simulation. Second, the difference in working conditions of the whole operation cycle is not considered, which leads to poor robustness. Third, the limitation of the objective function only satisfies the total pressure difference matching, and does not represent that the distribution of the local pressure of each node in the pipe network is consistent with the actual situation, which easily leads to the phenomenon that the total pressure difference is matched, but there is a serious water imbalance in the local area. SUMMARY

[0004] The technical problem solved by the present application is that the prior art cannot solve the problem that different heating pipe networks are difficult to solve due to artificial trial and error and the use of theoretical formulas, resulting in a long time-consuming and large deviation from the actual roughness, which leads to the inability of the hydraulic simulation model to correctly specify the operation of the heating pipe network.

[0005] To solve the above technical problems, the present application provides the following technical solutions: a heating pipe network roughness optimization method based on hydraulic simulation, comprising the following steps:

[0006] Step S1, according to the topology structure of heating, the heating pipe network is divided into W regions, and based on the W regions, a hydraulic simulation model is constructed;

[0007] Step S2, obtain the heating data, the heating data includes initial cold period data, severe cold period data and end cold period data, arrange the heating data into the input parameter set of the hydraulic simulation model, and output the optimized roughness of the heating pipe network;

[0008] Step S3, according to the optimized roughness and the preset population number, a population of roughness of the heating pipe network is constructed, the fitness value of different individuals in the population is calculated through the pre-constructed roughness fitness function, and the optimized population is obtained;

[0009] Step S4, obtain the optimal roughness of the initial cold period, the severe cold period and the end cold period of the preset area, and calculate according to the preset proportion to obtain the optimal roughness of the preset area;

[0010] Step S5, after obtaining the optimal roughness of the W regions, adjust the main flow distribution to be consistent with the actual working condition, and verify the overall heating pipe network roughness.

[0011] Preferably, the step S1 comprises the following sub-steps:

[0012] Step S101, according to the topology structure of heating, the heating pipe network is divided into W regions, and a virtual heat source is preset at the junction of each region;

[0013] Step S102, refer to the measured operating parameters of the heat station near the junction, and check the elevation data, then preset the water supply temperature, supply and return water pressure and lift parameters of the virtual heat source;

[0014] Step S103, according to the pipe laying condition, heat source position, heat station construction position and elevation data of the W regions, a hydraulic simulation model is constructed.

[0015] Preferably, the step S2 comprises the following sub-steps:

[0016] Step S201, collecting operation data of the heating pipe network in the whole heating season, and obtaining heating data by cleaning abnormal data from the operation data, the heating data including initial cold period data, severe cold period data and end cold period data;

[0017] Step S202, arranging the initial cold period data into an initial cold period input parameter set, the initial cold period input parameter set including initial cold period heat source parameters, initial cold period heat exchange station parameters and initial cold period service pressure parameter set;

[0018] Step S203, arranging the severe cold period data into a severe cold period input parameter set, the severe cold period input parameter set including severe cold period heat source parameters, severe cold period heat exchange station parameters and severe cold period service pressure parameter set;

[0019] Step S204, arranging the end cold period data into an end cold period input parameter set, the end cold period input parameter set including end cold period heat source parameters, end cold period heat exchange station parameters and end cold period service pressure parameter set.

[0020] Preferably, the step S2 includes the following sub-steps:

[0021] Step S205, taking the initial cold period input parameter set, the severe cold period input parameter set and the end cold period input parameter set as input parameter sets of a hydraulic simulation model, and outputting an optimized roughness of the heating pipe network in the preset area.

[0022] Preferably, the step S3 includes the following sub-steps:

[0023] Step S301, constructing a roughness fitness function, and presetting a termination condition and physical constraint upper and lower limits of pipe roughness, the physical constraint upper and lower limits being a preset roughness upper limit a and a preset roughness lower limit b;

[0024] The calculation expression of the roughness fitness function is:

[0025]

[0026] Wherein, is a roughness fitness value, is a service pressure error proportion of an i-th pipe in an f-th area, if is 1, otherwise is 0, and k% is a preset percentage;

[0027] The calculation formula of the service pressure error proportion is:

[0028]

[0029] Wherein, is a service pressure error of an i-th pipe in an f-th area, a pressure of the initial cold period of the i-th pipe in the f-th region, a pressure of the initial cold period of the i-th pipe in the f-th region calculated;

[0030] In step S302, when the proportion of the preset total number of pipes is less than the preset termination condition, the optimization is stopped, and the preset total number of pipes is the number of pipes whose pressure error proportion is greater than the preset percentage.

[0031] Preferably, the step S3 further comprises the following sub-steps:

[0032] In step S303, an initial population of roughness of the heat supply pipe network is constructed according to a preset population quantity and the optimized roughness, and the initial population is:

[0033]

[0034] In step S304, the fitness values of different individuals in the population are obtained through the roughness fitness function.

[0035] The logic for obtaining the fitness value is:

[0036] The roughness of the j-th individual in the f-th region is obtained.

[0037] The initial cold period input parameter set and the roughness of the j-th individual in the f-th region are input into the hydraulic simulation model as input parameters, and the calculated pressure is output.

[0038] The pressure error of the i-th pipe of the j-th individual in the f-th region is calculated, and the calculation expression is:

[0039]

[0040] wherein, the pressure error of the i-th pipe of the j-th individual in the f-th region, the calculated pressure of the i-th pipe of the j-th individual in the f-th region, the pressure of the initial cold period of the i-th pipe of the j-th individual in the f-th region;

[0041] The pressure error of all pipes of the j-th individual in the f-th region is calculated, and the fitness value of the j-th individual in the f-th region is obtained.

[0042] From the fitness value of the j-th individual in the f-th region, the maximum pressure error of the j-th individual in the f-th region is obtained.

[0043] The optimal individual is selected, and the selection method of the optimal individual comprises:

[0044] In the fitness value of the population, the minimum pressure error of the population is obtained.

[0045] If the number of individuals with the same minimum utilization pressure error in the population is greater than or equal to 2, the maximum utilization pressure error of the individuals corresponding to the minimum utilization pressure error of the population is compared, and the individual corresponding to the smaller maximum utilization pressure error after comparison is taken as the optimal individual.

[0046] Preferably, the step S3 further comprises the following sub-steps:

[0047] Step S305: It is judged whether the minimum utilization pressure error of the optimal individual satisfies a preset termination percentage. If the preset termination percentage is satisfied, the optimization is ended. If the preset termination percentage is not satisfied, an optimized population is obtained through elite strategy selection, binary simulated crossover and polynomial mutation.

[0048] The logic for obtaining the optimized population is as follows:

[0049] The elite strategy selection is used to select individuals to construct a new population for the initial population, and the selection method for constructing the new population by using the elite strategy selection is the same as the selection method for the optimal individual.

[0050] The binary simulated crossover is used for the new population.

[0051] The polynomial mutation is used for the population after the simulated crossover to obtain the optimized population.

[0052] Preferably, the step S4 comprises the following sub-steps:

[0053] Step S401: The initial cold period input parameter set is taken as input, and step S3 is repeatedly executed. When the preset total number of pipelines accounts for less than a preset termination condition, the initial cold period optimal roughness of the preset region is obtained.

[0054] Step S402: The severe cold period input parameter set and the end cold period input parameter set are respectively taken as input, and step S3 is repeatedly executed. When the preset total number of pipelines accounts for less than a preset termination condition, the severe cold period optimal roughness and the end cold period optimal roughness of the preset region are respectively obtained.

[0055] Step S403: The initial cold period optimal roughness, the severe cold period optimal roughness and the end cold period optimal roughness are calculated according to a preset proportion to obtain the optimal roughness of the preset region.

[0056] Preferably, the step S5 comprises the following sub-steps:

[0057] Step S501: Each of the W regions is taken as the preset region, and steps S3 and S4 are repeatedly executed until the optimal roughness of the W regions is calculated.

[0058] Step S502, input the optimal roughness of the W regions into the set of hydraulic simulation models respectively, splice the hydraulic simulation models of the W regions, and cancel the virtual heat source at the junction of each region. In the spliced hydraulic simulation model, adjust the resistance value of each regional junction, adjust the flow distribution relationship, and until the flow distribution ratio of the main line of each region in the hydraulic simulation model is consistent with the actual working condition data.

[0059] Step S503, if the preset pipe total number ratio of the overall heating pipe network is less than the preset termination condition, output the optimization result, if the preset pipe total number ratio is greater than or equal to the preset termination condition, return to step S2 and step S3 to re-optimize.

[0060] A heating pipe network roughness optimization system based on hydraulic simulation, comprising a construction module, a processing module, an optimization module, an optimization module and a verification module;

[0061] The construction module is used to divide the heating pipe network into W regions according to the topological structure of heating, and to construct a hydraulic simulation model based on the W regions;

[0062] The processing module is used to obtain heating data, the heating data including initial cold period data, severe cold period data and end cold period data, to arrange the heating data into an input parameter set of the hydraulic simulation model, and to output the optimized roughness of the heating pipe network, the input parameter set including initial cold period input parameter set, severe cold period input parameter set and end cold period input parameter set;

[0063] The optimization module is used to construct a roughness fitness function, to construct a population of roughness of the heating pipe network according to the optimized roughness and a preset population number, to calculate the fitness value of different individuals in the population through the roughness fitness function, and to obtain the optimized population through elite strategy selection, binary simulation crossover and polynomial mutation;

[0064] The optimization module is used to construct a roughness fitness function, to construct a population of roughness of the heating pipe network according to the optimized roughness and a preset population number, to calculate the fitness value of different individuals in the population through the roughness fitness function, and to obtain the optimized population through elite strategy selection, binary simulation crossover and polynomial mutation;

[0065] The verification module is used to obtain the optimal roughness of the W regions, adjust the W region main flow distribution to be consistent with the actual working condition, and optimize and verify the overall heating pipe network roughness.

[0066] The present application has the advantages that the advantages of fusion genetic algorithm and actual operation data are combined, the operation data in three working conditions of initial cold period, severe cold period and final cold period are obtained through historical data operation, the optimal roughness in each working condition is optimized through the improved genetic algorithm, and finally the optimal roughness is calculated by setting the proportion, compared with artificial trial and error and theoretical formula, the calculation efficiency and calculation accuracy are improved, and more accurate data support is provided for realizing stable operation of the heat supply pipe network. BRIEF DESCRIPTION OF DRAWINGS

[0067] Figure 1 A step flow chart of a heat supply pipe network roughness optimization method based on hydraulic simulation is provided for an embodiment of the present application.

[0068] Figure 2 A basic flow diagram of a heat supply pipe network roughness optimization system based on hydraulic simulation is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0069] In order to make the above-mentioned objects, features and advantages of the present application more obvious and easy to understand, the specific embodiments of the present application will be described in detail below with reference to the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments.

[0070] Embodiment 1, with reference to Figure 1 , provides a heat supply pipe network roughness optimization method based on hydraulic simulation, including the following steps:

[0071] Step S1, according to the topological structure of heat supply, the heat supply pipe network is divided into W regions, and a hydraulic simulation model is constructed based on the W regions.

[0072] Step S2, obtain heat supply data, the heat supply data includes initial cold period data, severe cold period data and final cold period data, arrange the heat supply data into an input parameter set of the hydraulic simulation model, output the optimized roughness of the heat supply pipe network, and the input parameter set includes an initial cold period input parameter set, a severe cold period input parameter set and a final cold period input parameter set.

[0073] Step S3, construct a roughness fitness function, construct a roughness population of the heat supply pipe network according to the optimized roughness and a preset population number, calculate the fitness value of different individuals in the population through the roughness fitness function, and obtain the optimized population through elite strategy selection, binary simulation crossover and polynomial mutation.

[0074] Step S4, respectively obtain the optimal roughness of the initial cold period, the severe cold period and the final cold period of the preset region, calculate the initial cold period, the severe cold period and the final cold period according to a preset proportion, and obtain the optimal roughness of the preset region.

[0075] Step S5, after obtaining the optimal roughness of the W regions, adjusting the W region main stem flow distribution to be consistent with the actual working condition, and optimizing and verifying the overall heating pipe network roughness.

[0076] The present application solves the problems of calculation dimension explosion and poor robustness of single working condition fitting in traditional heating pipe network roughness identification through the overall architecture of partition modeling, time period optimization and weighted fusion, greatly reduces the search space of single optimization by dividing the complex large network into small regions according to the topology, improves the algorithm convergence speed, and overcomes the problem of roughness parameter overfitting under single working condition by introducing the data of three typical periods of initial cold period, severe cold period and final cold period for multi-dimensional verification and weighting, so that the finally optimized roughness parameter can more truly reflect the physical characteristics of the pipe network in the whole heating season, and the full working condition prediction accuracy of the hydraulic simulation model is significantly improved.

[0077] Step S1 includes the following sub-steps:

[0078] Step S101, according to the topological structure of heating, the heating pipe network is divided into W regions, and a virtual heat source is preset at the gateway of each region.

[0079] Step S102, referring to the measured operating parameters of the heat station near the gateway, and checking the elevation data, the water supply temperature, the supply and return water pressure and the lift parameters of the virtual heat source are preset.

[0080] Step S103, W regions are constructed into a set of hydraulic simulation models according to the pipeline laying condition, heat source position, heat station construction position and elevation data. Wherein w is the total number of regions, g f is the hydraulic simulation model of the fth region.

[0081] Step S1, the region division method based on topological structure instead of simple physical properties such as pipe diameter is innovatively put forward, and the virtual heat source and near-end measured calibration technology is ingeniously introduced, realizing the real decoupling of each region in hydraulics. This not only makes parallel computing possible, greatly shortens the modeling and calculation time of super large scale pipe network, but also ensures the high fidelity of the boundary conditions of the segmented sub-models through the calibration of the near-end measured data, avoiding the boundary error accumulation caused by model splitting.

[0082] Step S2 includes the following sub-steps:

[0083] Step S201, the operation data of the heating pipe network in the whole heating season is collected, and after the abnormal data cleaning, the heating data including the initial cold period data, the severe cold period data and the final cold period data are obtained.

[0084] It should be noted that the initial cold period, the severe cold period and the final cold period are divided according to the outdoor daily average temperature and the heating operation time stage of the heating season. The following is one of the exemplary division methods:

[0085] The initial cold period is the stage from the beginning of the heating season to before the outdoor daily average temperature drops to-5℃. The initial cold period corresponds to 15 to 30 days at the beginning of the heating season.

[0086] The severe cold period is the stage when the outdoor daily average temperature is continuously lower than-5℃. The severe cold period corresponds to the coldest 2 to 3 months in the middle of the heating season, at which time the heating pipe network is in a high load operation state.

[0087] The final cold period is the stage from when the outdoor daily average temperature rises above-5℃ to the end of the heating season. The final cold period corresponds to 15 to 30 days at the end of the heating season.

[0088] Those skilled in the art can understand that the length of the heating season and the change of air temperature are different in different regions. The above specific temperature and days are only examples and do not constitute a limitation on the present application. The core purpose of the division is to obtain typical working condition data of the heating pipe network in the initial cold period, the severe cold period and the final cold period.

[0089] Step S202, the initial cold period data is sorted into an initial cold period input parameter set , wherein, is the initial cold period heat source parameter of the fth region, is the initial cold period heat exchange station parameter of the fth region and the initial cold period pressure parameter set of the fth region .

[0090] The initial cold period heat source parameter is:

[0091]

[0092] , wherein, is the initial cold period heat source water supply pressure of the fth region, is the initial cold period heat source return water pressure of the fth region, is the initial cold period heat source water supply temperature of the fth region.

[0093] The initial cold period heat exchange station parameter is:

[0094]

[0095] , wherein y is the total number of heat exchange stations of the rth region, is the initial cold period boundary parameter of the ith heat exchange station of the fth region, including the initial cold period load and the initial cold period return water temperature of the heat exchange station.

[0096] The initial cold period pressure is:

[0097]

[0098] wherein n is the total number of the service pressure of the fth region, is the initial cold period service pressure of the ith pipe of the fth region.

[0099] Step S203, the cold period data is sorted into a cold period input parameter set wherein, is the cold period heat source parameter of the fth region, is the cold period heat exchange station parameter of the fth region and the cold period service pressure parameter set of the fth region .

[0100] The cold period heat source parameter is:

[0101]

[0102] wherein, is the cold period heat source water supply pressure of the fth region, is the cold period heat source return water pressure of the fth region, is the cold period heat source water supply temperature of the fth region.

[0103] The cold period heat exchange station parameter is:

[0104]

[0105] wherein y is the total number of the heat exchange station of the fth region, is the cold period boundary parameter of the ith heat exchange station of the fth region, including the heat exchange station cold period load and the cold period return water temperature.

[0106] The cold period service pressure is:

[0107]

[0108] wherein n is the total number of the service pressure of the fth region, is the cold period service pressure of the ith pipe of the fth region.

[0109] Step S204, the end cold period data is sorted into an end cold period input parameter set wherein, is the end cold period heat source parameter of the fth region, is the end cold period heat exchange station parameter of the fth region and the end cold period service pressure parameter set of the fth region .

[0110] The end cold period heat source parameter is:

[0111]

[0112] wherein, is the supply water pressure of the terminal cold period heat source of the fth region, is the return water pressure of the terminal cold period heat source of the fth region, is the supply water temperature of the terminal cold period heat source of the fth region.

[0113] The terminal cold period heat exchange station parameters are:

[0114]

[0115] wherein, y is the total number of heat exchange stations of the fth region, is the terminal cold period boundary parameter of the ith heat exchange station of the fth region, including the terminal cold period load and the terminal cold period return water temperature.

[0116] The terminal cold period service pressure is:

[0117]

[0118] wherein, n is the total number of service pressures of the fth region, is the terminal cold period service pressure of the ith pipe of the fth region.

[0119] Step S2 includes the following sub-steps:

[0120] Step S205, taking the initial cold period input parameter set, the severe cold period input parameter set and the terminal cold period input parameter set as the input parameter set of the hydraulic simulation model, outputting the optimized roughness of the preset region heating pipe network wherein, n is the total number of pipe roughnesses of the fth region, is the optimized roughness of the ith pipe of the fth region.

[0121] Step S2, the input parameter set covering the initial cold period, the severe cold period and the terminal cold period is constructed, the input parameter set including the heat source, the heat exchange station and the service pressure, the innovation point is that the nonlinear hydraulic characteristics of the heating pipe network generated with the change of the meteorological condition and the operation strategy are captured from the time dimension, the three groups of boundary condition data with great difference are cleaned and arranged, a training set covering the whole working condition is provided for the subsequent algorithm, and it is ensured that the finally recognized roughness is not a special solution for a moment, but a general solution adapting to the whole heating season. It is clear that the pipe roughness is taken as the optimization variable, and the multi-time phase parameter set is taken as the boundary condition for driving the hydraulic simulation model. The physical model and the data driven method are closely combined to provide a clear input and output interface for the genetic algorithm, so that the optimization process can be targeted and can directly inverse the roughness affecting the hydraulic disorder.

[0122] Step S3 includes the following sub-steps:

[0123] Step S301, construct the roughness fitness function, and preset the termination condition as t and the physical constraint upper and lower limits of pipe roughness, the physical constraint upper limit of roughness is a, the physical constraint lower limit of roughness is b, and the default termination condition is 5%.

[0124] The calculation expression of the roughness fitness function is:

[0125]

[0126] Wherein, is the roughness fitness value, that is, the error of a group of roughnesses in the fth region, is the error proportion of the service pressure of the ith pipe in the fth region, if is 1, otherwise 0, and k% is a preset percentage, is the error of a group of roughnesses in the fth region.

[0127] The calculation formula of the error proportion of the service pressure of the ith pipe in the fth region is:

[0128]

[0129] Wherein, is the error of the service pressure of the ith pipe in the fth region, is the service pressure of the ith pipe in the fth region in the initial cold period, is the calculated service pressure of the ith pipe in the fth region in the initial cold period.

[0130] Step S302, when the preset pipe total number proportion is less than the preset termination condition t, stop optimization, the preset pipe total number is the number of pipes with a service pressure error proportion greater than the preset percentage k%. When there are 24 pipe roughnesses in the hydraulic simulation model, the error proportion of the 24 pipe roughnesses is calculated after each iteration, such as [30%, 25%, …, 2%], and the error is (2 / 24) x 100 = 8.33%.

[0131] Step S3 further includes the following substeps:

[0132] Step S303, according to the preset population number m and the optimized roughness R f , the initial population of the roughness of the heating pipe network is constructed , the initial population is:

[0133]

[0134] Step S304, the fitness values of different individuals in the population are obtained through the roughness fitness function , wherein, is the fitness value of the jth individual in the fth region, and m is the population size.

[0135] Obtaining the fitness value The logic is as follows:

[0136] Obtaining the roughness of one individual wherein, is the roughness of the ith pipe of the jth individual in the fth region.

[0137] Inputting the set of initial cold period input parameters and the roughness of the ith pipe of the jth individual in the fth region as input parameters, inputting the hydraulic simulation model g f , and outputting the calculated utilization pressure wherein, is the calculated utilization pressure of the ith pipe of the jth individual in the fth region.

[0138] Calculating the utilization pressure error of the ith pipe of the jth individual in the fth region , and the calculation expression is as follows:

[0139]

[0140] wherein, is the utilization pressure error of the ith pipe of the jth individual in the fth region, is the ith calculated utilization pressure of the jth individual in the fth region, is the ith initial cold period utilization pressure of the jth individual in the fth region.

[0141] Calculating the utilization pressure error of all pipes of the jth individual, and obtaining the fitness value of the jth individual in the fth region .

[0142] From the fitness value of the jth individual in the fth region, obtaining the maximum utilization pressure error of the jth individual .

[0143] Selecting an optimal individual u, and the selection method of the optimal individual u includes:

[0144] In the fitness values of the population, obtaining the minimum utilization pressure error ;

[0145] If there are two or more individuals whose minimum utilization pressure errors are consistent, comparing the maximum utilization pressure errors of the corresponding individuals , and selecting the individual with the smaller maximum utilization pressure error as the optimal individual.

[0146] Step S304, through the specific logic of population construction and fitness evaluation in genetic algorithm, especially the double selection strategy of comparing the number of violations first and then the maximum error, i.e. the theory of wooden barrel, which solves the problem that the traditional algorithm easily hides the local serious disorder by only looking at the average error. By further screening the individual with the smallest maximum error under the same number of violations, it is ensured that the final scheme not only meets the macroscopic indicators, but also eliminates the extremely severe hydraulic imbalance points at the micro level, greatly improving the hydraulic balance stability of the heating system.

[0147] Step S3 also includes the following sub-steps:

[0148] Step S305, judge whether the minimum consumption pressure error of the optimal individual meets the preset termination percentage If the preset termination percentage is met, the optimization is ended, if the preset termination percentage is not met, the optimized population is obtained through elite strategy selection, binary simulated crossover and polynomial mutation.

[0149] The logic for obtaining the optimized population is.

[0150] For the existing population Use the elite strategy to select individuals to construct a new population Use the elite strategy to select individuals to construct a new population The selection method is the same as that of the optimal individual.

[0151] For the new population Use binary simulated crossover;

[0152] For the population after simulated crossover Use polynomial mutation to obtain the optimized population .

[0153] Step S305, the elite strategy selection, binary simulated crossover and polynomial mutation and other advanced genetic operation operators are introduced, compared with the standard genetic algorithm, the introduction of these operators significantly enhances the ability of the algorithm to jump out of the local optimal trap, at the same time, the elite strategy ensures that the excellent genes, i.e. the roughness combination that meets the hydraulic law, are not destroyed. This makes the algorithm quickly and stably converge to the global optimal solution when dealing with high-dimensional, nonlinear pipe network hydraulic equation set inversion problems.

[0154] Step S3, the fitness function is innovatively constructed with the number of pipelines with excessive pressure error ratio as the core index, instead of the traditional total pressure difference or total flow error. This design directly anchors the actual heating quality of the end users of the heating pipe network, i.e., the pressure is the key to determine the user's cold and warm, so that the optimization direction is more in line with the actual engineering demand. At the same time, the upper and lower limits of the physical constraints and the specific percentage termination condition are set, effectively preventing the algorithm from producing mathematical solutions that violate physical common sense, and ensuring the engineering usability of the optimization results.

[0155] Step S4 includes the following sub-steps:

[0156] Step S401, repeat step S3 with the initial cold period input parameter set as input until the minimum pressure error of the optimal individual meets the termination condition t, and obtain the optimal roughness of the preset area in the initial cold period.

[0157] Step S402, respectively input the severe cold period input parameter set and the end cold period input parameter set, repeat step S3 until the minimum pressure error of the optimal individual meets the termination condition t, and respectively obtain the optimal roughness of the preset area in the severe cold period and the end cold period.

[0158] It should be noted that step S3 is described as an example of the initial cold period, when this step is executed for the severe cold period or the end cold period, the initial cold period input parameter set in step S3 should be understood as being replaced by the severe cold period input parameter set or the end cold period input parameter set, and the initial cold period pressure in the formula of step S3 should also be replaced by the severe cold period pressure or the end cold period pressure. Through the above parameter replacement and repeated execution of step S3, when the total number of preset pipelines is less than the preset termination condition, the optimal roughness of the preset area in the severe cold period and the end cold period is obtained respectively.

[0159] Step S403, calculate the optimal roughness of the preset area according to the preset proportion of the optimal roughness in the initial cold period, the optimal roughness in the severe cold period and the optimal roughness in the end cold period, and the calculation expression is:

[0160]

[0161] Wherein, K f is the optimal roughness of the preset area, K f,1 is the optimal roughness of the initial cold period, K f,2 is the optimal roughness of the severe cold period, K f,3 is the optimal roughness of the end cold period, x1 is the preset weight proportion of the initial cold period, x2 is the preset weight proportion of the severe cold period, and x3 is the preset weight proportion of the end cold period.

[0162] Step S4, which improves the robustness of the model, discards the traditional idea of calculating data, and instead allows the algorithm to train in three different working conditions. Finally, the final result is obtained by weighted fusion. This mechanism is similar to the idea of ensemble learning in machine learning. By fusing the optimal solutions in different scenarios, random errors and system biases in a single scenario are eliminated, making the final determined roughness parameter remain highly accurate when facing unknown future operating conditions.

[0163] Step S5 includes the following sub-steps:

[0164] Step S501, traverse W regions, and respectively take each region as a preset region. Repeat steps S3 and S4 until the optimal roughness of the W regions is calculated.

[0165] Step S502, input the optimal roughness of the W regions into the set of hydraulic simulation models, respectively. The hydraulic simulation models of the W regions are spliced, and the virtual heat source at the junction of each region is removed. In the spliced hydraulic simulation model, adjust the resistance value of each regional junction, adjust the flow distribution relationship, until the flow distribution ratio of each regional main line in the hydraulic simulation model is consistent with the actual working condition data.

[0166] Step S503, if the preset pipe total number ratio of the overall heating pipe network is less than the preset termination condition t, output the optimization result, if the preset pipe total number ratio is greater than or equal to the preset termination condition t, return to step S2 and step S3 to optimize again.

[0167] Step S5, a closed-loop verification process from local optimal regression to global optimal is designed. Through a series of operations such as model splicing, removing virtual heat sources and flow distribution callback, the physical connection state of the complete pipe network is truly restored, eliminating the boundary mismatching problem that may be left by regional segmentation. Through the final check from a global perspective, it is ensured that the local optimization results can be seamlessly integrated into the overall system, achieving perfect unity from micro parameter correction to macro system balance.

[0168] Embodiment 2, referring to Figure 2 provides a heating pipe network roughness optimization system based on hydraulic simulation, which includes a construction module, a processing module, an optimization module, an optimization module, and a verification module.

[0169] The construction module is used to divide the heating pipe network into W regions according to the topological structure of the heating, and to construct a hydraulic simulation model based on the W regions.

[0170] The construction module integrates topology analysis and regional disassembly functions, can automatically identify weakly coupled nodes of the pipe network for partitioning, and automatically configure virtual heat source parameters, realizing automation and standardization of the modeling process, so that even in the face of super-large complex pipe networks with tens of thousands of nodes, a high-precision sub-model set for parallel computing can be quickly generated, greatly reducing the difficulty and time cost of manual modeling.

[0171] The processing module is used to obtain heating data, the heating data including initial cold period data, severe cold period data and end cold period data, arrange the heating data into an input parameter set of the hydraulic simulation model, and output an optimized roughness of the heating pipe network, the input parameter set including an initial cold period input parameter set, a severe cold period input parameter set and an end cold period input parameter set.

[0172] The processing module has powerful multi-source data fusion and cleaning capability, can automatically filter out effective samples of the initial cold period, the severe cold period and the end cold period from massive historical data, and convert the effective samples into a standard input parameter set recognizable by the simulation model, ensuring the purity and representativeness of the input data and eliminating the interference of noise data, laying a solid data foundation for subsequent high-precision optimization.

[0173] The optimization module is used to construct a roughness fitness function, construct a population of roughness of the heating pipe network according to the optimized roughness and a preset population quantity, calculate the fitness values of different individuals in the population through the roughness fitness function, and obtain an optimized population through elite strategy selection, binary simulated crossover and polynomial mutation.

[0174] The optimization module has an improved adaptive genetic algorithm engine built-in, integrates a unique double fitness evaluation mechanism, i.e., the number of violations and the maximum error, converts the complex fluid mechanics inverse problem into a solvable mathematical optimization problem, and uses the powerful global search capability of the intelligent algorithm to quickly lock the roughness combination that meets the physical law in the vast solution space, solving the dilemma of traditional manual parameter tuning.

[0175] The optimization module is used to obtain the optimal roughness of the initial cold period, the severe cold period and the end cold period of the preset area, calculate the initial cold period, the severe cold period and the end cold period according to a preset ratio, and obtain the optimal roughness of the preset area.

[0176] The optimization module executes a time-period parallel optimization and weighted fusion strategy, and plays the role of a senior decision maker, does not blindly trust the calculation results of a single working condition, but calculates a roughness parameter with the most universality by comprehensively evaluating the performance of the pipe network under different weather conditions, thereby significantly improving the running stability of the heating pipe network digital twin system in the whole life cycle.

[0177] The verification module is used to obtain the optimal roughness of the W regions, adjust the trunk flow distribution of the W regions to be consistent with the actual working condition, and optimize and verify the overall heating pipe network roughness.

[0178] The verification module provides a unique global check function of the puzzle type, can dynamically adjust the trunk flow distribution to simulate the real physical connection, and builds the last line of defense of optimization. Only the parameters passing the global hydraulic balance test can be output, so as to ensure that the final model delivered to the user is not only convergent in mathematics, but also real and reliable in engineering physics, and can directly guide the actual production scheduling.

[0179] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present application, which should be covered in the scope of the claims of the present application.

Claims

1. A method for optimizing the roughness of heating pipe networks based on hydraulic simulation, characterized in that, Includes the following steps: Step S1: Based on the heating topology, the heating network is divided into W regions, and a hydraulic simulation model is constructed based on the W regions. Step S2: Obtain heating data, which includes data from the early cold period, the severe cold period, and the late cold period. Organize the heating data into an input parameter set for the hydraulic simulation model and output the optimized roughness of the heating network. Step S3: Based on the optimized roughness and the preset population size, construct a population for the roughness of the heating pipe network. Calculate the fitness values ​​of different individuals in the population using a pre-constructed roughness fitness function to obtain the optimized population. Step S3 includes the following sub-steps: Step S301: Construct a roughness fitness function and preset a termination condition and physical constraint upper and lower limits for pipe roughness. The physical constraint upper and lower limits are preset roughness upper limit as a and preset roughness lower limit as b. The roughness fitness function is calculated as follows: ， in, This represents the roughness fitness value. Let f be the percentage of available pressure error for the i-th pipeline in the f-th region. =1 otherwise =0, k% is a preset percentage; The formula for calculating the percentage of asset pressure error is as follows: ， in, Let f be the percentage of available pressure error for the i-th pipeline in the f-th region. The available pressure in the first stage of the cold period of the i-th pipeline in the f-th region. The available pressure is calculated for the i-th pipeline in the f-th region during the initial cold period; Step S302: When the percentage of the preset total number of pipelines is less than the preset termination condition, the optimization is stopped. The preset total number of pipelines is the number of pipelines whose available pressure error percentage is greater than a preset percentage. Step S4: Obtain the optimal roughness of the early cold period, severe cold period and late cold period of the preset area, and calculate it according to the preset ratio to obtain the optimal roughness of the preset area. Step S5: After obtaining the optimal roughness of W regions, adjust the main flow distribution to be consistent with the actual operating conditions, and then verify the roughness of the overall heating network.

2. The method for optimizing the roughness of a heating pipe network based on hydraulic simulation as described in claim 1, characterized in that, Step S1 includes the following sub-steps: Step S101: According to the heating topology, the heating network is divided into W regions, and a virtual heat source is preset at the junction of each region. Step S102: After referring to the measured operating parameters of the heat station near the gate and verifying the elevation data, preset the water supply temperature, water supply and return pressure and head parameters of the virtual heat source. Step S103: Construct a hydraulic simulation model for the W regions according to the pipeline laying conditions, heat source locations, heating station construction locations, and elevation data.

3. The method for optimizing the roughness of a heating pipe network based on hydraulic simulation as described in claim 2, characterized in that, Step S2 includes the following sub-steps: Step S201: Collect the operation data of the heating network throughout the heating season. After cleaning the abnormal data, obtain the heating data, which includes data from the early cold period, the severe cold period, and the late cold period. Step S202: Organize the data of the early cold period into an input parameter set for the early cold period. The input parameter set for the early cold period includes the heat source parameters, the heat exchange station parameters, and the resource pressure parameter set for the early cold period. Step S203: Organize the severe cold period data into a severe cold period input parameter set, which includes severe cold period heat source parameters, severe cold period heat exchange station parameters, and severe cold period available pressure parameter set; Step S204: Organize the data of the last cold period into a set of input parameters for the last cold period. The set of input parameters for the last cold period includes the heat source parameters, the heat exchange station parameters, and the available pressure parameters for the last cold period.

4. The method for optimizing the roughness of a heating pipe network based on hydraulic simulation as described in claim 3, characterized in that, Step S2 includes the following sub-steps: Step S205: Use the input parameter sets for the early cold period, severe cold period, and late cold period as the input parameter sets for the hydraulic simulation model, and output the optimized roughness of the preset regional heating network.

5. The method for optimizing the roughness of a heating pipe network based on hydraulic simulation as described in claim 4, characterized in that, Step S3 further includes the following sub-steps: Step S303: Based on the preset population size and the optimized roughness, construct an initial population for the roughness of the heating pipe network. The initial population is: ， Step S304: Obtain the fitness values ​​of different individuals in the population through the roughness fitness function; The logic for obtaining the fitness value is as follows: Obtain the roughness of the j-th individual in the f-th region; The initial cold period input parameter set and the roughness of the j-th individual in the f-th region are used as input parameters and input into the hydraulic simulation model to output the calculated available pressure. The available pressure error ratio of the ith pipeline for the j-th individual in the f-th region is calculated using the following expression: ， in, Let i be the percentage of available pressure error for the j-th individual and the ith pipeline in the f-th region. For the i-th individual in the f-th region, calculate the resource pressure. For the j-th individual in the f-th region, the resource pressure during the i-th early winter period; Calculate the available pressure error of all pipelines of the j-th individual in the f-th region, and obtain the fitness value of the j-th individual in the f-th region; The maximum resource pressure error of the j-th individual in the f-th region is obtained from the fitness value of the j-th individual in the f-th region. The optimal individual is selected, and the methods for selecting the optimal individual include: The minimum resource pressure error of the population is obtained from the fitness values ​​of the population. If there are two or more individuals in the population with the same minimum resource pressure error, then the maximum resource pressure error of the individuals corresponding to the minimum resource pressure error of the population is compared, and the individual with the smaller maximum resource pressure error after comparison is taken as the optimal individual.

6. The method for optimizing the roughness of a heating pipe network based on hydraulic simulation as described in claim 5, characterized in that, Step S3 further includes the following sub-steps: Step S305: Determine whether the minimum resource pressure error of the best individual meets the preset termination percentage. If it meets the preset termination percentage, the optimization ends. If it does not meet the preset termination percentage, the optimized population is obtained through elite strategy selection, binary simulated crossover and polynomial mutation. The logic for obtaining the optimized population is as follows: For the initial population, an elite strategy is used to select individuals to construct a new population. The selection method for using the elite strategy to select individuals to construct the new population is the same as the selection method for the optimal individual. Binary simulated crossover was used on the new population; Polynomial mutation is applied to the population after simulated crossover to obtain an optimized population.

7. The method for optimizing the roughness of a heating pipe network based on hydraulic simulation as described in claim 6, characterized in that, Step S4 includes the following sub-steps: Step S401: Using the initial cold period input parameter set as input, repeat step S3. When the proportion of the total number of preset pipes is less than the preset termination condition, obtain the optimal roughness of the initial cold period in the preset region. Step S402: Take the input parameter set of the severe cold period and the input parameter set of the late cold period as inputs respectively, and repeat step S3. When the proportion of the total number of preset pipelines is less than the preset termination condition, obtain the optimal roughness of the severe cold period and the optimal roughness of the late cold period of the preset region respectively. Step S403: Calculate the optimal roughness of the early cold period, the optimal roughness of the severe cold period, and the optimal roughness of the late cold period according to a preset ratio to obtain the optimal roughness of the preset region.

8. The method for optimizing the roughness of a heating pipe network based on hydraulic simulation as described in claim 7, characterized in that, Step S5 includes the following sub-steps: Step S501: Traverse the W regions, take each region as the preset region, and repeat steps S3 and S4 until the optimal roughness of the W regions is calculated. Step S502: Input the optimal roughness of the W regions into the hydraulic simulation model set, stitch the hydraulic simulation models of the W regions together, and remove the virtual heat source at the gate of each region. In the stitched hydraulic simulation model, adjust the resistance value of the gate of each region and adjust the flow distribution relationship until the flow distribution ratio of the main line of each region in the hydraulic simulation model is consistent with the actual working condition data. Step S503: If the percentage of the total number of preset pipes in the overall heating network is less than the preset termination condition, the optimization result is output; if the percentage of the total number of preset pipes is greater than or equal to the preset termination condition, the process returns to steps S2 and S3 to perform optimization again.

9. A hydraulic simulation-based heating network roughness optimization system, applied in the hydraulic simulation-based heating network roughness optimization method as described in any one of claims 1-8, characterized in that, It includes a construction module, a processing module, an optimization module, an optimization search module, and a verification module; The construction module is used to divide the heating network into W regions according to the heating topology, and to construct a hydraulic simulation model based on the W regions. The processing module is used to acquire heating data, which includes data from the early cold period, the severe cold period, and the late cold period. The heating data is organized into an input parameter set for a hydraulic simulation model, and the optimized roughness of the heating network is output. The input parameter set includes input parameter sets for the early cold period, the severe cold period, and the late cold period. The optimization module is used to construct a roughness fitness function, build a population of roughness of the heating pipe network according to the optimized roughness and the preset population size, calculate the fitness value of different individuals in the population through the roughness fitness function, and obtain the optimized population through elite strategy selection, binary simulated crossover and polynomial mutation. The optimization module is used to obtain the optimal roughness of the early cold period, severe cold period and late cold period of the preset area respectively, and calculate the early cold period, severe cold period and late cold period according to the preset ratio to obtain the optimal roughness of the preset area. The verification module is used to obtain the optimal roughness of the W regions, adjust the main flow distribution of the W regions to be consistent with the actual working conditions, and then optimize and verify the roughness of the overall heating network.

Images