Emergency command and field scheduling system and method based on heat supply network model

Abstract

The invention relates to the technical field of operation state scheduling, and discloses an emergency command and field scheduling system and method based on a heat supply network model. According to the method, firstly, the heat supply pipeline flow, the heat supply pipeline pressure and the heat supply pipeline temperature are calculated, a heat supply network model is established, and the pressure, the temperature and the flow of the heat supply pipeline are commanded, regulated and controlled in real time according to abnormal data; predicting the pressure, the temperature and the flow of the heat supply pipeline by using a BP neural network to obtain prediction data of the heat supply pipeline; heat supply pipeline leakage points are identified and verified through heat supply network model heat supply pipeline prediction data, meanwhile, a badger optimization algorithm is used for optimizing an electric power emergency material conveying path, and the material dispatching efficiency is guaranteed; and finally, a thermal dispatching command platform and a thermal database are established, so that temperature, pressure, flow and leakage point data of the heat supply pipeline are conveniently regulated, controlled and stored, and later query is facilitated.

Description

Technical Field

[0001] The present invention relates to the technical field of operation status scheduling, and in particular to an emergency command and on-site scheduling system and method based on a heat network model. Background Art

[0002] The Chinese invention patent application with publication number CN117350445A discloses an intelligent emergency command system and method based on artificial intelligence. Specifically, it includes: first, obtaining the location where the danger occurs under the management of the power plant monitoring platform, receiving the material dispatch request of the location where the danger occurs, and then obtaining the traffic information of the location where the danger occurs and the location where the danger does not occur under the management of the power plant monitoring platform, generating an emergency material dispatch route corresponding to the traffic information, and marking the emergency material dispatch route; secondly, for the marked emergency material dispatch route, judging and identifying the characteristic distribution of the emergency material dispatch route according to the impact degree of the danger, calculating the characteristic distribution of the emergency material dispatch route to obtain the first characteristic index; according to the characteristic distribution of different dangers occurring on the emergency material dispatch route, calculating the characteristic distribution of different dangers occurring on the emergency material dispatch route to calculate the second characteristic index; according to the first characteristic index and the second characteristic index, the emergency material dispatch route is evaluated for the early warning index, and the early warning index is sent to the emergency command personnel to assist the emergency command personnel to complete the personnel arrangement work for the dredging and maintenance of the emergency material dispatch route in the power plant.

[0003] When an emergency safety accident occurs in a power plant, the accident alarm takes a long time, and the emergency command response of the relevant departments of the power plant is not timely, which will lead to the failure to deal with the emergency safety accident in time, thus causing a chain of accidents. In the traditional on-site dispatch of emergency safety accidents, the relevant departments of the power plant do not plan the material dispatch path, nor do they have a platform suitable for emergency accident dispatch. In addition, they lack the support of high-tech such as big data, and on-site dispatch is prone to untimely dispatch and chaotic dispatch. Summary of the invention

[0004] In view of the problems in the related art, the present invention provides an emergency command and on-site dispatching system and method based on a heating network model to overcome the above-mentioned technical problems existing in the existing related art.

[0005] To solve the above technical problems, the present invention is achieved through the following technical solutions:

[0006] The present invention is an emergency command and on-site dispatching method based on a heating network model, comprising the following steps:

[0007] S1. Obtain the heating data of the power plant in the heating network model, generate the heating pipeline matrix of the power plant, establish a heating network model based on the dynamic characteristics of the heating network, realize real-time regulation of the pressure, temperature and flow of the heating pipeline, and obtain a processed heating network model;

[0008] S2. Obtain the pressure, temperature and flow data of the heating pipeline according to the processed heating network model, record it as the heating pipeline data set, train the BP neural network to obtain a trained BP neural network, and use a genetic algorithm to optimize the trained BP neural network to achieve the prediction of the pressure, temperature and flow of the heating pipeline, and obtain the heating pipeline prediction data;

[0009] S3, according to the processed heating network model and the predicted data of the heating pipeline, the suspected leakage point of the heating pipeline is obtained, the monitoring image is processed by image recognition technology, the suspected leakage point of the heating pipeline is verified to obtain the leakage point of the heating pipeline, if the leakage point of the heating pipeline is found, the transportation route of the emergency power materials is planned, the leakage point of the heating pipeline is repaired, and the transportation route of the emergency power materials is optimized by using the honey badger optimization algorithm;

[0010] S4. Establish a thermal dispatch command platform and thermal database to display and store the temperature, pressure, flow and leakage point data of the heating pipeline in real time.

[0011] The invention first establishes a heating pipeline matrix of a power plant, calculates the flow rate, pressure and temperature of each section of the heating pipeline after numbering the heating pipeline, establishes a heating network model based on the dynamic characteristics of the heating network, and adjusts the pressure, temperature and flow rate of the heating pipeline in real time according to abnormal data, and comprehensively considers various factors to build a model to obtain more accurate results; secondly, the pressure, temperature and flow rate data of the heating pipeline are obtained according to the heating network model, the BP neural network is trained to obtain a trained BP neural network, and the trained BP neural network is optimized using a genetic algorithm. By calculating the variable mutation probability, the population gradually converges to the optimal solution, and the pressure, temperature and flow rate of the heating pipeline are predicted, and the heating pipeline prediction data is obtained. The optimized BP neural network converges faster and the predicted value is closer to Actual value; then the suspected heating pipeline leakage point is obtained from the heating network model and the heat pipeline prediction data, and the suspected heating pipeline leakage point matrix is ​​generated from the monitoring image. First, the gray value is linearly processed to increase the contrast, and then the image is segmented. The pixel gray value is replaced based on the regional growth principle. Finally, the Canny algorithm is used to extract the edge of the heating pipeline leakage point, and the heating pipeline leakage point identification and verification are completed. After locating the heating pipeline leakage point, the power emergency material delivery route is planned; the honey badger optimization algorithm is used to optimize the power emergency material delivery route. The honey badger finds the nectar source according to the strength of the smell and continuously updates the iterative delivery position. The algorithm converges quickly and has strong development capabilities; finally, a thermal dispatch command platform and a thermal database are established to facilitate the regulation and storage of heating pipeline temperature, pressure, flow and leakage point data.

[0012] Preferably, the S1 comprises the following steps:

[0013] S11, obtaining the heating pipeline data of the power plant in the heating network model, numbering the heating pipelines in sequence, and recording them as the heating pipeline set A = {a1, a2, a3, ..., a i}, where a i Represents the heating pipeline numbered i, assuming that each heating pipeline in the heating pipeline set has j pipeline nodes, and generates the power plant heating pipeline matrix B as follows:

[0014]

[0015] Among them, a ij represents the jth pipeline node of the heating pipeline numbered i;

[0016] S12, selecting a row in the power plant heating pipeline matrix as a heating pipeline parameter set in Indicates the number The jth pipeline node of the heating pipeline of the heating pipeline; set the specific friction resistance of the heating pipeline to b, the length of the heating pipeline to c, the equivalent length of resistance to b′, the absolute roughness of the heating pipeline to c′, the density of the heat transfer medium to b″, the diameter of the heating pipeline to c″, the resistance coefficient of the heating pipeline to α, the local resistance coefficient of the heating pipeline to β, and the number is The pressure loss of the heating pipeline is C1, numbered The heating pipeline flow is C2, and the calculation formula is as follows:

[0017]

[0018] Set the i1th pipeline node flow of the heating pipeline to The number is The heating pipeline flow is C2, and the calculation formula is as follows:

[0019]

[0020] Set the upper and lower limits of the water supply temperature in the heating pipe to be d′ max and d′ min The upper and lower limits of the return water temperature of the heating pipe are d′ m ' ax and d′ m ' in , the water supply temperature of the heating pipe is d′ and d′ min <d′<d′ max , the return water temperature of the heating pipe is d″ and d′ m ' in <d″<d′ m ' ax , the specific heat capacity of the heat transfer medium is e', numbered The hot water flow rate of the jth pipe node of the heating pipe is e″, then the number is The calculation formula of the thermal power C′ of the jth pipeline node of the heating pipeline is as follows:

[0021] C′=e′e″·(d′-d″);

[0022] Set the number to The outlet temperature of the heating pipe is d1, numbered The inlet temperature of the heating pipe is d2, numbered The calculation formula of heating pipe temperature C3 is as follows:

[0023]

[0024] S13, the number is The heating pipe pressure and number are The heating pipe flow and number are The temperature of the heating pipeline is obtained, the pressure of the heating network pipeline, the flow rate of the heating network pipeline and the temperature of the heating network pipeline are obtained, and a heating network model is established; the pressure of the heating pipeline to be regulated, the temperature of the heating pipeline to be regulated and the flow rate of the heating pipeline to be regulated are obtained from the heating network model;

[0025] The upper pressure threshold is set to ξ1, the lower pressure threshold is set to ξ2, the upper flow threshold is set to ψ1, the lower flow threshold is set to ψ2, the upper temperature threshold is set to ζ1, and the lower temperature threshold is set to ζ2; when the pressure of the heating pipe to be regulated is less than ξ2, the temperature of the heating pipe to be regulated and the flow of the heating pipe to be regulated are adjusted until the pressure of the heating pipe to be regulated is between ξ1 and ξ2; when the flow of the heating pipe to be regulated is less than ψ2, the pressure of the heating pipe to be regulated and the temperature of the heating pipe to be regulated are increased until the flow of the heating pipe to be regulated is between ψ1 and ψ2; when the temperature of the heating pipe to be regulated is less than ζ2, the pressure of the heating pipe to be regulated and the flow of the heating pipe to be regulated are increased until the temperature of the heating pipe to be regulated is between ζ1 and ζ2; realize real-time regulation of the pressure, temperature and flow of the heating pipe, and obtain a processed heat network model.

[0026] The invention obtains the power plant's heating pipeline data, establishes the power plant's heating pipeline matrix, calculates the heating pipeline flow, heating pipeline pressure and heating pipeline temperature, establishes a heating network model based on the dynamic characteristics of the heating network, and adjusts the pressure, temperature and flow of the heating pipeline in real time according to abnormal data. The modeling takes into account various factors comprehensively to obtain more accurate results.

[0027] Preferably, S2 comprises the following steps:

[0028] S21, the pressure, temperature and flow data of the heating pipeline are obtained according to the processed heating network model, and the heating pipeline data set A2={(a1′,a1″,a1″′),(a′2,a′2′,a′2″),(a3′,a3″,a3″′),...,(a e ′,a e ″,a e ″′)}, where a e ' represents the pressure of the heating pipe numbered e, a e ″ represents the temperature of the heating pipe numbered e, a e ″′ represents the flow rate of the heating pipeline numbered e; the heating pipeline data set is divided into a heating pipeline training set and heating pipe test set in Indicates the number The heating pipe pressure, Indicates the number The heating pipe temperature, Indicates the number The flow rate of heating pipeline;

[0029] Set the number of nodes in the input layer of the BP (multi-layer forward error back propagation) neural network to The number of output layer nodes is 3, the learning rate is x and x=0.7, the hidden layer activation function is the tansig function, the heating pipeline training set is normalized to obtain a normalized heating pipeline training set, the normalized heating pipeline training set is input into the BP neural network, and the network error is set to Continuously iterate, when the BP neural network error is less than When , the iteration stops and the trained BP neural network is obtained; otherwise, S21 is repeated until the BP neural network error is less than

[0030] S22, input the heating pipeline test set into the trained BP neural network, set the current iteration number and the maximum iteration number D, and iterate continuously. When the current iteration number is greater than D, stop iterating to obtain the final BP neural network; otherwise, use the genetic algorithm to optimize the trained BP neural network to obtain the final BP neural network. The specific process is as follows:

[0031] S221, setting the fitness function f of the trained BP neural network to calculate the formula as follows:

[0032]

[0033] in, Indicates the actual value, represents the predicted value of the trained BP neural network,

[0034] Set the population size to δ, use the selection function to select the best individual in the population with a probability of e′, calculate the fitness function value of the individual in the population, and record it as the fitness function set A5 = {e1, e2, e3, ..., e δ}, where e δ Represents the fitness function value of the δth individual; the fitness function set is sorted to obtain a processed fitness function value set, and then the distribution parameter is used to calculate the individual selection probability. The calculation formula is as follows:

[0035]

[0036] Among them, f′ represents the distribution parameter, e″ represents the individual selection probability;

[0037] The population is cross-operated. The process of cross-operation is to optimize the weights and biases of the trained BP neural network. Two individuals D1 and D2 are randomly selected from the population as parents, and a random value f″ in the range of [0,1] is generated for the parents. The random value is used to perform interpolation operation on the parents to generate two offspring individuals D1′ and D2′. The calculation formula is as follows:

[0038] D1′=D1f″+D2(1-f″), D2′=D1(1-f″)+D2f″;

[0039] The population is mutated. The process of population mutation is the process of narrowing the local search range of the trained BP neural network. The current generation of the population is set to D′, the maximum generation of the population is set to D″, the population selects the mutation direction and starts mutation, and the mutated individuals are obtained. The mutation probability e″′ is calculated as follows:

[0040]

[0041] S222, the population selects the best individual of the first iteration with the probability of the best individual, and the trained BP neural network performs the first iteration operation at this time, and adjusts the weight and bias of the trained BP neural network after the best individual of the first iteration is cross-operated; the best individual of the second iteration is selected from the best individual of the first iteration, and the trained BP neural network performs the second iteration operation; S222 is repeated until the current iteration number D′ of the trained BP neural network is greater than D″, and the iteration is stopped to obtain the final BP neural network;

[0042] S23, inputting the historical heating pipeline pressure, temperature and flow data as the historical heating pipeline data set into the final BP neural network, and the final BP neural network outputs the predicted values ​​of the heating pipeline pressure, temperature and flow to obtain the heating pipeline prediction data.

[0043] The invention obtains the pressure, temperature and flow data of the heating pipeline through the heat network model, recorded as the heating pipeline data set, trains the BP neural network to obtain the trained BP neural network, and uses a genetic algorithm to optimize the trained BP neural network. By realizing a variable mutation probability, the population gradually converges to the optimal solution, and the pressure, temperature and flow of the heating pipeline are predicted, and the heating pipeline prediction data is obtained. The optimized BP neural network converges faster, and the predicted value is closer to the actual value, achieving a good effect.

[0044] Preferably, S3 comprises the following steps:

[0045] S31, calculate the first suspected heating pipe leakage point according to the processed heating network model, and then obtain the second suspected heating pipe leakage point according to the heating pipe prediction data, and record the suspected heating pipe leakage point where the first suspected heating pipe leakage point and the second suspected heating pipe leakage point overlap as the suspected heating pipe leakage point; obtain the monitoring image of the suspected heating pipe leakage point, set the horizontal dimension of the pixel point of the monitoring image of the suspected heating pipe leakage point to i', and the vertical dimension of the pixel point of the monitoring image of the suspected heating pipe leakage point to j', and generate the suspected heating pipe leakage point matrix B' as follows,

[0046]

[0047] Among them, a′ i′j′ Represents the pixel points of the horizontal i′ dimension and the vertical j′ dimension in the matrix of suspected heating pipeline leakage points;

[0048] S32, perform grayscale linear processing on the matrix of suspected heating pipeline leakage points, set the grayscale transformation slope to g, the grayscale value to h, and the grayscale transformation range to Gray value conversion function Calculate and obtain the grayscale value after the grayscale value is linearly processed. When the grayscale value of the pixel in the suspected heating pipeline leakage point matrix is ​​less than the grayscale value after the grayscale value is linearly processed, use the grayscale value after the grayscale value is linearly processed to replace the grayscale value of the pixel in the suspected heating pipeline leakage point matrix; otherwise, retain the grayscale value of the pixel in the suspected heating pipeline leakage point matrix to obtain the processed suspected heating pipeline leakage point matrix;

[0049] S33, performing image segmentation on the pixel points in the processed suspected heating pipe leakage point matrix, calculating the absolute value of the difference between the grayscale value of the pixel point in the processed suspected heating pipe leakage point matrix and the grayscale value of the surrounding pixel points, setting the fourth threshold value to ω4, and when the absolute value of the difference between the grayscale value of the pixel point in the processed suspected heating pipe leakage point matrix and the grayscale value of the surrounding pixel points is less than ω4, the corresponding surrounding pixel points are recorded as seed pixel points;

[0050] Calculate the absolute value of the difference between the grayscale value of the seed pixel and the grayscale value of the pixels around the seed pixel, and then calculate the average of the grayscale value of the seed pixel and the grayscale value of the pixels around the seed pixel, which is recorded as Set the gray value of the seed pixel and the gray value of the pixels around the seed pixel as h′, and the gray value threshold ξ is calculated as follows:

[0051]

[0052] When the absolute value of the difference between the grayscale value of the seed pixel and the grayscale value of the pixels around the seed pixel is less than ξ, the pixels around the corresponding seed pixel are recorded as new seed pixels; otherwise, repeat until S33 to obtain a new seed pixel; the first cumulative grayscale histogram h″ is obtained from the grayscale values ​​of the pixels in the suspected heating pipe leakage point matrix after processing and the grayscale values ​​of the surrounding pixels, and the first cumulative grayscale histogram h″′ is obtained from the grayscale value of the seed pixel and the grayscale value of the pixels around the seed pixel, and the grayscale histogram threshold is set to ψ. When max|h″-h″′| is less than ψ, stop image segmentation; otherwise, continue image segmentation until max|h″-h″′| is less than ψ, and obtain the image segmented suspected heating pipe leakage point matrix;

[0053] S34, use the Canny algorithm to extract edges, establish a rectangular coordinate system XOY for the suspected heating pipe leakage point matrix of the image segmentation, the coordinate origin O is the lower left corner pixel of the suspected heating pipe leakage point matrix of the image segmentation, calculate the partial derivatives of the suspected heating pipe leakage point matrix of the image segmentation on the X-axis and Y-axis to obtain the partial derivative matrix X′ and the partial derivative matrix Y′, then calculate the X-axis gradient amplitude and X-axis gradient direction of the partial derivative matrix X′, calculate the Y-axis gradient amplitude and Y-axis gradient direction of the partial derivative matrix Y′, and obtain the image The gradient amplitude and gradient direction of the segmented suspected heating pipe leakage point matrix; the image segmented suspected heating pipe leakage point matrix is ​​compared with the grayscale value and gradient amplitude of the pixel points of the image segmented suspected heating pipe leakage point matrix along the gradient direction, when the grayscale value of the pixel point of the image segmented suspected heating pipe leakage point matrix is ​​greater than the gradient amplitude, the corresponding image segmented suspected heating pipe leakage point matrix pixel point is retained, otherwise the corresponding image segmented suspected heating pipe leakage point matrix pixel point grayscale value is set to 0, and the processed suspected heating pipe leakage point matrix is ​​obtained;

[0054] S35, locating the leakage area by the processed suspected heating pipe leakage point matrix, acquiring the next frame image of the monitoring image of the suspected heating pipe leakage point, obtaining the processed suspected heating pipe leakage point matrix of the next frame image, checking whether the leakage area exists, if the leakage area exists, the suspected heating pipe leakage point is the heating pipe leakage point; otherwise, the suspected heating pipe leakage point is not the heating pipe leakage point;

[0055] After the leakage point of the heating pipeline is found, the location of the leakage point of the heating pipeline is located according to the monitoring position, and the transportation path of the emergency power materials is planned. The leakage point of the heating pipeline is repaired, and the transportation path of the emergency power materials is optimized using the Honey Badger optimization algorithm. The specific steps are as follows:

[0056] S351, set the starting point of the power emergency material transportation route to the initial position of the honey badger, the transportation cost is k1, the terrain cost is k2, the boundary cost is k3, and the fitness function of the honey badger optimization algorithm is F=min(k1+k2+k3); the honey badger collects honey, set the random number k′ and k′∈[0,1], the upper bound of the honey badger activity area is l1, and the lower bound of the honey badger activity area is l2, then the calculation formula of the honey badger initial position x1 is as follows,

[0057] x1=l2+k′(l1-l2)

[0058] When the honey badger searches for nectar, the nectar emits fragrance to generate a fragrance density factor. A random number k″ is set, and k″∈[0,1]. The intensity between the honey badger and the nectar source is m, and the distance between the honey badger and the nectar source is n. The calculation formula for the honey badger's olfactory intensity G is as follows:

[0059]

[0060] As time t increases during the honey badger's search for nectar, the fragrance density factor changes continuously, and the power emergency material delivery path changes direction continuously. The fragrance density factor calculation formula is as follows:

[0061]

[0062] Where F′ represents the fragrance density factor, k″′ represents a random number and k″′≥1, t max Indicates the maximum number of iterations;

[0063] S352, determine the direction of the maximum fragrance density factor, which is the direction of the power emergency material transportation path, recorded as m'. During the process of honey badger digging for nectar sources, the honey badger changes its position to obtain a new position. According to the direction of the power emergency material transportation path, the first transportation position x2 is obtained. The calculation formula is as follows:

[0064]

[0065] Among them, x′ represents the location of the nectar source, represents the predation ability of the honey badger, p1, p2 and p3 represent random numbers between 0 and 1;

[0066] The honey badger continues to search for honey sources and constantly updates the direction of the power emergency material delivery route to obtain a new delivery position, and obtains the second delivery position x3. The calculation formula is as follows:

[0067] x3=x′+m′·F′·p4·n

[0068] Among them, p4 represents a random number between 0 and 1;

[0069] The honey badger continuously searches for honey sources, continuously updates the delivery location, and updates the fitness function value under the power emergency material delivery path. It continuously iterates according to the fitness function value under the power emergency material delivery path. The current number of iterations is set to time t. When the current number of iterations is greater than t max , stop the iteration, and the path of the honey badger looking for the nectar source is the optimized power emergency material transportation path; otherwise, repeat S251 and S252 until the current number of iterations is greater than t max .

[0070] The invention generates a matrix of suspected heating pipe leakage points through monitoring images, first performs linear processing of grayscale values ​​to increase contrast, and then uses the Canny algorithm to extract the edges of the heating pipe leakage points after image segmentation and region growth principle processing, completes the identification and verification of the heating pipe leakage points, and starts planning the emergency power material delivery route after locating the heating pipe leakage points; uses the honey badger optimization algorithm to optimize the emergency power material delivery route, and uses the honey badger to find the nectar source according to the strength of the smell and continuously updates the iterative delivery position. The algorithm converges quickly and has strong development capabilities.

[0071] Preferably, S4 comprises the following steps:

[0072] S41. Establish a heat dispatching command platform to display the temperature, pressure and flow data of the heating pipeline in real time. When the temperature, pressure and flow data of the heating pipeline are abnormal, adjust the temperature, pressure and flow data of the heating pipeline in real time; when a leakage point of the heating pipeline is found, locate the position of the leakage point of the heating pipeline, and inspect and repair the leakage point of the heating pipeline according to the optimized power emergency material transportation route;

[0073] S42. Establish a thermal database to store past heating pipeline temperatures, pressures, flows, heating pipeline prediction data, and leakage point data.

[0074] This embodiment also discloses an emergency command and on-site dispatching system based on a heat network model, which specifically includes: a heat network model building module, a heat supply pipeline data control module, a heat supply pipeline data prediction module, a heat pipeline leakage point positioning module and an electric power emergency material transportation path optimization module;

[0075] The heat network model building module is used to build a pressure, flow and temperature model of the heating pipeline;

[0076] The heating pipeline data control module is used to use the model to calculate the pressure, flow and temperature of the heating pipeline and to perform real-time control on abnormal situations;

[0077] The heating pipeline data prediction module is used to train the BP neural network and use the BP neural network to predict the heating pipeline data;

[0078] The heating pipeline leakage point positioning module is used to detect leakage points on the monitoring image using image processing to verify the location of the suspected heating pipeline leakage point;

[0079] The power emergency material delivery path optimization module is used to optimize the power emergency material delivery path using the honey badger optimization algorithm.

[0080] The present invention has the following beneficial effects:

[0081] 1. The invention obtains the heating pipeline data of the power plant, establishes the heating pipeline matrix of the power plant, calculates the flow rate, pressure and temperature of the heating pipeline, establishes a heating network model based on the dynamic characteristics of the heating network, and adjusts the pressure, temperature and flow rate of the heating pipeline in real time according to abnormal data. The model is built by comprehensively considering various factors to obtain more accurate results.

[0082] 2. The invention obtains the pressure, temperature and flow data of the heating pipeline through the heat network model, trains the BP neural network to obtain a trained BP neural network, and uses a genetic algorithm to optimize the trained BP neural network. By realizing a variable mutation probability, the population gradually converges to the optimal solution, thereby realizing the prediction of the pressure, temperature and flow of the heating pipeline. The optimized BP neural network converges faster and the predicted value is closer to the actual value.

[0083] 3. The invention generates a matrix of suspected heating pipe leakage points through monitoring images, first performs linear processing of grayscale values, and then uses the Canny algorithm to extract the edges of the heating pipe leakage points after image segmentation and region growth principle processing, completes the identification and verification of the heating pipe leakage points, and uses the honey badger optimization algorithm to optimize the power emergency material delivery path after locating the heating pipe leakage point. The honey badger finds the nectar source according to the strength of the smell and continuously updates the iterative delivery position. The algorithm converges quickly and has strong development capabilities.

[0084] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the drawings required for describing the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, they can also obtain drawings based on these drawings without paying creative work.

[0086] Figure 1 The present invention provides a flow chart of emergency command and on-site dispatching by an emergency command and on-site dispatching system based on a heat network model. DETAILED DESCRIPTION

[0087] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0088] In the description of the present invention, it is necessary to understand that the terms "opening", "upper", "lower", "top", "middle", "inside" and the like indicating orientation or positional relationship are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the components or elements referred to must have a specific orientation, be constructed and operate in a specific orientation, and therefore cannot be understood as limiting the invention.

[0089] This embodiment discloses an emergency command and on-site dispatch method based on a heat network model, which specifically includes the following contents:

[0090] S1. Obtain the heating data of the power plant in the heating network model, generate the heating pipeline matrix of the power plant, establish a heating network model based on the dynamic characteristics of the heating network, realize real-time regulation of the pressure, temperature and flow of the heating pipeline, and obtain a processed heating network model;

[0091] The S1 comprises the following steps:

[0092] S11, obtaining the heating pipeline data of the power plant in the heating network model, numbering the heating pipelines in sequence, and recording them as the heating pipeline set A = {a1, a2, a3, ..., a i}, where a i Represents the heating pipeline numbered i, assuming that each heating pipeline in the heating pipeline set has j pipeline nodes, and generates the power plant heating pipeline matrix B as follows:

[0093]

[0094] Among them, a ij represents the jth pipeline node of the heating pipeline numbered i;

[0095] S12, selecting a row in the power plant heating pipeline matrix as a heating pipeline parameter set in Indicates the number The jth pipeline node of the heating pipeline of the heating pipeline; set the specific friction resistance of the heating pipeline to b, the length of the heating pipeline to c, the equivalent length of resistance to b′, the absolute roughness of the heating pipeline to c′, the density of the heat transfer medium to b″, the diameter of the heating pipeline to c″, the resistance coefficient of the heating pipeline to α, the local resistance coefficient of the heating pipeline to β, and the number is The pressure loss of the heating pipeline is C1, numbered The heating pipeline flow is C2, and the calculation formula is as follows:

[0096]

[0097] Set the i1th pipeline node flow of the heating pipeline to The number is The heating pipeline flow is C2, and the calculation formula is as follows:

[0098]

[0099] Set the upper and lower limits of the water supply temperature in the heating pipe to be d′ max and d′ min The upper and lower limits of the return water temperature of the heating pipe are d′ m ' ax and d′ m ' in , the water supply temperature of the heating pipe is d′ and d′ min <d′<d′ max , the return water temperature of the heating pipe is d″ and d′ m ' in <d″<d′ m ' ax , the specific heat capacity of the heat transfer medium is e', numbered The hot water flow rate of the jth pipe node of the heating pipe is e″, then the number is The calculation formula of the thermal power C′ of the jth pipeline node of the heating pipeline is as follows:

[0100] C′=e′e″·(d′-d″);

[0101] Set the number to The outlet temperature of the heating pipe is d1, numbered The inlet temperature of the heating pipe is d2, numbered The calculation formula of heating pipe temperature C3 is as follows:

[0102]

[0103] S13, based on the pressure of the heating pipeline numbered i to i, the flow rate of the heating pipeline numbered i to i and the heating pipeline numbered i to i, The temperature of the heating pipeline is obtained, the pressure of the heating network pipeline, the flow rate of the heating network pipeline and the temperature of the heating network pipeline are obtained, and a heating network model is established; the pressure of the heating pipeline to be regulated, the temperature of the heating pipeline to be regulated and the flow rate of the heating pipeline to be regulated are obtained from the heating network model;

[0104] Set the upper pressure threshold to ξ1, the lower pressure threshold to ξ2, the upper flow threshold to ψ1, the lower flow threshold to ψ2, the upper temperature threshold to ζ1, and the lower temperature threshold to ζ2; when the pressure of the heating pipeline to be regulated is less than ξ2, the temperature of the heating pipeline to be regulated and the flow of the heating pipeline to be regulated are adjusted until the pressure of the heating pipeline to be regulated is between ξ1 and ξ2; when the flow of the heating pipeline to be regulated is less than ψ2, increase the pressure of the heating pipeline to be regulated and the temperature of the heating pipeline to be regulated until the flow of the heating pipeline to be regulated is between ψ1 and ψ2; when the temperature of the heating pipeline to be regulated is less than ζ2, increase the pressure of the heating pipeline to be regulated and the flow of the heating pipeline to be regulated until the temperature of the heating pipeline to be regulated is between ζ1 and ζ2; realize real-time regulation of the pressure, temperature and flow of the heating pipeline, and obtain a processed heat network model;

[0105] S2. Obtain the pressure, temperature and flow data of the heating pipeline from the processed heating network model, record them as the heating pipeline data set, train the BP neural network to obtain a trained BP neural network, and use a genetic algorithm to optimize the trained BP neural network to predict the pressure, temperature and flow of the heating pipeline, and obtain the heating pipeline prediction data;

[0106] The S2 comprises the following steps:

[0107] S21. The processed heat network model obtains the pressure, temperature and flow data of the heating pipeline, and obtains the heating pipeline data set A2={(a1′,a1″,a1″′),(a′2,a′2′,a′2″),(a3′,a3″,a3″′),...,(a e ′,a e ″,a e ″′)}, where a e ' represents the pressure of the heating pipe numbered e, a e ″ represents the temperature of the heating pipe numbered e, a e ″′ represents the flow rate of the heating pipeline numbered e; the heating pipeline data set is divided into a heating pipeline training set and heating pipe test set in Indicates the number The heating pipe pressure, Indicates the number The heating pipe temperature, Indicates the number The flow rate of heating pipeline;

[0108] Set the number of nodes in the BP neural network input layer to The number of output layer nodes is 3, the learning rate is x and x=0.7, the hidden layer activation function is the tansig function, the heating pipeline training set is normalized to obtain a normalized heating pipeline training set, the normalized heating pipeline training set is input into the BP neural network, and the network error is set to Continuously iterate, when the BP neural network error is less than When , the iteration stops and the trained BP neural network is obtained; otherwise, S21 is repeated until the BP neural network error is less than

[0109] S22, input the heating pipeline test set into the trained BP neural network, set the current iteration number and the maximum iteration number D, and iterate continuously. When the current iteration number is greater than D, stop iterating to obtain the final BP neural network; otherwise, use the genetic algorithm to optimize the trained BP neural network to obtain the final BP neural network. The specific process is as follows:

[0110] S221, setting the fitness function f of the trained BP neural network to calculate the formula as follows:

[0111]

[0112] in, Indicates the actual value, represents the predicted value of the trained BP neural network,

[0113] Set the population size to δ, use the selection function to select the best individual in the population with a probability of e′, calculate the fitness function value of the individual in the population, and record it as the fitness function set A5 = {e1, e2, e3, ..., e δ}, where e δ Represents the fitness function value of the δth individual; the fitness function set is sorted to obtain a processed fitness function value set, and then the distribution parameter is used to calculate the individual selection probability. The calculation formula is as follows:

[0114]

[0115] Among them, f′ represents the distribution parameter, e″ represents the individual selection probability;

[0116] The population is crossovered, and the process of crossover is to optimize the weights and biases of the trained BP neural network; two individuals D1 and D2 are randomly selected from the population as parents, and a random value f″ in the range of [0,1] is generated for the parents. The random value is used to perform interpolation operation on the parents to generate two offspring individuals D1′ and D2′. The calculation formula is as follows:

[0117] D1′=D1f″+D2(1-f″), D2′=D1(1-f″)+D2f″;

[0118] The population performs mutation operation, which is the process of narrowing the local search range of the trained BP neural network; the current generation of the population is set to D′, the maximum generation of the population is set to D″, the population selects the mutation direction and starts to mutate, and the mutated individuals are obtained. The mutation probability e″′ is calculated as follows:

[0119]

[0120] S222, the population selects the best individual of the first iteration with the probability of the best individual, and the trained BP neural network performs the first iteration operation at this time, and adjusts the weight and bias of the trained BP neural network after the best individual of the first iteration is cross-operated; the best individual of the second iteration is selected from the best individual of the first iteration, and the trained BP neural network performs the second iteration operation; S222 is repeated until the current iteration number D′ of the trained BP neural network is greater than D″, and the iteration is stopped to obtain the final BP neural network;

[0121] S23, inputting the historical pressure, temperature and flow data of the heating pipeline as the historical heating pipeline data set into the final BP neural network, and the final BP neural network outputs the predicted values ​​of the pressure, temperature and flow of the heating pipeline to obtain the heating pipeline prediction data;

[0122] S3, according to the processed heating network model and the predicted data of the heating pipeline, the suspected leakage point of the heating pipeline is obtained, the monitoring image is processed by image recognition technology, the suspected leakage point of the heating pipeline is verified to obtain the leakage point of the heating pipeline, if the leakage point of the heating pipeline is found, the transportation route of the emergency power materials is planned, the leakage point of the heating pipeline is repaired, and the transportation route of the emergency power materials is optimized by using the honey badger optimization algorithm;

[0123] The S3 comprises the following steps:

[0124] S31, calculate the first suspected heating pipe leakage point by the processed heating network model, and then obtain the second suspected heating pipe leakage point by the heating pipe prediction data, and record the suspected heating pipe leakage point where the first suspected heating pipe leakage point and the second suspected heating pipe leakage point overlap as the suspected heating pipe leakage point; obtain the monitoring image of the suspected heating pipe leakage point, set the horizontal dimension of the pixel point of the monitoring image of the suspected heating pipe leakage point to i', and the vertical dimension of the pixel point of the monitoring image of the suspected heating pipe leakage point to j', and generate the suspected heating pipe leakage point matrix B' as follows,

[0125]

[0126] in, Represents the pixel points of the horizontal i′ dimension and the vertical j′ dimension in the matrix of suspected heating pipeline leakage points;

[0127] S32, perform grayscale linear processing on the matrix of suspected heating pipeline leakage points, set the grayscale transformation slope to g, the grayscale value to h, and the grayscale transformation range to Gray value conversion function Calculate and obtain the grayscale value after the grayscale value is linearly processed. When the grayscale value of the pixel in the suspected heating pipeline leakage point matrix is ​​less than the grayscale value after the grayscale value is linearly processed, use the grayscale value after the grayscale value is linearly processed to replace the grayscale value of the pixel in the suspected heating pipeline leakage point matrix; otherwise, retain the grayscale value of the pixel in the suspected heating pipeline leakage point matrix to obtain the processed suspected heating pipeline leakage point matrix;

[0128] S33, performing image segmentation on the pixel points in the processed suspected heating pipe leakage point matrix, calculating the absolute value of the difference between the grayscale value of the pixel point in the processed suspected heating pipe leakage point matrix and the grayscale value of the surrounding pixel points, setting the fourth threshold value to ω4, and when the absolute value of the difference between the grayscale value of the pixel point in the processed suspected heating pipe leakage point matrix and the grayscale value of the surrounding pixel points is less than ω4, the corresponding surrounding pixel points are recorded as seed pixel points;

[0129] Calculate the absolute value of the difference between the grayscale value of the seed pixel and the grayscale value of the pixels around the seed pixel, and then calculate the average of the grayscale value of the seed pixel and the grayscale value of the pixels around the seed pixel, which is recorded as Set the gray value of the seed pixel and the gray value of the pixels around the seed pixel as h′, and the gray value threshold ξ is calculated as follows:

[0130]

[0131] When the absolute value of the difference between the grayscale value of the seed pixel and the grayscale value of the pixels around the seed pixel is less than ξ, the pixels around the corresponding seed pixel are recorded as new seed pixels; otherwise, repeat until S33 to obtain a new seed pixel; the first cumulative grayscale histogram h″ is obtained from the grayscale values ​​of the pixels in the suspected heating pipe leakage point matrix after processing and the grayscale values ​​of the surrounding pixels, and the first cumulative grayscale histogram h″′ is obtained from the grayscale value of the seed pixel and the grayscale value of the pixels around the seed pixel, and the grayscale histogram threshold is set to ψ. When max|h″-h″′| is less than ψ, stop image segmentation; otherwise, continue image segmentation until max|h″-h″′| is less than ψ, and obtain the image segmented suspected heating pipe leakage point matrix;

[0132] S34, use the Canny algorithm to extract edges, establish a rectangular coordinate system XOY for the suspected heating pipe leakage point matrix of the image segmentation, the coordinate origin O is the lower left corner pixel of the suspected heating pipe leakage point matrix of the image segmentation, calculate the partial derivatives of the suspected heating pipe leakage point matrix of the image segmentation on the X-axis and Y-axis to obtain the partial derivative matrix X′ and the partial derivative matrix Y′, then calculate the X-axis gradient amplitude and X-axis gradient direction of the partial derivative matrix X′, calculate the Y-axis gradient amplitude and Y-axis gradient direction of the partial derivative matrix Y′, and obtain the image segmentation. The gradient amplitude and gradient direction of the suspected heating pipe leakage point matrix of the image segmentation are compared along the gradient direction with the grayscale value and gradient amplitude of the pixel points of the suspected heating pipe leakage point matrix of the image segmentation. When the grayscale value of the pixel point of the suspected heating pipe leakage point matrix of the image segmentation is greater than the gradient amplitude, the corresponding pixel point of the suspected heating pipe leakage point matrix of the image segmentation is retained, otherwise the grayscale value of the pixel point of the suspected heating pipe leakage point matrix of the image segmentation is set to 0, so as to obtain the processed suspected heating pipe leakage point matrix;

[0133] S35, locating the leakage area by the processed suspected heating pipe leakage point matrix, acquiring the next frame image of the monitoring image of the suspected heating pipe leakage point, obtaining the processed suspected heating pipe leakage point matrix of the next frame image, checking whether the leakage area exists, if the leakage area exists, the suspected heating pipe leakage point is the heating pipe leakage point; otherwise, the suspected heating pipe leakage point is not the heating pipe leakage point;

[0134] After the leakage point of the heating pipeline is found, the location of the leakage point of the heating pipeline is located according to the monitoring position, and the transportation path of the emergency power materials is planned. The leakage point of the heating pipeline is repaired, and the transportation path of the emergency power materials is optimized using the Honey Badger optimization algorithm. The specific steps are as follows:

[0135] S351, set the starting point of the power emergency material transportation route to the initial position of the honey badger, the transportation cost is k1, the terrain cost is k2, the boundary cost is k3, and the fitness function of the honey badger optimization algorithm is F=min(k1+k2+k3); the honey badger collects honey, set the random number k′ and k′∈[0,1], the upper bound of the honey badger activity area is l1, and the lower bound of the honey badger activity area is l2, then the calculation formula of the honey badger initial position x1 is as follows,

[0136] x1=l2+k′(l1-l2)

[0137] When the honey badger searches for nectar, the nectar emits fragrance to generate a fragrance density factor. A random number k″ is set, and k″∈[0,1]. The intensity between the honey badger and the nectar source is m, and the distance between the honey badger and the nectar source is n. The calculation formula for the honey badger's olfactory intensity G is as follows:

[0138]

[0139] As time t increases during the honey badger's search for nectar, the fragrance density factor changes continuously, and the power emergency material delivery path changes direction continuously. The fragrance density factor calculation formula is as follows:

[0140]

[0141] Where F′ represents the fragrance density factor, k″′ represents a random number and k″′≥1, t max Indicates the maximum number of iterations;

[0142] S352, determine the direction of the maximum fragrance density factor, which is the direction of the power emergency material transportation path, recorded as m'. During the process of honey badger digging for nectar sources, the honey badger changes its position to obtain a new position. According to the direction of the power emergency material transportation path, the first transportation position x2 is obtained. The calculation formula is as follows:

[0143]

[0144] Among them, x′ represents the location of the nectar source, represents the predation ability of the honey badger, p1, p2 and p3 represent random numbers between 0 and 1;

[0145] The honey badger continues to search for honey sources and constantly updates the direction of the power emergency material delivery route to obtain a new delivery position, and obtains the second delivery position x3. The calculation formula is as follows:

[0146] x3=x′+m′·F′·p4·n

[0147] Among them, p4 represents a random number between 0 and 1;

[0148] The honey badger continuously searches for honey sources, continuously updates the delivery location, and updates the fitness function value under the power emergency material delivery path. It continuously iterates according to the fitness function value under the power emergency material delivery path. The current number of iterations is set to time t. When the current number of iterations is greater than t max , stop the iteration, and the path of the honey badger looking for the nectar source is the optimized power emergency material transportation path; otherwise, repeat S251 and S252 until the current number of iterations is greater than t max ;

[0149] S4. Establish a thermal dispatch command platform and thermal database to display and store the temperature, pressure, flow and leakage point data of the heating pipeline in real time;

[0150] The S4 comprises the following steps:

[0151] S41. Establish a heat dispatching command platform to display the temperature, pressure and flow data of the heating pipeline in real time. When the temperature, pressure and flow data of the heating pipeline are abnormal, adjust the temperature, pressure and flow data of the heating pipeline in real time; when a leakage point of the heating pipeline is found, locate the position of the leakage point of the heating pipeline, and inspect and repair the leakage point of the heating pipeline according to the optimized power emergency material transportation route;

[0152] S42. Establish a thermal database to store past heating pipeline temperatures, pressures, flows, heating pipeline prediction data, and leakage point data.

[0153] This embodiment also discloses an emergency command and on-site dispatching system based on a heat network model, which specifically includes: a heat network model building module, a heat supply pipeline data control module, a heat supply pipeline data prediction module, a heat pipeline leakage point positioning module and an electric power emergency material transportation path optimization module;

[0154] The heat network model building module is used to build a pressure, flow and temperature model of the heating pipeline;

[0155] The heating pipeline data control module is used to use the model to calculate the pressure, flow and temperature of the heating pipeline and to perform real-time control on abnormal situations;

[0156] The heating pipeline data prediction module is used to train the BP neural network and use the BP neural network to predict the heating pipeline data;

[0157] The heating pipeline leakage point positioning module is used to detect leakage points on the monitoring image using image processing to verify the location of the suspected heating pipeline leakage point;

[0158] The power emergency material delivery path optimization module is used to optimize the power emergency material delivery path using the honey badger optimization algorithm.

[0159] In the description of this specification, the description with reference to the terms "one embodiment", "example", "specific example", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner.

[0160] The preferred embodiments of the invention disclosed above are only used to help explain the invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the invention, so that those skilled in the art can understand and use the invention well.

Claims

1. An emergency command and on-site dispatching method based on a heating network model, characterized in that: The steps include: S1. Obtain the heating data of the power plant in the heating network model, generate the heating pipeline matrix of the power plant, establish a heating network model based on the dynamic characteristics of the heating network, realize real-time regulation of the pressure, temperature and flow of the heating pipeline, and obtain a processed heating network model; S2. According to the processed heat network model, the pressure, temperature and flow data of the heating pipeline are obtained, recorded as the heating pipeline data set, the BP neural network is trained to obtain a trained BP neural network, and the trained BP neural network is optimized to realize the prediction of the pressure, temperature and flow of the heating pipeline, and obtain the heating pipeline prediction data; S3. Obtain suspected heating pipeline leakage points according to the processed heating network model and heating pipeline prediction data, process monitoring images by image recognition technology, verify suspected heating pipeline leakage points to obtain heating pipeline leakage points, and if heating pipeline leakage points are found, plan emergency power material transportation routes, inspect and repair heating pipeline leakage points, and optimize emergency power material transportation routes; S4. Establish a thermal dispatch command platform and thermal database to display and store the temperature, pressure, flow and leakage point data of the heating pipeline in real time.

2. The method for emergency command and on-site dispatching based on a heating network model according to claim 1 is characterized in that: The S1 comprises the following steps: S11, obtaining the heating pipeline data of the power plant in the heating network model, and generating a heating pipeline matrix of the power plant; S12. Select a row in the heating pipeline matrix of the power plant as a set of heating pipeline parameters, calculate the flow, temperature and pressure of the heating pipeline, and establish a heating network model; obtain the pressure of the heating pipeline to be regulated, the temperature of the heating pipeline to be regulated and the flow of the heating pipeline to be regulated from the heating network model, and regulate the pressure of the heating pipeline to be regulated, the temperature of the heating pipeline to be regulated and the flow of the heating pipeline to be regulated in real time to obtain a processed heating network model.

3. The method for emergency command and on-site dispatching based on a heating network model according to claim 2 is characterized in that: The S2 comprises the following steps: S21, according to the processed heat network model, obtain the pressure, temperature and flow data of the heating pipeline, obtain the heating pipeline data set, and divide the heating pipeline data set into a heating pipeline training set and a heating pipeline test set; set the number of nodes in the BP neural network input layer to The number of output layer nodes is 3, the learning rate is x and x=0.7, the hidden layer activation function is the tansig function, the heating pipeline training set is normalized to obtain a normalized heating pipeline training set, the normalized heating pipeline training set is input into the BP neural network, and the network error is set to Continuously iterate, when the BP neural network error is less than When , the iteration stops and the trained BP neural network is obtained; otherwise, S21 is repeated until the BP neural network error is less than S22, input the heating pipeline test set into the trained BP neural network, set the current iteration number and the maximum iteration number D, and iterate continuously. When the current iteration number is greater than D, stop iterating to obtain the final BP neural network; otherwise, use the genetic algorithm to optimize the trained BP neural network to obtain the final BP neural network; S23, inputting the historical heating pipeline pressure, temperature and flow data as the historical heating pipeline data set into the final BP neural network, and the final BP neural network outputs the predicted values ​​of the heating pipeline pressure, temperature and flow to obtain the heating pipeline prediction data.

4. The method for emergency command and on-site dispatching based on a heating network model according to claim 3 is characterized in that: The S22 comprises the following steps: S221, setting the fitness function f of the trained BP neural network to calculate the formula as follows: in, Indicates the actual value, represents the predicted value of the trained BP neural network, Set the population size to δ, use the selection function to select the best individual in the population with a probability of e′, calculate the fitness function value of the individuals in the population, sort the fitness function values ​​of the individuals in the population to get the processed fitness function value set, and then use the distribution parameter to calculate the individual selection probability. The calculation formula is as follows: Among them, f′ represents the distribution parameter, e″ represents the individual selection probability; The population is cross-operated. The process of cross-operation is the process of optimizing the weights and biases of the trained BP neural network. Two individuals D1 and D2 are randomly selected from the population as parents, and a random value f″ in the range of [0,1] is generated for the parents. The random value is used to perform interpolation operation on the parents to generate two offspring individuals D1′ and D2′. The calculation formula is as follows: D1′=D1f″+D2(1-f″), D2′=D1(1-f″)+D2f″; The population is mutated. The population mutation process is the process of narrowing the local search range of the trained BP neural network. The current generation of the population is set to D′, the maximum generation of the population is set to D″, the population selects the mutation direction and starts to mutate, and the mutated individuals are obtained. The mutation probability e″′ is calculated as follows:

5. The method for emergency command and on-site dispatching based on a heating network model according to claim 4 is characterized in that: The population selects the best individual of the first iteration with the probability of the best individual, and the trained BP neural network performs the first iteration operation at this time. After the best individual of the first iteration is subjected to a crossover operation, the weight and bias of the trained BP neural network are adjusted; the best individual of the second iteration is selected from the best individual of the first iteration, and the trained BP neural network performs the second iteration operation; S222 is continuously repeated until the current iteration number D′ of the trained BP neural network is greater than D″, and the iteration is stopped to obtain the final BP neural network.

6. The method for emergency command and on-site dispatching based on a heating network model according to claim 3 is characterized in that: The S3 comprises the following steps: S31, obtaining a suspected heating pipeline leakage point from the processed heating network model and heating pipeline prediction data; obtaining a monitoring image of the suspected heating pipeline leakage point, processing the monitoring image by image recognition technology, verifying the suspected heating pipeline leakage point to obtain the heating pipeline leakage point; S32. After a leakage point in the heating pipeline is found, the position of the leakage point in the heating pipeline is located according to the monitoring position, and an emergency power material transportation route is planned. The leakage point in the heating pipeline is repaired, and the emergency power material transportation route is optimized using the Honey Badger optimization algorithm.

7. The method for emergency command and on-site dispatching based on a heating network model according to claim 6 is characterized in that: The S32 comprises the following steps: S321. Set the starting point of the power emergency material transportation route as the initial position of the honey badger, the transportation cost is k1, the terrain cost is k2, the boundary cost is k3, and the fitness function of the honey badger optimization algorithm is F=min(k1+k2+k3); the honey badger collects honey, set the random number k′ and k′∈[0,1], the upper bound of the honey badger activity area is l1, and the lower bound of the honey badger activity area is l2, then the calculation formula of the honey badger initial position x1 is as follows, x1=l2+k′(l1-l2) When the honey badger searches for nectar, the nectar emits fragrance to generate a fragrance density factor. A random number k″ is set, and k″∈[0,1]. The intensity between the honey badger and the nectar source is m, and the distance between the honey badger and the nectar source is n. The calculation formula for the honey badger's olfactory intensity G is as follows: As time t increases during the honey badger's search for nectar, the fragrance density factor changes continuously, and the power emergency material delivery path changes direction continuously. The fragrance density factor calculation formula is as follows: Where F′ represents the fragrance density factor, k″′ represents a random number and k″′≥1, t max Indicates the maximum number of iterations; S322. Determine the direction of the maximum fragrance density factor, which is the direction of the power emergency material delivery path, recorded as m′. During the process of digging for nectar sources, the honey badger changes its position to obtain a new position, and obtains the first delivery position x2 according to the direction of the power emergency material delivery path; the honey badger continues to look for nectar sources, constantly updates the direction of the power emergency material delivery path to obtain a new delivery position, and obtains the second delivery position x3.

8. The method for emergency command and on-site dispatching based on a heating network model according to claim 3 is characterized in that: The honey badger continuously searches for honey sources, continuously updates the delivery location, and updates the fitness function value under the power emergency material delivery path. It continuously iterates according to the fitness function value under the power emergency material delivery path. The current number of iterations is time t. When the current number of iterations is greater than t max , stop the iteration, and the path of the honey badger looking for the nectar source is the optimized power emergency material transportation path; otherwise, repeat S251 and S252 until the current number of iterations is greater than t max .

9. The method for emergency command and on-site dispatching based on a heating network model according to claim 4 is characterized in that: The S4 comprises the following steps: S41. Establish a thermal dispatching command platform to display the temperature, pressure and flow data of the heating pipeline in real time; establish a thermal database to store the past temperature, pressure, flow, heating pipeline prediction data and leakage point data of the heating pipeline.

10. Implement the emergency command and on-site dispatching system based on the heating network model as claimed in any one of claims 1 to 9, characterized in that: Specifically include: heat network model building module, heat supply pipeline data control module, heat supply pipeline data prediction module, heat supply pipeline leakage point location module and power emergency material transportation route optimization module; The heat network model building module is used to build a pressure, flow and temperature model of the heating pipeline; The heating pipeline data control module is used to use the model to calculate the pressure, flow and temperature of the heating pipeline and to perform real-time control on abnormal situations; The heating pipeline data prediction module is used to train the BP neural network and use the BP neural network to predict the heating pipeline data; The heating pipeline leakage point positioning module is used to detect leakage points on the monitoring image using image processing to verify the location of the suspected heating pipeline leakage point; The power emergency material delivery path optimization module is used to optimize the power emergency material delivery path using the honey badger optimization algorithm.

Images