Pump station adjusting method and device based on digital twinning, electronic equipment and storage medium

Abstract

The invention relates to the technical field of information processing, in particular to a pump station adjusting method and device based on digital twinning, electronic equipment and a storage medium, and the method comprises the steps: obtaining real-time data and historical data of a pump station, and synchronizing the real-time data and the historical data to a digital twinning platform; obtaining a scheduling strategy generated by the digital twin platform based on the real-time data and the historical data; and adjusting the operation mode of the pump station based on the scheduling strategy. The operation process of the pump station is reproduced based on the digital twinborn technology, and the scheduling strategy is formulated by using the real-time data and historical data of the pump station, so that the fine regulation and optimization of the operation state of the pump station can be realized, and the energy efficiency is improved by optimizing the operation parameters of the pump station; the energy consumption is reduced, the change of external environment and urban demand can be quickly responded, and the operation mode of the pump station is dynamically adjusted.

Description

Technical Field

[0001] The present invention relates to the field of information processing technology, and in particular to a pump station regulation method, device, electronic equipment, and storage medium based on digital twins. Background Art

[0002] In the current field of water resources management, pumping stations, as an important part of water conservancy projects, are traditionally operated mainly by manual monitoring and experience-based control, lacking intelligent and sophisticated management methods. Summary of the invention

[0003] In view of this, the object of the present invention is to provide a pump station regulation method, device, electronic device and storage medium based on digital twin.

[0004] In a first aspect, an embodiment of the present invention provides a pump station adjustment method based on digital twin, the method comprising: Obtain real-time and historical data of the pumping station, and synchronize the real-time and historical data to the digital twin platform; Obtain the scheduling strategy generated by the digital twin platform based on real-time data and historical data; Adjust the operation mode of the pumping station based on the dispatching strategy; Among them, the digital twin platform includes an intelligent control model, which includes an initial correction model, a pump station unit evaluation model, a flow optimization distribution model within the pump station, and a real-time optimization scheduling model of the pump station, which are connected in sequence; the initial correction model includes a water level prediction model and a pump group characteristic curve correction model.

[0005] In combination with the first aspect, the step of obtaining real-time data and historical data of the pump station and synchronizing the real-time data and historical data to the digital twin platform includes: Obtain drawings of pump station units, manage information, and build an initial digital twin model; The initial digital twin model is corrected based on the pump station to obtain the digital twin model of the pump station; The pump station digital twin model and the intelligent control model are integrated into the digital twin platform.

[0006] In combination with the first aspect, the step of obtaining a scheduling strategy generated by the digital twin platform based on the real-time data and the historical data includes: Based on the water level prediction model, the water level prediction sequence scheme in the short term in the future is obtained. At the same time, based on the pump unit characteristic curve correction model, the corrected pump station water pump unit device curve is obtained. Based on the evaluation model of the pump station unit, the operation efficiency / energy consumption of the pump station is used to determine whether the current period is optimized, and the evaluation plan of the pump station unit is obtained; When the evaluation plan of the pump station unit shows that the current unit status is operating in the non-efficient area, the flow optimization allocation model in the pump station and the real-time optimization scheduling model of the pump station are called to obtain the corresponding optimal number of units started and operating flow, and the optimal flow allocation strategy for units in different time periods.

[0007] In combination with the first aspect, the step of obtaining a corrected pump station water pump unit device curve based on the pump unit characteristic curve correction model includes: Obtain pump station characteristic curves and long sequence operation data during the operation of pump station units; The real number coded genetic algorithm is used to calibrate and optimize the parameters of the preset formula of the pump characteristic curve to obtain the optimization result; Based on the optimization results, the target water pump characteristic curve is drawn to update the water pump unit characteristic curve in a rolling manner.

[0008] In combination with the first aspect, based on the evaluation model of the pump station unit, the steps of determining whether to optimize the current period by the pump station operation efficiency / energy consumption and obtaining the evaluation scheme of the pump station unit include: According to the actual project scheduling and operation conditions at different time scales, the water situation data and engineering situation data in the real-time monitoring data are analyzed; Determine the evaluation indicators in the evaluation index system of pump station units and quantify the evaluation indicators; The analytic hierarchy process is used to determine the weights of evaluation indicators and establish a hierarchical structure; Construct a judgment matrix and quantify the scale of the importance relationship between two adjacent levels; Based on the judgment matrix, the relative importance order of each element in this level to an element in the previous level is calculated; The scores and weights of each evaluation index are combined and summed to obtain the overall evaluation result.

[0009] In combination with the first aspect, the step of calculating the relative importance order of each element in the current level to an element in the previous level based on the judgment matrix also includes: Calculate the first ratio of the consistency index and the random consistency; Determine whether the ratio is less than a preset value; If so, determine that the judgment matrix meets the preset requirements; If not, the judgment matrix is ​​adjusted and modified until the second ratio calculated again is smaller than the preset value.

[0010] Combined with the first aspect, when the evaluation scheme of the pump station unit shows that the current unit state is operating in the non-efficient area, the flow optimization allocation model in the pump station and the real-time optimization scheduling model of the pump station are called to obtain the corresponding optimal number of units to be started and the operating flow, and the optimal flow allocation strategy for units in different time periods, including: After processing the characteristic curve of the water pump device, calculate the efficiency of the water pump unit pumping device; Calculate the optimal solution for economic operation within the pumping station; The real-time optimization model of the pump station determines whether it can meet the water diversion requirements corresponding to the optimal solution under high-zone operation; If not, with the goal of minimizing the unit water extraction cost, combined with the optimal solution, the given scheduling time and water transfer volume, the optimal unit operation plan that meets the water transfer task requirements is formulated.

[0011] In a second aspect, the present application provides a pump station regulating device based on digital twin, the device comprising: Data synchronization module, used to obtain real-time data and historical data of the pump station, and synchronize the real-time data and historical data to the digital twin platform; The generation module is used to obtain the scheduling strategy generated by the digital twin platform based on real-time data and historical data; A regulation module, used to adjust the operation mode of the pump station based on the dispatching strategy; Among them, the digital twin platform includes an intelligent control model, which includes an initial correction model, a pump station unit evaluation model, a flow optimization distribution model within the pump station, and a real-time optimization scheduling model of the pump station, which are connected in sequence; the initial correction model includes a water level prediction model and a pump group characteristic curve correction model.

[0012] In a third aspect, the present application provides an electronic device, which includes a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the above method.

[0013] In a fourth aspect, the present application provides a storage medium, in which computer program instructions are stored. When the computer program instructions are read and executed by a processor, the above method is executed.

[0014] The embodiments of the present invention bring the following beneficial effects: The embodiments of the present invention provide a pump station adjustment method, device, electronic device, and storage medium based on digital twins, the method comprising: obtaining real-time data and historical data of the pump station, and synchronizing the real-time data and historical data to the digital twin platform; obtaining a scheduling strategy generated by the digital twin platform based on the real-time data and historical data; adjusting the operation mode of the pump station based on the scheduling strategy; wherein the digital twin platform comprises an intelligent control model, the intelligent control model comprises an initial correction model, a pump station unit evaluation model, a flow optimization distribution model within the pump station, and a real-time optimization scheduling model of the pump station connected in sequence; the initial correction model comprises a water level prediction model and a pump group characteristic curve correction model.

[0015] This application reproduces the operation process of the pump station based on digital twin technology, and uses the real-time data and historical data of the pump station to formulate scheduling strategies, which can achieve fine control and optimization of the operating status of the pump station, thereby optimizing the operating parameters of the pump station to improve energy efficiency and reduce energy consumption, and can quickly respond to changes in the external environment and urban demand, and dynamically adjust the pump station operation mode.

[0016] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without paying any creative work.

[0019] Figure 1 A schematic flow chart of a pump station regulation method based on digital twins provided in an embodiment of the present invention; Figure 2 A schematic diagram of the principle of a pump station regulation method based on digital twins provided in an embodiment of the present invention; Figure 3 A schematic diagram of the connections of the modules in the pump station regulation method based on digital twins provided in an embodiment of the present invention; Figure 4 A schematic diagram of a characteristic curve of synchronous motor efficiency in a pump station regulation method based on digital twins provided in an embodiment of the present invention; Figure 5 A schematic diagram of the structure of a pump station regulating device based on digital twin provided in an embodiment of the present invention; Figure 6 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention.

[0020] Reference numerals: 10-data synchronization module, 20-generation module, 30-adjustment module; 130 - processor, 131 - memory, 132 - bus, 133 - communication interface. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0022] In order to facilitate the understanding of this embodiment, the technical terms designed in this application are briefly introduced below.

[0023] The concept of digital twin technology is a digital idea and technical means. It takes the integrated fusion of data and models as its basis and core. It builds accurate digital mapping of physical objects in real time in digital space, simulates, verifies, predicts and controls the entire life cycle of physical entities based on data integration and analytical prediction, and finally forms an optimized closed loop of intelligent decision-making.

[0024] With the rapid development of digital twin technology, this technology has been widely used in many fields such as industry, urban construction and smart agriculture. It can accurately map the dynamic behavior of physical entities and match a large amount of sensor data, greatly improving the operational efficiency and reliability of the system.

[0025] After introducing the technical terms involved in this application, the application scenarios and design concepts of the embodiments of this application are briefly introduced.

[0026] At present, pump station management mainly relies on manual monitoring and experience-based control, lacking intelligent and sophisticated adjustment methods.

[0027] Based on this, the embodiments of the present application provide a pump station adjustment method, device, electronic device, and storage medium based on digital twins.

[0028] Example 1 This application provides a pump station adjustment method based on digital twins, combined with Figure 1 As shown, the method includes: S110, obtain the real-time data and historical data of the pump station, and synchronize the real-time data and historical data to the digital twin platform.

[0029] S120, obtaining a scheduling strategy generated by the digital twin platform based on real-time data and historical data.

[0030] S130: Adjust the operation mode of the pump station based on the dispatching strategy.

[0031] Among them, the digital twin platform includes an intelligent control model, which includes an initial correction model, a pump station unit evaluation model, a flow optimization distribution model within the pump station, and a real-time optimization scheduling model of the pump station, which are connected in sequence; the initial correction model includes a water level prediction model and a pump group characteristic curve correction model.

[0032] The digital twin-based pump station adjustment method provided in the present application collects data of the pump station in real time, combines the collected data with the digital twin model, reproduces the operation process of the pump station through the digital twin model, and formulates a scheduling strategy using the real-time and historical data of the pump station, thereby ensuring the real-time and accuracy of the scheduling of the pump station group, and can achieve fine-grained control and optimization of the operating status of the pump station, thereby improving energy efficiency and reducing energy consumption by optimizing the operating parameters of the pump station, and can quickly respond to changes in the external environment and urban demand, and dynamically adjust the pump station operation mode.

[0033] Before step S110, the following steps are specifically included: S010, obtain drawings of the pump station units, manage the information, and build an initial digital twin model.

[0034] Specifically, 3DMAX software is used to construct the initial digital twin model of the pump station unit. The modeling process is as follows: (1) Arrange drawings: Arrange the completion drawings of buildings, structures, water machines, etc. required for modeling parts such as pump stations, flood gates, and water intake gates.

[0035] (2) Project decomposition: Based on the modeling scope, personnel division of labor and twin modeling decomposition are carried out. Each group independently inverts each component and finally integrates and assembles the model.

[0036] (3) Creation of family library: The buildings and equipment components in the pump station project are not universal. During the twin modeling process, it is necessary to create a family library for the components based on the large-scale drawings. The scope of the family library includes doors and windows of the pump station plant, electromechanical equipment, etc.

[0037] (4) Graphic management and information management: Graphic management and information management are performed before the model is established. Graphic management mainly includes model color matching, line type and layer determination, and the appearance color matching of the twin model is determined according to the actual decoration plan. Information management mainly includes model geometric information and non-geometric information. Geometric information includes model shape, size, position, etc. Non-geometric information includes project parameters, equipment parameters, manufacturers, etc. Geometric information is determined by the completion drawings, and non-geometric information is provided by the pump station operation and maintenance unit, and attributes are input in the twin model processing.

[0038] (5) Software modeling: Using 3DMAX software, combined with the professional drawings, the twin model is inverted strictly according to the design to ensure the consistency between the twin model and the physical pump station. After the inversion of each professional model is completed, the model is assembled to form the initial digital twin model.

[0039] S020: Correct the initial digital twin model based on the pump station to obtain a digital twin model of the pump station.

[0040] It is understandable that after the initial digital twin model is established, the quality of the initial digital twin model is audited. The audit content includes: whether the model meets the initial design accuracy standard; whether the component positioning is accurate; whether the key parts are consistent with the drawings and the actual situation, etc. If the consistency requirements are not met, timely corrections are made, and finally a digital twin model of the pump station that meets the requirements is obtained.

[0041] S030, integrate the pump station digital twin model and intelligent control model into the digital twin platform.

[0042] It can be understood that the digital twin model of the pump station has a communication connection module, which can be connected to the intelligent control model in step S120. The model training of the intelligent control model can be based on the real-time data and historical data of the digital twin model of the pump station, and the scheduling strategy can be generated based on the model training results. The intelligent control model and the pump station establish a communication connection to obtain real-time data. Specifically: A sensor network is arranged in each key part of the pump station, and real-time data collection can be performed on each key part of the pump station, and the data can be synchronized to the digital twin model in real time. The main data are as follows: The data collected from the main and auxiliary units include at least: real-time operation data and historical data of water pumps and motors; real-time and historical data of water supply pumps, drainage pumps, vacuum pumps, and fans.

[0043] The data collected for electrical equipment includes at least: incoming lines, PTs, outgoing line interval data, high-voltage cabinet data, switch cabinet data, and excitation equipment data of GIS equipment.

[0044] The data collected for the power generation and transformation system at least includes: data of hydropower high and low voltage cabinets, various parameters of power generation and transformation equipment such as generators, main substations, and booster substations.

[0045] The data collected by the vibration monitoring and flow measurement system include at least: vibration monitoring perception data of the unit in all directions and ultrasonic flow velocity, flow rate and other data of the pump station flow channel.

[0046] In combination with the first aspect, step S120 includes: S121, based on the water level prediction model, a water level prediction sequence scheme in the short term in the future is obtained. At the same time, based on the pump unit characteristic curve correction model, a corrected pump station water pump unit device curve is obtained.

[0047] S122, based on the evaluation model of the pump station unit, determine whether to optimize the current period through the pump station operation efficiency / energy consumption, and obtain an evaluation plan for the pump station unit.

[0048] S123, when the evaluation plan of the pump station unit shows that the current unit status is operating in the non-high efficiency zone, the flow optimization allocation model in the pump station and the real-time optimization scheduling model of the pump station are called to obtain the corresponding optimal number of units to be started and the operating flow, and the optimal flow allocation strategy for the units in different time periods.

[0049] In combination with the first aspect, the real-time data of the pump station includes at least real-time water condition data; the historical data includes at least historical water condition data.

[0050] The step of obtaining a water level prediction sequence scheme in the short term in the future based on the water level prediction model in step S121 includes: S1211, using real-time water situation data and historical water situation data, the input layer, hidden layer and output layer in the water level prediction model based on the BP neural network are trained in sequence, and a water level prediction sequence solution is output.

[0051] It is understandable that processing water level information according to actual needs of the project and combining medium- and long-term measured data to establish a suitable pump station water level prediction model is of great significance to pump station regulation, water volume scheduling, building safety, etc. In this embodiment, a BP neural network model with strong nonlinear fitting ability and wide application is used to establish a complex mapping relationship between water status and work status, so as to achieve accurate perception of water level and flow rate.

[0052] Before actual application, the model is built and trained and verified to obtain a water level prediction model that meets the accuracy requirements for application in the testing phase.

[0053] It is worth noting that since the water level status of the water diversion project has the characteristic of changing in an annual cycle, the data length used in the training phase should be greater than or equal to a full year of water level monitoring data.

[0054] BP neural network refers to an artificial neural network that uses the back propagation (BP) algorithm. The back propagation in the BP algorithm actually transmits the calculated prediction error, and then adjusts the weights and thresholds between layers according to the error size. Therefore, a complete process includes two parts. First, the input data is forwarded through the network calculation, and the error is calculated after a prediction value is obtained. The error is then transmitted backward to adjust the network parameters.

[0055] The BP neural network consists of an input layer, an output layer, and a hidden layer in between. Each layer contains a different number of neurons, and the connections between neurons in different layers have corresponding weight values. In addition to the input layer, a certain amount of deviation, namely the threshold, is added to the neurons. When the error between the predicted value and the target value of the output layer exceeds the specified range, the error is transmitted in reverse, and the weights and thresholds of the previous layers are adjusted.

[0056] Then, the specific steps for training a three-layer BP neural network with one hidden layer are as follows: Combination Figure 3 As shown in the figure, the water level, flow and other water monitoring data before and after the pump station are typical time series data. The above water condition data are cleaned and used as the input data set of the above BP neural network, of which 70% are used as the training set, 15% as the validation set, and 15% as the test set. It can be understood that the above BP neural network is trained through the training set; the validation set is used to verify the effect of the trained BP neural network, adjust the parameters, and prevent overfitting; the test set is used to finally evaluate the generalization ability of the BP neural network. Specifically, the water condition detection data includes the number of pump station startups, the average opening of the pump station units, the pump station flow, the water level above the station, the water level below the station, and the number of pump stations at the current moment, and the output is the water level above the station (or the water level below the station) at the current moment.

[0057] In the process of reverse adjustment of weights in BP neural network, the activation function needs to be derived, and the activation function used must meet the differentiable condition. Commonly used activation functions include sigmoid and tansig functions. The general expression of the S-type function is as follows:

[0058] When a=0, b=1, k=1, the above formula can be transformed into

[0059] Combination Figure 2 As shown in the figure, the value range of the S function is (0, 1); when net=0, y=f(net)=0.5; the slope of y=f(net) is larger when net∈(-0.6, 0.6), and changes less in other intervals.

[0060] The specific training process is as follows: ① Assign initial values ​​to the weight matrix: Assign the weight W between the input layer and the hidden layer ji and the weight W between the hidden layer and the output layer kj Randomly assign a set of non-zero and small initial values.

[0061] ② Set the network structure and hyperparameters, such as: set the training target, number of iterations, learning rate, momentum coefficient and other parameters, and then perform the following training calculation steps Ⅰ to VI for each sample in turn until the error convergence condition or the upper limit of the training number is reached, and then stop training.

[0062] Ⅰ) Calculate the output y of each layer starting from the input layer i ,y j and k Among them, the BP neural network data set consists of input data X and output data Y, and generally both are column vectors.

[0063]

[0064]

[0065] The BP neural network in this embodiment uses a linear function as the input layer activation function and a sigmoid function as the hidden layer activation function. The input and output relationship in the network is as follows: For the input layer: The input data of the input layer is the water situation data at each moment , due to the use of linear functions, the output .

[0066] For the hidden layer: The input value of a neuron in the hidden layer is equal to the sum of the input value connected to it multiplied by the corresponding weight, plus the threshold value attached to the neuron, as shown in the following formula:

[0067] In the formula, n represents the number of neurons in the input layer, and j represents the number of neurons in the hidden layer.

[0068] The output value of the hidden layer is the value calculated by the sigmoid function, as shown in the following formula:

[0069] For the output layer: The input value of the output layer neuron is equal to the sum of the input value connected to it multiplied by the corresponding weight plus the threshold value attached to the neuron, as shown in the following formula:

[0070] Similarly, the output value of the output layer is the value after the sigmoid function is calculated, as shown in the following formula:

[0071] II) Calculate the error between the predicted value and the true value of each neuron in the output layer, as shown in the following formula:

[0072] III) After obtaining the error of the output layer prediction value, the error of each neuron in the previous hidden layer is calculated according to the following formula:

[0073] IV) Use the error to adjust the weights and thresholds of each layer:

[0074]

[0075] In the formula, m is the number of iterations, c is the momentum coefficient, which generally ranges from 0.9 to 1; l r It is the learning rate, which has a great influence on convergence and iteration step size, and its value generally ranges from 0.1 to 3.

[0076] V) After a feedforward propagation calculation, determine the error calculated by the output layer With the training goals set size.

[0077] VI) If , then stop training; otherwise, execute III and IV, pass the error forward for calculation, adjust the weights and thresholds between layers, execute I and II, re-perform feedforward forward propagation, and the training ends when the training termination conditions are met.

[0078] After that, the performance of the BP neural network is evaluated. The evaluation indicators of the neural network model are usually selected according to the different research problems. For prediction problems, the evaluation indicators usually include mean square error (MSE), root mean square error (RMSE), mean absolute percentage error (MAPE) and determination coefficient (R 2 ) can be selected according to actual use requirements and are not limited here. Their expressions are:

[0079]

[0080]

[0081]

[0082] In the above formulas, n represents the total number of samples, i represents the i-th sample, and y irepresents the network prediction value of the i-th sample, Represents the corresponding sample true value.

[0083] In combination with the first aspect, in step S121, the step of obtaining a corrected pump station water pump unit device curve based on the pump unit characteristic curve correction model includes: S1212, obtaining the pump station characteristic curve and the long sequence operation data during the operation of the pump station unit.

[0084] S1213, using a real number coded genetic algorithm to calibrate and optimize the parameters of the setting formula of the water pump characteristic curve to obtain an optimization result.

[0085] S1213, based on the optimization result, draw the water pump characteristic curve to roll over and update the water pump unit characteristic curve.

[0086] It is understandable that in the actual operation of the pump station, the uncontrollable emergency operation of the unit due to weather, human operation, emergency events, etc., and the blade wear caused by cavitation and vibration of the pump station unit all year round will cause the hydraulic characteristic curve of the unit to deviate from the design value. Therefore, the pump station characteristic curve combines the long sequence operation data of the pump station unit during operation, and uses the real number coded genetic algorithm to calibrate and optimize the preset formula parameters of the water pump characteristic curve, so as to further draw the water pump characteristic curve and realize the rolling update of the characteristic curve relationship of the water pump unit.

[0087] In view of the fact that the actual monitoring data are all in the high-efficiency area, different operating points are divided into confidence intervals such as the high-efficiency area, the non-high-efficiency area, and the high-efficiency area to increase the correction accuracy of the curve.

[0088] The pump characteristic curve refers to the characteristic curve between the flow rate and head, flow rate and power, flow rate and efficiency, and flow rate and water suction height of the pump at a certain speed. It is understandable that for a pump, at its fixed frequency, there is an area where the pump can operate at a higher efficiency. For the power frequency pump, its theoretical curve conforms to the following empirical formula:

[0089]

[0090] in , , , , is the parameter of the characteristic curve, H is the height, and Q is the water flow of the pumping station.

[0091] Since the update of the pump station characteristic curve can be converted into the solution of the parameters in the above formula, the object for finding the optimal solution is five coefficient parameters, and each of them has its own value range. If binary encoding is used, it will be difficult to decode. Therefore, real number encoding is used here. The specific implementation steps are as follows: ① Determine the search range for each parameter based on the results of the least squares method.

[0092] ②Initialize the population size.

[0093] ③ Establish the corresponding fitness function based on the objective function.

[0094] ④Select appropriate selection operator, crossover operator and mutation operator.

[0095] ⑤Iterate the results based on the operators in steps ③ and ④ and the fitness function, and output the final result.

[0096] In this application, since a real-number coded genetic algorithm is used, the corresponding crossover operator and mutation operator are also different from those in the binary genetic algorithm: 1) Crossover operator The traditional crossover operator for real-coded genetic algorithms is to add a random ratio of the two individuals to be crossed, that is:

[0097]

[0098] Generally, the value of α is limited to [0, 1], so the single-point crossover method will limit its global search capability to a certain extent.

[0099] On this basis, the method of this project introduces a second random quantity β to enhance the global search ability of the crossover operator, namely:

[0100]

[0101] If the value of the new individual after calculation by the above formula exceeds the limited search domain, the value calculated by the traditional method is used instead.

[0102] 2) Mutation Operator Commonly used mutation operators in real-coded genetic algorithms are: uniform mutation, boundary mutation, non-uniform mutation, and Gaussian mutation. Uniform mutation refers to randomly selecting a number in the search domain to directly replace the value of the gene position that needs to be mutated. This method is the easiest to implement, but it does not take into account that the population will be closer to the optimal solution as the iteration proceeds. Therefore, in the later stages of the iteration, its progressive nature will affect the speed at which the population approaches the optimal solution; boundary mutation is a variation of uniform mutation. Its principle is that when performing boundary mutation operations, one of the two corresponding boundary genes of the locus is randomly selected to replace the original gene value, which has the same problem as uniform mutation. Both non-uniform mutation and Gaussian mutation randomly select values ​​around the current value of the gene position that needs to be mutated, and as the iteration progresses, the mutation area will be narrowed. The method of this project selects non-uniform mutation. The specific mutation operator is as follows. Assume that the value range of the mutation point is [ X min , X max ]:

[0103] In the formula, △ (t,y) is a random number in the range [0, y] that conforms to a non-uniform distribution, satisfying that as the evolutionary generation t increases, △ (t,y) The probability close to 0 also increases, that is:

[0104] Where r is a random number in the range [0, 1] that conforms to a uniform distribution, T represents the maximum evolutionary generation, and b is a system parameter that determines the degree of dependence of the random number perturbation on the evolutionary generation t.

[0105] 3) Algorithm configuration Before using the algorithm to fit the characteristic curve of the water pump, a series of parameter configurations need to be given to enable the algorithm to play its role. Different from the binary coding algorithm, the search interval of each bit in the chromosome is no longer a simple 0 and 1. The chromosomes composed of the coefficients that need to be optimized for the fitted polynomial here each have different search domains. If the search domain is too large, the size of the population and the number of iterations in the current algorithm may not be able to find the optimal solution; but if the population size and the number of iterations are increased, the solution time of the entire algorithm will be greatly increased. Therefore, how to give a reasonable search domain for each coefficient to be solved has become an issue that cannot be ignored in parameter configuration. Specifically, since the order of magnitude of each coefficient of the polynomial is not in the same series, a unified search domain cannot be given. Therefore, it is thought of using the least squares method to fit the data once, and then define 30% of the coefficient above and below as the search domain based on the obtained coefficient. In addition, in the improved real number coded genetic algorithm, the coefficients obtained by the least squares method are also used as the initial value of the optimal solution, so that the algorithm can obtain a solution that is not worse than that obtained by the least squares method during the optimization process.

[0106] In addition, since in actual working conditions, water pumps generally operate in the high-efficiency zone, the collected data points may not cover the entire operating range of the water pump, and the collected data points vary according to the running time of the water pump itself and the change of external water consumption. The distribution of data points obtained in a short time is generally uneven. And for different pumps, there will be a situation where they run in the non-high-efficiency zone for a short time. Therefore, if you simply draw the curve directly without judging all the data points, the huge density difference of the data points will have a great deviation effect on the overall drawn curve. Therefore, we need to first divide the operating range of the water pump, and then draw and analyze it. Since the operating points of the water pump are mainly concentrated in the high-efficiency zone during operation, the area can be divided accordingly, that is, to the left of the high-efficiency zone, the high-efficiency zone, and to the right of the high-efficiency zone. Naturally, if there are more data points distributed in the area, the higher the confidence of the area, which means that the credibility of the drawn curve is higher. The data collected in the actual operation are randomly selected, and some of the data are used for effect verification.

[0107] Among them, data collection and processing include: collecting the measured data of the pump station's long sequence flow, head, blade placement angle, unit operating efficiency, etc., and establishing a flow-head-blade placement angle historical data summary table and a flow-head-efficiency historical data summary table.

[0108] The correction of the pump station characteristic curve includes: using the least squares method to find the shortest distance between the historical data point and the original pump station characteristic curve, and by moving the characteristic curve, obtaining the pump station characteristic curve and corresponding formula that conforms to the actual operating conditions, and obtaining the optimization results of each parameter, and then drawing the target water pump characteristic curve to roll over and update the water pump unit characteristic curve.

[0109] In combination with the first aspect, step S142 includes: S1421, according to the actual project scheduling and operation conditions at different time scales, the water situation data and engineering situation data in the real-time monitoring data are analyzed.

[0110] S1422, determine the evaluation indicators in the evaluation indicator system of the pump station unit and quantify the evaluation indicators.

[0111] S1423, use the analytic hierarchy process to determine the evaluation index weights and establish a hierarchical structure.

[0112] S1424, construct a judgment matrix and quantify the scale of the importance relationship between two adjacent levels; S1425, based on the judgment matrix, the relative importance order of each element in the current level to an element in the previous level is calculated; S1426 combines the scores and weights of each evaluation indicator to perform weighted summation to obtain an overall evaluation result.

[0113] It is understandable that the water delivery status of the pumping station in each channel of the water diversion project is directly related to the safety of the project. Especially in the case of real-time dispatching and operation, it is necessary to control the water levels upstream and downstream of the pumping station. Therefore, it is necessary to establish a corresponding pumping station unit evaluation model to characterize the real-time operation status of the pumping station. The pumping station unit evaluation model should be able to comprehensively consider various dispatching indicators, judge the comprehensive water status of the pumping station, and meet the dispatching management personnel's control over the overall real-time operation status of the pumping station. Specifically, by analyzing the changes in water level, flow, water volume, energy consumption, cost and other state quantities during the water diversion period (combined with Figure 3 As shown in the figure, an evaluation index system is established from the aspects of scheme execution, water diversion effect, and operating cost. Based on the hierarchical analysis method, a monthly / daily scale water scheduling effect evaluation model is established to meet the quantitative water scheduling and project scheduling operation conditions. The scheduling effect is evaluated from the aspects of plan completion, water supply guarantee level, water transmission stability, water supply efficiency, etc., and a reference is provided for the adjustment of water diversion tasks in the next stage.

[0114] The pump station unit evaluation model can review and specifically summarize the operation status of the pump station within a year, summarize and calculate relevant information such as water level, water volume, unit efficiency, etc., evaluate the health of the pump station units during a specific scheduling period from the perspectives of safety, efficiency, and benefits, determine the startup sequence of the pump station units, and achieve energy-saving and efficient operation of the pump station during the water diversion period.

[0115] S1221, based on the actual project scheduling and operation status at different time scales, statistical summary of water conditions, construction conditions and other related data.

[0116] S1222, determine the evaluation indicators in the evaluation index system of the pump station unit and quantify the evaluation indicators. Specifically, the steps of determining the evaluation indicators include: the evaluation index system of the pump station unit consists of three levels, the first-level indicators include: efficiency, safety, and benefit; the second-level indicators include: program completion, water supply guarantee rate, and water supply efficiency; the third-level indicators include: water supply plan completion, water diversion volume, water supply guarantee rate, water supply efficiency, etc. The steps of quantifying the evaluation indicators include: ① Water supply plan completion: By comparing the planned water diversion volume and the actual water diversion volume, the completion of the water supply plan is evaluated. The plan completion status is divided into the monthly / daily plan completion rate, and the calculation method is as follows:

[0117] Where: PCR is the plan completion rate; AWS is the actual water supply; PWS is the planned water supply.

[0118] The scoring is directly based on the plan completion index, and 100 points will be awarded when the plan completion index is greater than 100%.

[0119] ② Water diversion volume: By comparing the planned water diversion volume and the actual water diversion volume of each water diversion gate, the water diversion volume of the overall water diversion gate is evaluated. The evaluation object is mainly each water diversion gate, which is divided into two time scales: month / day. The calculation method is as follows:

[0120] In the formula: WOR is the water diversion rate compliance rate of each water diversion gate; AWdg is the actual water diversion volume of each water diversion gate; PWdg is the planned water diversion volume of each water diversion gate.

[0121] The scoring is based on the average of the calculation results of each water diversion gate. When WOR is greater than 100%, 100 points will be awarded.

[0122] ③ Water supply guarantee rate: Compare and analyze the actual water supply guarantee times and total times in different periods to evaluate the degree of water supply guarantee. The actual water supply guarantee times in different periods can be expressed by the number of days when the actual water supply is equal to the water demand. When the actual water supply is equal to the water demand, the water supply is considered to be guaranteed. The indicator is expressed as a monthly / daily water supply guarantee rate, and the calculation method is as follows:

[0123] In the formula, GR is the water supply guarantee rate; GT is the actual water supply guarantee times of the entire line; AT is the total times; AW is the actual water supply; WD is the water demand; T is the total number of days. The water supply guarantee index is directly used for scoring.

[0124] ④ Water supply efficiency: The water supply efficiency index is mainly measured by water loss. Water loss is obtained by loss rate. The loss rate is the ratio of river water loss within a period of time to the river inflow within the period of time.

[0125] S1223, use the analytic hierarchy process to determine the evaluation index weights and establish a hierarchical structure.

[0126] It is understandable that the quantitative representation of the relative importance of each indicator in the entire indicator system and whether the weight determination is reasonable will have a decisive impact on the comprehensive evaluation results and evaluation quality. In this application, the hierarchical analysis method is used to determine the weights of the evaluation indicators. The basic steps are as follows: establish a hierarchical structure, construct a judgment matrix, single sorting of the hierarchy, consistency verification, and calculate the evaluation results of the scheme.

[0127] In combination with the first aspect, step S1223 of establishing a hierarchical structure includes: S12231 According to the results of problem analysis, the elements of the system are divided into the following three levels.

[0128] ①The highest level: the target level. This level contains the overall goal that the system is to achieve.

[0129] ② Middle layer: Principle layer. In this layer are the various principles to be adopted to achieve the overall goal.

[0130] ③The lowest level: the solution level. In this level are various feasible solutions that the system wants to choose.

[0131] S1224, construct a judgment matrix and quantify the scale of the importance relationship between two adjacent levels.

[0132] Among them, construct a judgment matrix: analyze the relative importance of the next level element to the previous element and express it in the form of a judgment matrix. For example, if the element c in the C level is related to P1, P2…P in the next level, n There is a connection. See Table 1.

[0133] Table 1 is a schematic diagram of the judgment matrix.

[0134]

[0135] The element P in the matrix ij Indicates P i With P j The relative importance of the comparison is estimated by introducing a suitable scale and performing P i With P j The AHP method uses a scaling method to give a quantitative scale for the comparison of different situations: 1 - Equally important. Indicates that two elements are equally important to a certain attribute.

[0136] 3 - Slightly important. It indicates that when comparing two elements, one element is slightly more important than the other.

[0137] 5 - Apparently important. It indicates that when comparing two elements, one element is apparently more important than the other.

[0138] 7 - Very important. It indicates that when comparing two elements, one element is very more important than the other.

[0139] 9 - Extremely important. It indicates that when comparing two elements, one element is extremely more important than the other.

[0140] 2, 4, 6, 8 - Represent a quantitative scale when a compromise is needed between the above two criteria.

[0141] The reciprocals of the above numbers - Represent inverse comparison.

[0142] Obviously, when i = j, there is P ij = 1; when i < j, there is P ij = 1 / P ji Step S1425 can usually be calculated by the root method. The calculation steps are as follows: ① Calculate the product of the elements in each row of the judgment matrix:

[0143] ② Calculate the nth root of M i :

[0144] ③ Normalize Wi: .

[0145] Among them, after S1225, it also includes: S1227, calculate the consistency index C.I. and the random consistency ratio C.R.

[0146] S1228, judge whether C.I. is less than 0.1.

[0147] If so, execute step S12281; if not, execute step S12282.

[0148] S12281, determine that the judgment matrix is consistent.

[0149] S1229, determine that the judgment matrix is inconsistent.

[0150] It can be understood that the consistency check is performed after step S1425, the purpose of which is to evaluate the quality of the single sorting and total sorting of the hierarchy, which is mainly accomplished by calculating the consistency index CI and the random consistency ratio CR.

[0151] The calculation formula of CI is: CI = (the maximum feature of the judgment matrix - the number of elements at this level) / (the number of elements at this level - 1).

[0152] The calculation formula of CR is: CR=CI÷RL. RL is a coefficient, called a random index. Judgment matrices of different orders (n) should take different RI values. For example, when n=3, RI should be 0.58; when n=7, RI should be 1.32, and so on.

[0153] When CR<0.1, the judgment matrix is ​​considered to have acceptable and satisfactory consistency. Otherwise, the judgment matrix needs to be adjusted and corrected to satisfy CR<0.1 and thus have satisfactory consistency.

[0154] Before actual application, the evaluation model of the pump station unit is constructed. Specifically: Data collection: Based on the real-time project scheduling and operation status, the scheduling operation data, project operation parameters and other information in the pump station project are collected and counted.

[0155] Based on the preset evaluation rules, establish an evaluation index system: The real-time evaluation index system of the pumping station consists of four parts: operation safety, operation efficiency, power generation efficiency, and unit selection of the pumping station.

[0156] Comprehensively quantify multiple evaluation indicators: In order to construct the real-time status model of the pump station, it is necessary to comprehensively consider the impact of various indicators and quantify each indicator. The evaluation indicators of the pump station are mainly quantified. The quantification process of the other building indicators can refer to this example to quantify the evaluation indicators based on the engineering parameter information of each building. After all the evaluation indicators are quantified, the calculation formula in the model method is used to obtain the real-time status evaluation score of the pump station.

[0157] The evaluation index includes multiple preset grade evaluation indexes corresponding to the water levels in the forebay of the pumping station as shown in Table 2. Thus, the water level in the forebay of the pumping station can be evaluated at seven grades for the current operating status.

[0158] Table 2 is a schematic diagram of the water level classification of the pump station forebay.

[0159]

[0160] It also includes: early warning time level division evaluation index. As shown in Table 3, the early warning time level division uses six levels to evaluate the current operating status.

[0161] Table 3 is the classification table of warning time levels.

[0162]

[0163] It also includes: water level fluctuation evaluation index, as shown in Table 4, the water level fluctuation grade can be divided into eight grades to evaluate the current operating status.

[0164] Table 4 is a table for the classification of water level fluctuation levels.

[0165]

[0166] It also includes: water level high / low limit warning value evaluation indicators. Combined with Table 5, the water level high / low limit warning value level division can be used to evaluate the current operating status at eight levels.

[0167] Table 5 Evaluation indicators for classification of high / low water level limits.

[0168]

[0169] Evaluation indicators of pump station efficiency warning values ​​are shown in Table 6. Pump station efficiency warning value levels can be divided into four levels to evaluate the current operating status.

[0170] Table 6 is the evaluation index table for the classification of pump station efficiency.

[0171]

[0172] The evaluation index of pump station operation temperature is shown in Table 7. The temperature warning value of the pump station can be divided into five levels to evaluate the current operation status.

[0173] Table 7 shows the evaluation indicators for station operating temperature level classification.

[0174]

[0175] In combination with the first aspect, step S123 includes: S1231, after processing the characteristic curve of the water pump device, calculate the efficiency of the water pump device of the water pump unit.

[0176] S1232, calculate the optimal solution for economic operation in the pump station.

[0177] S1233, the real-time optimization model of the pump station determines whether it can meet the water diversion requirements corresponding to the optimal solution under high-zone operation.

[0178] If not, execute step S1234.

[0179] With the goal of minimizing the unit water extraction cost, the optimal unit operation plan that meets the water diversion task requirements is formulated by combining the optimal solution, the given scheduling time and the water diversion volume.

[0180] It is understandable that the operation of a pump station needs to consider how the total flow rate is reasonably distributed among the units in the station to ensure that the pump station operates economically and reasonably under the conditions of existing equipment, improve the operating efficiency of the pump station, and reduce the cost of water extraction. Therefore, it is necessary to build an optimal flow distribution model within the pump station to carry out optimization work at the unit level and the pump station level to achieve the goal of maximizing the project benefits under the premise of ensuring the safe operation and scheduling of the project.

[0181] The flow optimization distribution model in the pump station is actually a spatial optimization problem of the flow distribution among the units in the pump station. The units that can be put into operation in the pump station are artificially numbered in sequence, and each unit is abstracted as a stage. In this way, the flow optimization problem becomes an optimization problem of a multi-stage decision-making process, which can be solved by using an optimization algorithm. Because the unit selection in the design stage is based on the design working conditions of the pump station, there are often some unfavorable working conditions that cannot be achieved during the actual operation. For the operating working points of each pump station, when the flow is constant, how to distribute the flow to each unit can ensure that each pump station can operate within the high-efficiency range. According to the design data and operation characteristics, the flow optimization distribution model in the pump station is constructed to analyze the distribution results of the units in the station corresponding to different flow rates when the pump station operation efficiency reaches the highest under different operating flow and head combinations of each pump station, and provide a basis for the operation scheduling when the upstream flow changes during the actual water diversion process.

[0182] The solution process of the flow optimization distribution model in the pump station is as follows: (1) Objective function The goal of the first-level model is to distribute the flow of the units in the pumping station to maximize its total efficiency while satisfying various equality and inequality constraints. The objective function of a pumping station containing n units is expressed as:

[0183] In the formula, The j-th pump station has a flow rate of , lift The total efficiency of the case; is the total flow of the pumping station in the kth period, is the head of the j-th pump station; and are the flow rate and efficiency of the i-th unit of the j-th pumping station respectively.

[0184] (2) Decision variables The flow rate of each unit in the pump station is used as the decision variable. The smaller the discrete step size is, the higher the calculation accuracy is, but the amount of calculation increases significantly.

[0185] (3) Constraints The constraints include total flow constraint and unit flow capacity constraint, and their mathematical expressions are shown in the following formulas.

[0186]

[0187]

[0188] In the formula, , are the minimum and maximum flow rates allowed to pass through the i-th unit respectively.

[0189] The model is based on deterministic algorithms such as dynamic programming algorithms that do not involve random variables. The test report of the water pump device corresponding to the actual installation unit of the project and the device test report of the matching motor were used during the model development. Considering the parameter deviation caused by the wear of the water pump unit, the relevant parameters of the model are summarized in the configuration file, which is easy for humans to read, write and modify.

[0190] Step S143, if the evaluation scheme of the pump station unit obtained in step S142 shows that the current unit state is operating in the non-high efficiency zone, when the flow optimization distribution model in the pump station is called, first, the characteristic curve of the water pump device is processed, including flow-head-efficiency fitting discrete processing and flow-head-blade placement angle / speed fitting discrete processing; secondly, the efficiency of the water pump unit's pumping device is calculated; then, the optimal solution for economic operation in the pump station is calculated.

[0191] Among them, the steps of flow-head-efficiency fitting discrete processing are as follows: extract the corresponding operating points of flow-head-efficiency from the prototype characteristic curve of the pump station, use Matlab to fit the extracted operating points into a third-order surface, and at the same time obtain the flow-head-efficiency relationship. Use the fitted formula to discretize the flow-head-efficiency curve with a step size of 0.1.

[0192] The steps of flow-head-blade placement angle / speed fitting discrete processing are as follows: extract the operating points corresponding to flow-head-blade placement angle on the prototype characteristic curve, use Matlab to fit the extracted operating points into a second-order surface, and at the same time obtain the flow-head-blade placement angle relationship. Use the fitted formula to discretize the flow-head-blade placement angle curve with a step size of 0.1.

[0193] The steps of calculating the efficiency of the pumping device include: it is known that the main carrier for completing the pumping task during the operation of the pumping station is the pumping device, which is composed of a water pump, a power machine, a transmission device, a pipeline and its accessories. The efficiency of the pumping device is used as the actual operating efficiency in the calculation of the pumping station. The calculation formula for the efficiency of the pumping device of the pumping station is:

[0194] In the formula They represent the efficiency of the pumping device, Indicates the efficiency of the pump device, Indicates transmission efficiency, Indicates motor efficiency, Indicates the frequency conversion efficiency.

[0195] For mechanical transmission with direct connection between motor and pump, . Frequency conversion efficiency of PWM high voltage inverter is around 96%, when the pump station does not adjust the frequency = 1. It is known that the motor efficiency of large motors is about 94%, that is, =94%. When the load is greater than 50%, it can be considered that the motor efficiency remains basically unchanged.

[0196] For the motor efficiency at different operating points of a specific unit, the efficiency of the output of a single unit at different operating points of a pump station can be calculated according to the following formula:

[0197]

[0198] In the formula, is the motor output power, kw; is the density of water ; is the acceleration due to gravity, m / s²; is the flow rate of the i-th unit, m 3 / s; is the head of the i-th unit, m; is the rated power of the motor, kW.

[0199] When sought ≥50%, the motor efficiency is calculated as 94%; when <50%, the motor efficiency is based on Figure 4 Take its slope for calculation.

[0200] After determining the transmission efficiency, motor efficiency and frequency conversion efficiency under each pump station operating condition, the efficiency of the pumping device of each pump station can be calculated according to the formula with a discretization step of 0.1 within the feasible domain of the pump station operating condition.

[0201] Afterwards, one-dimensional and two-dimensional pumping device characteristic curves can be drawn based on the calculated pumping device efficiency. Furthermore, after the pumping device efficiency surface of the pumping station is determined based on the one-dimensional and two-dimensional pumping device characteristic curves, the optimal solution for economic operation in the pumping station can be calculated.

[0202] Among them, the calculation of the pump station specifically includes two parts: operational flow range analysis and optimization calculation.

[0203] Analysis of the operable flow range of pump stations at all levels: First, based on the characteristic curves of each unit in the pump station and the design head range, the operable flow range of a single unit is calculated, and then the operable flow range of each pump station when different numbers of units in the pump station are put into operation is preliminarily determined, and a "flow-head feasible domain" analysis is performed.

[0204] After the actual operation of the project, there will be certain differences in the performance between different water pump units, which means that the operating flow ranges and head ranges of different units cannot be simply accumulated. The above flow ranges need to be redefined using a similar method.

[0205] Optimization calculation within a single-stage pump station: The efficiency of each pump unit in the pump station at discrete flow and head operating points is used as the input of the pump station flow optimization allocation model to calculate the optimal efficiency of the pump station at all operating points and the optimal flow allocation within the station. Based on the model calculation results, an optimal allocation plan for the number of pump station start-up units and unit flow is generated to meet the optimal operation of the pump station during the water diversion process.

[0206] Due to the influence of factors such as water level fluctuations and changes in operating conditions, the pump station does not always operate under the optimal operating conditions obtained by the flow optimization distribution model within the pump station. Therefore, it is necessary to build a real-time optimization scheduling model for the pump station based on the real-time water conditions and working conditions of the project, with the goal of maximizing the comprehensive benefits of the project, and optimize the operation of the pump station in real time.

[0207] Specifically, when S1232 provides the optimal solution, the real-time optimization scheduling model of the pump station needs to judge whether it can meet the water transfer requirements and operate in the high-efficiency area in step S1233. If it cannot be met, then in step S1234, based on the optimal solution of the economic operation model in the pump station, under the conditions of given scheduling time and water transfer volume tasks, with the goal of minimizing unit water lifting cost, the optimal unit operation plan that meets the water transfer task requirements is obtained.

[0208] Among them, step S1234 specifically identifies the upstream and downstream water level boundary sequence of the pump station, the initial operating flow, the model scheme calculation start time and other information. Under the condition of meeting the actual operation requirements in the scheduling process, the objective function, decision variables and constraints used in the model scheme calculation are as follows: 1) Objective function

[0209] in, is the total pumping device cost, unit: yuan; W is the total water transfer volume during the dispatch period, unit: m 3 . The flow rate of the new station in each period, unit: m 3 / h; It is the operation time of a single station at different pumping station flow rates, unit: h; It is the number of stage-by-stage control in the scheduling period.

[0210] 2) Decision variables Since it involves segmented operation time during the scheduling period, the operating flow of each unit in the pumping station, the startup status of the pumping station, the operating time of the pumping station and the number of operating stages of the pumping station are used as decision variables to calculate the economic operation model within the station with the goal of minimizing the unit water extraction cost.

[0211] 3) Constraints ①River constraints

[0212] In the formula, is the lower limit water level of the river, m; is the upper limit water level of the river, m.

[0213] ②Satisfy water transfer tasks

[0214] In the formula, is the actual water transfer volume; The water transfer task within the scheduling time of the model calculation plan.

[0215] ③Resonance constraint Collect the operating points where resonance occurs during the operation of the pump station units and exclude such operating conditions.

[0216] Among them, a real-time optimization and scheduling model of the pump station is constructed before actual application, including: data collection, model constraints, simulation and calculation of channel water level change process, real-time optimization and regulation, and handling of regulation failure.

[0217] Specifically, data collection includes: collecting and arranging the real-time downstream water level, real-time upstream water level, real-time flow rate and downstream water level or lake water level of the next pumping station that needs to be regulated from the database, and making them into model input data.

[0218] Model constraints include: the water transfer requirements within the scheduling time generated by the ten-day scheduling model received by the real-time optimization scheduling model or issued by the flood control and drought relief command unit as model constraints.

[0219] The simulation calculation of channel water level change process includes: first, through the flow imbalance of adjacent pumping stations or canals, the storage model is used to calculate the change process of the tank storage between pumping stations. Then, through the water level-storage relationship, channel water body generalization and other methods, according to the change process of the tank storage between adjacent pumping stations, the channel water level change process is calculated: ① The water level-storage results calculated by the one-dimensional hydrodynamic numerical simulation model, and the water level change process calculated by the tank storage change process between adjacent pumping stations; ② According to the basic parameters of the channel, the channel water body is generalized, and the change in water height is simulated and calculated based on the volume change process of the water body to obtain the change process of the channel water level.

[0220] Real-time optimization and control include: according to the simulated channel water level change process, the head change process of the pump station to be regulated is obtained. According to the flow and head sequence of the pump station, and the flow-head-efficiency result table of the pump station, each level of pump station is tested for the high-efficiency area. If the pump station can be guaranteed to operate in the high-efficiency area during the dispatch time, the pump station will not be regulated during the dispatch time.

[0221] If it is not possible to ensure that the pumping station to be regulated always operates in the high-efficiency zone, the scheduling decision is optimized. The flow rate and control time after regulation of each level of pumping station are set as optimization parameters, and the particle swarm optimization algorithm (PSO) is used to carry out the real-time optimization and regulation process with the goal of minimizing the unit water extraction cost of the pumping station. The regulation time and the flow rate before and after regulation are output to obtain a scheduling plan that meets the constraints.

[0222] The control failure processing includes: if it is impossible to generate a scheduling decision that can meet the water transfer task conditions given by the ten-day scheduling model, or the generated scheduling decision cannot be operated in the high-efficiency area, then feedback will be given to the ten-day scheduling model, and the ten-day scheduling model will modify the water transfer task according to the feedback content. If the generated scheduling decision suggestion can meet the water transfer requirements given by the ten-day scheduling model and operate in the high-efficiency area, the generated scheduling decision suggestion will be sent to the scheduling personnel for confirmation.

[0223] In the second aspect, the present application provides a pump station adjustment device based on digital twin, combined with Figure 5 As shown, the device includes: a data synchronization module 10, a generation module 20 and an adjustment module 30.

[0224] The data synchronization module 10 is used to obtain the real-time data and historical data of the pump station, and synchronize the real-time data and historical data to the digital twin platform.

[0225] The generation module 20 is used to obtain the scheduling strategy generated by the digital twin platform based on real-time data and historical data.

[0226] The adjustment module 30 is used to adjust the operation mode of the pump station based on the scheduling strategy.

[0227] Among them, the intelligent control end includes a digital twin platform including an intelligent control model. The intelligent control model includes an initial correction model, a pump station unit evaluation model, a flow optimization distribution model within the pump station, and a real-time optimization scheduling model of the pump station, which are connected in sequence; the initial correction model includes a water level prediction model and a pump group characteristic curve correction model.

[0228] In a third aspect, the present application provides an electronic device, Figure 6 As shown, the electronic device includes a memory 131 and a processor 130. The memory 131 is used to store computer programs, and the processor 130 runs the computer programs to enable the electronic device to perform the above method.

[0229] Furthermore, combined with Figure 6 The electronic device shown further includes a bus 132 and a communication interface 133 , and the processor 130 , the communication interface 133 and the memory 131 are connected via the bus 132 .

[0230] The memory 131 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 133 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used. The bus 132 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 6 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0231] The processor 130 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 130. The above processor 130 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in the embodiments of the present invention can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module may be located in a storage medium mature in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 131, and the processor 130 reads the information in the memory 131 and completes the steps of the method of the above embodiment in combination with its hardware.

[0232] In a fourth aspect, an embodiment of the present application provides a storage medium, in which computer program instructions are stored. When the computer program instructions are read and executed by a processor, the above method is executed.

[0233] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0234] In addition, in the description of the embodiments of the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the internal communication of two components. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0235] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0236] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0237] Finally, it should be noted that the above embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above embodiments, those skilled in the art should understand that any person skilled in the art can still modify the technical solutions recorded in the above embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A pump station regulation method based on digital twin, characterized in that: The method comprises: Acquire real-time data and historical data of the pump station, and synchronize the real-time data and historical data to the digital twin platform; Obtaining a scheduling strategy generated by the digital twin platform based on the real-time data and historical data; Based on the scheduling strategy, adjusting the operation mode of the pump station; Among them, the digital twin platform includes an intelligent control model, which includes an initial correction model, a pump station unit evaluation model, a flow optimization distribution model within the pump station, and a pump station real-time optimization scheduling model connected in sequence; the initial correction model includes a water level prediction model and a pump group characteristic curve correction model.

2. The method according to claim 1, characterized in that Before the step of obtaining the real-time data and historical data of the pump station and synchronizing the real-time data and historical data to the digital twin platform, the steps include: Obtain drawings of pump station units, manage information, and build an initial digital twin model; Correcting the initial digital twin model based on the pump station to obtain a digital twin model of the pump station; The pump station digital twin model and the intelligent control model are integrated into the digital twin platform.

3. The method according to claim 1, characterized in that The step of obtaining the scheduling strategy generated by the digital twin platform based on the real-time data and historical data includes: Based on the water level prediction model, a water level prediction sequence scheme in the short term in the future is obtained. At the same time, based on the pump unit characteristic curve correction model, a corrected pump station water pump unit device curve is obtained; Based on the evaluation model of the pump station unit, the operation efficiency / energy consumption of the pump station is used to determine whether the current period is optimized, and the evaluation plan of the pump station unit is obtained; When the evaluation plan of the pump station unit shows that the current unit status is operating in the non-efficient area, the flow optimization allocation model in the pump station and the real-time optimization scheduling model of the pump station are called to obtain the corresponding optimal number of units started and operating flow, and the optimal flow allocation strategy for units in different time periods.

4. The method according to claim 3, characterized in that The steps of obtaining a corrected pump station water pump unit device curve based on the pump unit characteristic curve correction model include: Obtain pump station characteristic curves and long sequence operation data during the operation of pump station units; The real number coded genetic algorithm is used to calibrate and optimize the parameters of the preset formula of the pump characteristic curve to obtain the optimization result; Based on the optimization result, a target water pump characteristic curve is drawn to roll over and update the water pump unit characteristic curve.

5. The method according to claim 3, characterized in that: Based on the evaluation model of the pump station unit, the steps of determining whether the current period is optimized by the pump station operation efficiency / energy consumption and obtaining the evaluation plan of the pump station unit include: According to the actual project scheduling and operation conditions at different time scales, the water situation data and construction situation data in the real-time monitoring data are statistically summarized; Determine the evaluation index in the evaluation index system of the pump station unit and quantify the evaluation index; The analytic hierarchy process is used to determine the weights of evaluation indicators and establish a hierarchical structure; Construct a judgment matrix and quantify the scale of the importance relationship between two adjacent levels; Based on the judgment matrix, the relative importance order of each element in the current level to an element in the previous level is calculated; The scores and weights of the evaluation indicators are combined and weighted to obtain an overall evaluation result.

6. The method according to claim 5, characterized in that The step of calculating the relative importance order of each element in the current level to an element in the previous level based on the judgment matrix also includes: Calculate the first ratio of the consistency index and the random consistency; Determining whether the ratio is less than a preset value; If so, determine that the judgment matrix meets the preset requirements; If not, the judgment matrix is ​​adjusted and corrected until the second ratio calculated again is smaller than the preset value.

7. The method according to claim 1, characterized in that When the evaluation scheme of the pump station unit shows that the current unit status is operating in the non-efficient area, the flow optimization allocation model in the pump station and the real-time optimization scheduling model of the pump station are called to obtain the corresponding optimal number of units to be started and the operating flow, and the optimal flow allocation strategy for units in different time periods, including: After processing the characteristic curve of the water pump device, calculate the efficiency of the water pump unit pumping device; Calculate the optimal solution for economic operation within the pumping station; The real-time optimization model of the pump station determines whether the water diversion requirements corresponding to the optimal solution can be met under high-zone operation; If not, with the goal of minimizing the unit water extraction cost, combined with the optimal solution, the given scheduling time and water transfer volume, formulate the optimal unit operation plan that meets the water transfer task requirements.

8. A pump station regulating device based on digital twin, characterized in that: The device includes: A data synchronization module, used to obtain real-time data and historical data of the pump station, and synchronize the real-time data and historical data to the digital twin platform; A generation module, used to obtain a scheduling strategy generated by the digital twin platform based on the real-time data and historical data; A regulating module, used for regulating the operation mode of the pump station based on the scheduling strategy; Among them, the digital twin platform includes an intelligent control model, which includes an initial correction model, a pump station unit evaluation model, a flow optimization distribution model within the pump station, and a pump station real-time optimization scheduling model connected in sequence; the initial correction model includes a water level prediction model and a pump group characteristic curve correction model.

9. An electronic device, characterized in that: The electronic device comprises a memory and a processor, wherein the memory is used to store a computer program, and the processor runs the computer program to enable the electronic device to execute the method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium stores computer program instructions, and when the computer program instructions are read and executed by a processor, the method according to any one of claims 1 to 7 is executed.

Images