Dynamic heat dissipation adjusting method and system for wind direction predictor of wind energy storage cabinet

Abstract

The invention discloses a dynamic heat dissipation adjusting method and system for a wind direction predictor of a wind energy storage cabinet, and belongs to the technical field of environment-friendly energy equipment, and the method comprises the steps: obtaining the actual working temperature of the wind direction predictor and related environment parameter data in real time, and storing the actual working temperature and related environment parameter data in a constructed parameter database; analyzing the data in the parameter database to construct a temperature model of the wind direction predictor; reading environmental parameter data in the parameter database, performing short-term prediction on environmental parameters, constructing an environmental prediction model, and generating an environmental prediction sequence; inputting the environment prediction sequence into the temperature model, calculating the predicted working temperature of the wind direction predictor, and formulating a heat dissipation regulation table; the actual working temperature of the wind direction predictor is continuously collected, the deviation between the actual working temperature and the predicted working temperature is calculated, the heat dissipation adjusting table is updated, and unnecessary energy consumption is reduced.

Description

Technical Field

[0001] The invention relates to a dynamic heat dissipation adjustment method and system for a wind direction predictor of a wind energy storage cabinet, belonging to the technical field of environmental protection energy equipment. Background Art

[0002] In wind energy storage, the heat dissipation structure of various equipment quickly discharges the heat inside the equipment, thereby keeping the equipment running within the normal operating temperature range, reducing the increase in energy consumption caused by overheating, and thus reducing carbon emissions. Through heat dissipation, frequent maintenance and replacement due to overheating of equipment can be avoided, achieving the goal of energy saving and environmental protection.

[0003] The existing Chinese patent with publication number CN117350159A discloses a method, device, equipment and storage medium for early warning of heat dissipation performance of a generator, which obtains the heat dissipation data of the generator in a historical time period, and then pre-processes the heat dissipation data of the generator to obtain the processed heat dissipation data, and then trains the preset temperature rise model according to the processed heat dissipation data to obtain the target temperature rise model, and then determines the heat dissipation performance coefficient based on the target temperature rise model, and performs a generator heat dissipation performance early warning according to the heat dissipation performance coefficient. The invention effectively trains the preset temperature rise model according to the processed heat dissipation data, and the obtained target temperature rise model can estimate the heat dissipation performance of the generator, and then determines the heat dissipation performance coefficient based on the target temperature rise model, and performs a generator heat dissipation performance early warning according to the heat dissipation performance coefficient.

[0004] Although the prior art processes the heat dissipation system of the generator in advance to avoid power loss caused by temporary shutdown of the generator, it does not take into account the actual temperature value of the equipment during the power generation process of the generator, especially the response strategy when there is a large difference between the estimated value and the actual value. Therefore, this application provides a dynamic heat dissipation adjustment method and system for a wind energy storage cabinet wind direction predictor, which predicts future heat dissipation needs, formulates a heat dissipation adjustment strategy, and dynamically adjusts and optimizes the heat dissipation adjustment strategy according to the actual heat generation of the equipment to improve the heat dissipation effect and save energy resources. Summary of the invention

[0005] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a dynamic heat dissipation adjustment method and system for a wind direction predictor of a wind energy storage cabinet. By predicting the working temperature of the wind direction predictor, a heat dissipation adjustment table is constructed, the predicted working temperature is compared with the safe temperature range of the wind direction predictor, and the expected target flow rate of the coolant in the heat dissipation device is calculated. When the time reaches the expected point, the flow rate of the coolant is controlled, and the actual working temperature is monitored in real time, and the current heat dissipation adjustment plan is adjusted in real time based on the actual working temperature.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A dynamic heat dissipation adjustment method for a wind energy storage cabinet wind direction predictor, comprising:

[0008] Step S1: acquiring the actual working temperature and related environmental parameter data of the wind direction predictor in real time, preprocessing the collected data, and saving them into the constructed parameter database;

[0009] Step S2: constructing a temperature model of the wind direction predictor by analyzing the data in the parameter database, wherein the temperature model is used to predict the operating temperature under given environmental parameters;

[0010] Step S3: reading the environmental parameter data in the parameter database, predicting the environmental parameters, building an environmental prediction model, and generating an environmental prediction sequence;

[0011] Step S4: inputting the environmental prediction sequence into the temperature model, calculating the predicted working temperature of the wind direction predictor, and formulating a heat dissipation adjustment table;

[0012] Step S5: continuously collecting the actual working temperature of the wind direction predictor, calculating the deviation between the actual working temperature and the predicted working temperature, and updating the heat dissipation adjustment table.

[0013] Specifically, the step S2 includes:

[0014] S2.1: Use regression analysis to build a temperature model, define the working temperature of the wind direction predictor as the dependent variable, and the environmental parameters as the independent variables; the objective function expression of the temperature model is as follows:

[0015]

[0016] In the formula, is the dependent variable of the model, is the number of independent variables in the model, is the first environmental parameters, For parameters The regression coefficient of

[0017] S2.2: Constructing the target parameter function ;

[0018] S2.3: Initialize parameter group , and set the initial number of iterations ;

[0019] S2.4: For the parameters in the target parameter function Solve partial derivatives and calculate the parameters The gradient value ;

[0020] S2.5: Parameters based on gradient values The value is updated, and the expression is as follows:

[0021]

[0022] In the formula, is the learning rate.

[0023] Specifically, the step S2 further includes:

[0024] S2.6: Set the iteration interval threshold of the learning rate update to , and calculate the current number of iterations with the iteration interval threshold The remainder between ;

[0025] S2.7: Based on the remainder Determine whether to perform learning rate attenuation calculation; if , the current number of iterations reaches the predetermined iteration interval threshold, and the learning rate is updated to ;like , the current number of iterations does not reach the predetermined iteration interval threshold, and the learning rate is not updated; where, To learn the decay factor;

[0026] S2.8: Update the number of iterations to , get the target parameter function The value of the parameter before updating and the updated values ​​of the parameters , and calculate the rate of change of the target parameter function ;

[0027] S2.9: Set the convergence threshold of the objective parameter function to The change rate threshold is , determine whether the target parameter function converges; if and , the target parameter function converges and enters S2.10; if or , the target parameter function does not converge, and returns to S2.4 to continue updating the parameters;

[0028] S2.10: Get the converged parameter group , the temperature model training is completed.

[0029] Specifically, step S3 includes:

[0030] S3.1: Constructing the environmental parameter matrix ;in, For the A sequence of environmental parameters, is the number of samples collected;

[0031] S3.2: Use recurrent neural network to build an environmental prediction model. The objective function expression is as follows:

[0032]

[0033] In the formula, is the output matrix of the environmental prediction model, is the activation function, is the connection weight matrix between the input layer and the hidden layer, is the input matrix of the environmental prediction model, is the bias matrix;

[0034] S3.3: Based on the environmental parameter matrix , defines the input matrix of the environmental prediction model , Theoretical output matrix ;in, is the sample size of the input matrix;

[0035] S3.4: Using the ant colony optimization algorithm to optimize the connection weight matrix in the environmental prediction model and the bias matrix Solve and optimize;

[0036] S3.5: Real-time acquisition Environmental sequence of moments , and output by the environmental prediction model The environmental prediction sequence at each moment .

[0037] Specifically, step S4 includes:

[0038] S4.1: The environmental prediction sequence Input the temperature model to calculate the predicted working temperature of the wind direction predictor and saving the predicted operating temperature into a constructed heat dissipation adjustment table;

[0039] S4.2: Set the safe temperature range of the wind direction predictor to ;

[0040] S4.3: Determine the predicted operating temperature Is it within the safe temperature range? , the predicted operating temperature is within the safe temperature range; if or , the predicted operating temperature is not within the safe temperature range, and the heat dissipation adjustment table is The predicted working temperature at each moment is marked as an early warning.

[0041] Specifically, the step S4 further includes:

[0042] S4.4: Obtain the predicted operating temperature with a warning mark in the heat dissipation adjustment table , calculate Estimated target flow rate of coolant at the moment ;

[0043] S4.5: Estimated target flow rate of coolant at the moment Save to the heat dissipation adjustment table.

[0044] Specifically, step S5 includes:

[0045] S5.1: When the time arrives At this time, adjust the coolant pump so that the coolant flow rate is , and obtain Actual working temperature at the moment ;

[0046] S5.2: Calculation Always predict operating temperature The actual working temperature The deviation between ;

[0047] S5.3: Set the temperature deviation threshold to ,judge Is the deviation of the time within the acceptable range? , The deviation of the time is within the acceptable range; if , If the deviation of the time is not within an acceptable range, the heat dissipation adjustment table is dynamically adjusted.

[0048] Specifically, adjusting the heat dissipation adjustment table in S5.3 further includes:

[0049] Calculated based on deviation value and temperature deviation threshold Proportional coefficient of flow rate adjustment at any time ;

[0050] like , adjust the coolant pump so that the coolant flow rate is ;

[0051] like , adjust the coolant pump so that the coolant flow rate is .

[0052] A dynamic heat dissipation adjustment system for a wind energy storage cabinet wind direction predictor, comprising: a data acquisition module, a prediction module and an adjustment module;

[0053] The data acquisition module is used to obtain the actual working temperature of the wind direction predictor and related environmental parameter data in real time, pre-process the collected data, and save them in the constructed parameter database;

[0054] The prediction module is used to predict environmental parameters, generate an environmental prediction sequence, and analyze the relationship between the working temperature of the wind direction predictor and the environmental parameters to predict the predicted working temperature of the environmental prediction sequence;

[0055] The adjustment module is used to construct a heat dissipation adjustment table according to the predicted working temperature, and compare the predicted working temperature with the safe temperature range of the wind direction predictor to formulate a heat dissipation adjustment strategy; continuously collect the actual working temperature of the wind direction predictor, and make local adjustments to the heat dissipation adjustment table by comparing it with the predicted working temperature.

[0056] Specifically, the prediction module includes an environment prediction unit and a temperature prediction unit;

[0057] The environment prediction unit is used to construct an environment prediction model. Environmental sequence of moments Input the environmental prediction model to generate The environmental prediction sequence at each moment ;

[0058] The temperature prediction unit is used to construct a temperature model, update the parameters by calculating the gradient values ​​of each parameter in the temperature model; and output the environmental prediction sequence Predicted operating temperature .

[0059] Specifically, the heat dissipation adjustment strategy and the update strategy are configured in the adjustment module;

[0060] The heat dissipation adjustment strategy determines whether the predicted working temperature of the wind direction predictor should be adjusted by setting the safe temperature range of the wind direction predictor, and calculates the target flow rate of the expected coolant. , saving the target flow rate into the heat dissipation adjustment table;

[0061] The update strategy continuously collects the actual operating temperature of the wind direction predictor, calculates the deviation between the actual operating temperature and the predicted operating temperature, and updates the heat dissipation adjustment table.

[0062] Beneficial effects of the present invention:

[0063] 1. Based on the constructed temperature model and environmental prediction model, the future operating temperature can be predicted, and a heat dissipation adjustment table can be formulated accordingly, which helps to take measures in advance to optimize heat dissipation management and prevent the equipment from consuming more energy to maintain normal operation due to overheating; at the same time, by continuously monitoring the deviation between the actual operating temperature and the predicted operating temperature, and updating the heat dissipation adjustment table in a timely manner, it can be ensured that the wind direction predictor always operates within the appropriate operating temperature range.

[0064] 2. Through accurate temperature prediction and heat dissipation management, unnecessary energy waste can be avoided. For example, when it is predicted that the temperature will not be too high in the future, the power consumption of the heat dissipation system can be appropriately reduced, thus achieving the goal of energy saving and consumption reduction. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 A schematic diagram of a dynamic heat dissipation adjustment method for a wind energy storage cabinet wind direction predictor;

[0066] Figure 2 A temperature prediction flow chart of a dynamic heat dissipation adjustment method for a wind direction predictor of a wind energy storage cabinet;

[0067] Figure 3 It is an environmental prediction flow chart of a dynamic heat dissipation adjustment method of a wind direction predictor for a wind energy storage cabinet;

[0068] Figure 4 A heat dissipation adjustment flow chart of a dynamic heat dissipation adjustment method for a wind direction predictor of a wind energy storage cabinet;

[0069] Figure 5 A heat dissipation adjustment update flow chart of a dynamic heat dissipation adjustment method for a wind direction predictor of a wind energy storage cabinet;

[0070] Figure 6 This is a structural diagram of a dynamic heat dissipation adjustment system for a wind direction predictor of a wind energy storage cabinet. DETAILED DESCRIPTION

[0071] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. The embodiments of the present invention and the technical features in the embodiments may be combined with each other unless there is a conflict.

[0072] Example 1

[0073] refer to Figures 1 to 5 As shown, this embodiment introduces a dynamic heat dissipation adjustment method of a wind energy storage cabinet wind direction predictor, comprising the following steps:

[0074] Step S1: Use the high-precision temperature sensor configured in the wind direction predictor to obtain the actual working temperature of the wind direction predictor. At the same time, according to the external environment detection equipment, such as anemometer, wind vane, thermometer and hygrometer, and barometer, collect relevant environmental parameter data, pre-process the collected data, and save it to the constructed parameter database; during the data collection process, set a unified initial collection time and sampling interval for the sampling equipment And sampling points , real-time recording of data collection time, so that the data has a unified timestamp, achieving data time synchronization, and facilitating subsequent statistics and analysis;

[0075] Step S2: Based on the data in the parameter database, the relationship between the actual working temperature of the wind direction predictor and the environmental parameters is analyzed, a temperature model of the wind direction predictor is constructed by using a regression analysis method, and the temperature model is trained until the temperature model predicts the working temperature of the wind direction predictor under given environmental parameters;

[0076] Step S3: Read the environmental parameter data in the parameter database, analyze the trend of historical environmental parameters, make short-term predictions on the environmental parameters, build an environmental prediction model, and generate an environmental prediction sequence; in addition, environmental prediction helps to optimize the layout of wind power generation equipment and improve energy utilization;

[0077] Step S4: input the environmental prediction sequence into the trained temperature model, calculate the predicted working temperature of the wind direction predictor, build a heat dissipation adjustment table based on the predicted working temperature, compare the predicted working temperature with the safe temperature range of the wind direction predictor, and formulate a heat dissipation adjustment strategy so that the wind direction predictor can maintain an appropriate temperature when working. For example, in an overheated environment, the device needs to consume more energy to maintain normal operation, which leads to increased energy consumption, avoiding performance degradation or damage of the wind direction predictor due to overheating during operation, and reducing the maintenance and replacement costs caused by equipment failure;

[0078] Step S5: Continuously collect the actual working temperature of the wind direction predictor to ensure the real-time data, compare the actual working temperature with the corresponding predicted working temperature, calculate the deviation, and locally adjust and update the heat dissipation adjustment table according to the deviation to ensure the effectiveness of the heat dissipation adjustment table while avoiding unnecessary energy consumption, thereby achieving the goal of energy conservation and emission reduction.

[0079] Specifically, the specific steps of step S2 include:

[0080] S2.1: Use regression analysis to build a temperature model, and define the working temperature of the wind direction predictor as the dependent variable of the model. The working temperature is affected by various environmental factors. For example, high humidity will increase the heat load of the wind direction predictor, resulting in an increase in the working temperature. When the air pressure is low, the air density is small and the heat dissipation capacity decreases, resulting in an increase in the working temperature. Therefore, various environmental parameters are defined as independent variables of the model, such as ambient temperature, humidity, wind speed, wind direction, and air pressure. The objective function expression of the model is as follows:

[0081]

[0082] In the formula, is the dependent variable of the model, is the number of independent variables in the model, and is the first environmental parameters, For parameters The regression coefficient of is the parameter group to be solved in the temperature model. By solving and optimizing the parameter group, the temperature model can accurately predict the working temperature of the wind direction predictor, which is recorded as the predicted working temperature of the wind direction predictor.

[0083] S2.2: In order to solve the specific value of the parameter group, construct the target parameter function , the objective parameter function is used to measure the difference between the predicted operating temperature and the actual operating temperature; the expression is as follows:

[0084]

[0085] In the formula, is the number of samples of historical data collected, For the The actual working temperature of the wind direction predictor in the samples, For the The predicted operating temperature of the environmental parameters in the samples is output by the temperature model, and , , , , Respectively The value of each independent variable in the sample;

[0086] S2.3: Parameter group Set an initial value and set the initial number of iterations , such as setting ;

[0087] S2.4: Parameters in the target parameter function Solve partial derivatives and calculate parameters The gradient value , the expression is as follows:

[0088]

[0089] In the formula, For the In the sample The values ​​of the independent variables, and , and set ;

[0090] S2.5: Parameters based on gradient values The value is updated, and the expression is as follows:

[0091]

[0092] In the formula, For the learning rate, set the initial learning rate In this embodiment, ,A smaller initial learning rate helps prevent the temperature model from skipping the optimal value of the parameter due to a large step size in the early stages of training;

[0093] S2.6: Set the iteration interval threshold of the learning rate update to , so that every The learning rate is updated in iterations, for example, , the learning rate is updated every 10 iterations; and the modulus operator is used Calculate the current iteration number and iteration interval threshold The remainder between , the expression is as follows:

[0094]

[0095] S2.7: Based on remainder Determine whether to perform learning rate attenuation calculation; if , the current number of iterations reaches the predetermined iteration interval threshold, and the learning rate is updated to ,in, is the learning attenuation factor and In this embodiment, , by regularly reducing the learning rate to make the temperature model close to the optimal solution, the step size of parameter update gradually decreases, which helps to adjust the parameters more finely and improve the adaptability and robustness of the model; if , the current number of iterations does not reach the predetermined iteration interval threshold, and the learning rate is not updated;

[0096] S2.8: Update the number of iterations to , get the target parameter function The value of the parameter before updating and the updated values ​​of the parameters , and calculate the rate of change of the target parameter function ;

[0097] S2.9: Set the convergence threshold of the objective parameter function to The change rate threshold is , to determine whether the target parameter function converges; if and , the target parameter function converges and enters S2.10; if or , the target parameter function does not converge, and returns to S2.4 to continue updating the parameters;

[0098] S2.10: Get the converged parameter group , substitute the parameter group into the temperature model, the temperature model training is now completed, and the temperature model is output for subsequent predictions.

[0099] Specifically, the specific steps of step S3 include:

[0100] S3.1: Read environmental parameter data from the parameter database and construct an environmental parameter matrix ;in, is the number of samples collected, based on the set sampling interval And sampling points The prediction interval can be calculated , For the A sequence of environmental parameters, and , , is the number of environmental parameters, is a sequence of parameters Middle Environmental parameter values, ;

[0101] S3.2: Use a recurrent neural network to build an environmental prediction model, which includes an input layer, a hidden layer, and an output layer. The objective function expression of the environmental prediction model is as follows:

[0102]

[0103] In the formula, is the output matrix of the environmental prediction model, is the activation function, is the connection weight matrix between the input layer and the hidden layer, and , For the The weight value corresponding to the environmental parameter value is is the input matrix of the environmental prediction model, is the bias matrix;

[0104] S3.3: Based on the environmental parameter matrix , defines the input matrix of the environmental prediction model , Theoretical output matrix ;in, is the sample size of the input matrix, and , when the input sequence is When , the output sequence of the environmental prediction model is , achieving the goal of monitoring environmental parameters every Numerical prediction of time periods;

[0105] S3.4: Using the ant colony optimization algorithm to solve and optimize the connection weight matrix and bias matrix in the environmental prediction model, and obtain the environmental prediction model;

[0106] S3.5: Real-time collection of current time Environmental sequence , and output through the environmental prediction model The environmental prediction sequence at each moment ;in, for The moment Environmental parameter values, The environmental prediction model predicts The moment An environmental parameter value.

[0107] Specifically, the solution steps using the ant colony optimization algorithm in S3.4 also include:

[0108] S3.41: Initialization parameters, including connection weight matrix , bias matrix , the number of ant colonies , initial value of iteration times ;

[0109] S3.42: Based on the known input sequence and objective function, calculate the probability of each ant choosing the next path, and move each ant to a new position according to the selection probability, and record the parameter values ​​corresponding to each ant moving to the new position in real time, including , and ;in, is the prediction output matrix of the environmental prediction model;

[0110] S3.43: Calculate the error between the theoretical output matrix and the predicted output matrix , the expression is as follows:

[0111]

[0112] S3.44: Record the parameter value corresponding to the minimum error of the current iteration, mark it as the local optimal solution, and perform pheromone update on the optimal solution of the current iteration;

[0113] S3.45: Set the number of optimization iterations to , to determine whether the number of optimization iterations has been reached; if , the number of optimization iterations is not reached, let , return to S3.42 and start a new round of iterative calculation; if , reaching the optimal number of iterations, output the current iteration and .

[0114] Specifically, the specific steps of step S4 include:

[0115] S4.1: The environmental prediction sequence at each moment Input the trained temperature model to calculate the predicted working temperature of the wind direction predictor , and save the predicted working temperature to the constructed heat dissipation adjustment table; wherein the heat dissipation adjustment table is used to save the predicted working temperature of the wind direction predictor, the initial flow rate of the coolant, the estimated target flow rate, the actual working temperature collected in real time after a period of time, and the actual flow rate judged based on the actual working temperature. The heat dissipation device in this embodiment is a liquid-cooled heat dissipation device. When the wind direction predictor is working, the heat generated is transferred to the liquid cooling pipe through the radiator, and the heat is absorbed by the circulating flow of the coolant. The initial flow rate of the coolant is set to be constant, which is ;

[0116] S4.2: Set the safe temperature range of the wind direction predictor to ;in, is the lower limit of the working temperature of the wind direction predictor, The upper limit of the working temperature of the wind direction predictor;

[0117] S4.3: Determine the predicted operating temperature Is it within the safe temperature range? , predict the operating temperature Within the safe temperature range; if or , predict the operating temperature Not within the safe temperature range, in the heat dissipation adjustment table The predicted working temperature at each moment is marked as an early warning;

[0118] S4.4: Get the predicted operating temperature with warning mark in the thermal adjustment table According to the heat balance equation, we can calculate Estimated target flow rate of coolant at the moment , the expression is as follows:

[0119]

[0120] In the formula, To predict the specific time corresponding to the working temperature, is the power consumption of the wind direction predictor, The heat generated during the operation of the wind direction predictor, is the heat transfer efficiency of the radiator, is the heat dissipation area of ​​the radiator, is the density of the coolant, is the specific heat capacity, is the difference between the target temperature and the predicted ambient temperature, , To predict the ambient temperature, Extraction from the environmental prediction sequence at each moment;

[0121] S4.5: Estimated target flow rate of coolant at the moment The flow rate is saved in the heat dissipation adjustment table so that the liquid cooling device can control the opening of the coolant pump or valve through the target flow rate at the corresponding time in the heat dissipation adjustment table.

[0122] Specifically, the specific steps of step S5 include:

[0123] S5.1: Continuously collect the actual working temperature of the wind direction predictor. When the time reaches At this time, adjust the coolant pump so that the coolant flow rate is , and obtain Actual working temperature at the moment ,Since the liquid cooling tube dissipates the heat transferred from the radiator into the air through the cooling fins, the external environmental parameters change slightly. At the same time, in order to cope with the error between the prediction model and the actual parameters, the dynamic adjustment of the heat dissipation adjustment strategy is realized;

[0124] S5.2: Calculation Always predict operating temperature The actual working temperature The deviation between , the expression is as follows:

[0125]

[0126] S5.3: Set the temperature deviation threshold to ,judge Is the deviation of the time within the acceptable range? , The deviation of the time is within the acceptable range; if , If the time deviation is not within the acceptable range, the heat dissipation adjustment table is dynamically adjusted.

[0127] Specifically, the specific steps of adjusting the heat dissipation adjustment table in S5.3 include:

[0128] Calculated based on deviation value and temperature deviation threshold Proportional coefficient of flow rate adjustment at any time , the expression is as follows:

[0129]

[0130] like , adjust the coolant pump so that the coolant flow rate is ;

[0131] like , adjust the coolant pump so that the coolant flow rate is .

[0132] Example 2

[0133] See also Figure 6 , another embodiment provided by the present invention: a dynamic heat dissipation adjustment system of a wind energy storage cabinet wind direction predictor, comprising: a data acquisition module, a prediction module and an adjustment module;

[0134] The data acquisition module is used to obtain the actual working temperature of the wind direction predictor according to the high-precision temperature sensor configured in the wind direction predictor, and collect relevant environmental parameter data according to external environment detection equipment such as anemometer, wind vane, thermometer and hygrometer, and barometer, pre-process the collected data, and save it to the constructed parameter database;

[0135] The prediction module is used to analyze the data in the parameter database, make short-term predictions on environmental parameters, build an environmental prediction model, generate an environmental prediction sequence, analyze the relationship between the actual working temperature of the wind direction predictor and the environmental parameters, predict the working temperature of the wind direction predictor under given environmental parameters, generate predicted working temperature based on the environmental prediction sequence, and transmit the prediction results to the adjustment module in real time;

[0136] The adjustment module is used to construct a heat dissipation adjustment table according to the prediction results, compare the predicted working temperature with the safe temperature range of the wind direction predictor, formulate a heat dissipation adjustment strategy, and update the heat dissipation adjustment table in real time, so that the wind direction predictor can maintain an appropriate temperature when working, avoid performance degradation or damage of the wind direction predictor due to overheating during operation, reduce the maintenance and replacement costs caused by equipment failure, and continuously collect the actual working temperature of the wind direction predictor. By comparing it with the predicted working temperature, the heat dissipation adjustment table is locally adjusted to ensure the effectiveness of the heat dissipation adjustment table, while avoiding unnecessary energy consumption, thereby achieving the goal of energy conservation and emission reduction.

[0137] Specifically, the prediction module includes an environment prediction unit and a temperature prediction unit;

[0138] The environmental prediction unit uses a recurrent neural network to build an environmental prediction model, and uses a training set to train the environmental prediction model, adjust and optimize the parameters in the environmental prediction model, so that the environmental prediction model can achieve short-term prediction of environmental parameters; obtain Environmental sequence of moments , and use the trained environment prediction model to output The environmental prediction sequence at each moment ;

[0139] The temperature prediction unit is used to analyze the relationship between the actual working temperature of the wind direction predictor and the environmental parameters. The temperature model is constructed using the regression analysis method. In order to solve the unknown parameters in the temperature model, the target parameter function is constructed and the gradient values ​​of each parameter are calculated. Each parameter is updated based on the gradient value, and the change rate of the target parameter function before and after the update is calculated to determine whether the target parameter function converges. If not, the parameters are continued to be updated. If converged, the specific parameter values ​​are output. At this time, the temperature model training is completed; and the environmental prediction sequence is output. Input to the trained temperature model to output the predicted operating temperature of the wind direction predictor .

[0140] Specifically, an ant colony optimization strategy is configured in the environmental prediction unit. The ant colony optimization strategy is used to solve the parameters in the environmental prediction model, move the ants to new positions according to the probability of each ant choosing a path, and record the parameter values ​​and objective function values ​​in real time. The error between the theoretical value and the actual value is calculated, and the pheromone is updated for the value with the minimum error until the iteration is completed, and the optimal parameter value of the current iteration is output.

[0141] Specifically, the heat dissipation adjustment strategy and update strategy are configured in the adjustment module;

[0142] The heat dissipation adjustment strategy sets the safe temperature range of the wind direction predictor to determine whether the predicted working temperature of the wind direction predictor should be adjusted for heat dissipation, and uses the heat balance equation to calculate the expected target flow rate of the coolant at the corresponding time. and save the estimated target flow rate to the corresponding position in the heat dissipation adjustment table, so that the liquid cooling device controls the opening of the coolant pump or valve through the target flow rate at the corresponding time in the heat dissipation adjustment table;

[0143] The update strategy continuously obtains the actual working temperature of the wind direction predictor, compares it with the predicted working temperature at the corresponding time, calculates the deviation, and locally adjusts and updates the heat dissipation adjustment table according to the deviation to deal with emergencies during the actual operation of the wind direction predictor.

[0144] In summary, the present invention analyzes environmental parameters and the actual working temperature of the wind direction predictor to construct an environmental prediction model to achieve short-term prediction of environmental parameters. At the same time, a temperature model is constructed to analyze the relationship between environmental parameters and the actual working temperature, so that the predicted working temperature of the wind direction predictor at the corresponding time is obtained according to the results of the short-term environmental prediction, and a heat dissipation adjustment strategy is formulated based on the predicted working temperature to avoid performance degradation or damage of the wind direction predictor due to overheating during operation. At the same time, the actual working temperature of the wind direction predictor is continuously collected and compared with the predicted working temperature at the corresponding time, and the heat dissipation adjustment strategy is adjusted in real time to avoid unnecessary energy waste.

[0145] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A dynamic heat dissipation adjustment method for a wind energy storage cabinet wind direction predictor, characterized in that: include: Step S1: acquiring the actual working temperature and related environmental parameter data of the wind direction predictor in real time, preprocessing the collected data, and saving them into the constructed parameter database; Step S2: constructing a temperature model of the wind direction predictor by analyzing the data in the parameter database, wherein the temperature model is used to predict the operating temperature under given environmental parameters; Step S3: reading the environmental parameter data in the parameter database, predicting the environmental parameters, building an environmental prediction model, and generating an environmental prediction sequence; Step S4: inputting the environmental prediction sequence into the temperature model, calculating the predicted working temperature of the wind direction predictor, and formulating a heat dissipation adjustment table; Step S5: continuously collecting the actual working temperature of the wind direction predictor, calculating the deviation between the actual working temperature and the predicted working temperature, and updating the heat dissipation adjustment table.

2. The dynamic heat dissipation adjustment method of a wind energy storage cabinet wind direction predictor according to claim 1 is characterized in that: The step S2 comprises: S2.1: Use regression analysis to build a temperature model, define the working temperature of the wind direction predictor as the dependent variable, and the environmental parameters as the independent variables; the objective function expression of the temperature model is as follows: ； In the formula, is the dependent variable of the model, is the number of independent variables in the model, is the first environmental parameters, For parameters The regression coefficient of S2.2: Constructing the target parameter function ; S2.3: Initialize parameter group , and set the initial number of iterations ; S2.4: For the parameters in the target parameter function Solve partial derivatives and calculate the parameters The gradient value ; S2.5: Parameters based on gradient values The value is updated, and the expression is as follows: ； In the formula, is the learning rate.

3. The dynamic heat dissipation adjustment method of a wind energy storage cabinet wind direction predictor according to claim 2 is characterized in that: The step S2 further comprises: S2.6: Set the iteration interval threshold of the learning rate update to , and calculate the current number of iterations The iteration interval threshold The remainder between ; S2.7: Based on the remainder Determine whether to perform learning rate attenuation calculation; if , the current number of iterations reaches the predetermined iteration interval threshold, and the learning rate is updated to ;like , the current number of iterations does not reach the predetermined iteration interval threshold, and the learning rate is not updated; where, To learn the decay factor; S2.8: Update the number of iterations to , get the target parameter function The value of the parameter before updating and the updated values ​​of the parameters , and calculate the rate of change of the target parameter function ; S2.9: Set the convergence threshold of the objective parameter function to The change rate threshold is , determine whether the target parameter function converges; if and , the target parameter function converges and enters S2.10; if or , the target parameter function does not converge, and returns to S2.4 to continue updating the parameters; S2.10: Get the converged parameter group , the temperature model training is completed.

4. The dynamic heat dissipation adjustment method of a wind energy storage cabinet wind direction predictor according to claim 3 is characterized in that: The step S3 comprises: S3.1: Constructing the environmental parameter matrix ;in, For the A sequence of environmental parameters, is the number of samples collected; S3.2: Use recurrent neural network to build an environmental prediction model. The objective function expression is as follows: ； In the formula, is the output matrix of the environmental prediction model, is the activation function, is the connection weight matrix between the input layer and the hidden layer, is the input matrix of the environmental prediction model, is the bias matrix; S3.3: Based on the environmental parameter matrix , defines the input matrix of the environmental prediction model , Theoretical output matrix ;in, is the sample size of the input matrix; S3.4: Using the ant colony optimization algorithm to optimize the connection weight matrix in the environmental prediction model and the bias matrix Solve and optimize; S3.5: Real-time acquisition Environmental sequence of moments , and output by the environmental prediction model The environmental prediction sequence at each moment .

5. The dynamic heat dissipation adjustment method of a wind energy storage cabinet wind direction prediction instrument according to claim 4 is characterized in that: The step S4 comprises: S4.1: Sequence the environmental prediction Input the temperature model to calculate the predicted working temperature of the wind direction predictor and saving the predicted operating temperature into a constructed heat dissipation adjustment table; S4.2: Set the safe temperature range of the wind direction predictor to ; S4.3: Determine the predicted operating temperature Is it within the safe temperature range? , the predicted operating temperature is within the safe temperature range; if or , the predicted operating temperature is not within the safe temperature range, and the heat dissipation adjustment table is The predicted working temperature at each moment is marked as an early warning.

6. The dynamic heat dissipation adjustment method of a wind energy storage cabinet wind direction prediction instrument according to claim 5 is characterized in that: The step S4 further comprises: S4.4: Obtain the predicted operating temperature with a warning mark in the heat dissipation adjustment table , calculate Estimated target flow rate of coolant at the moment ; S4.5: Estimated target flow rate of coolant at the moment Save to the heat dissipation adjustment table.

7. The dynamic heat dissipation adjustment method of a wind energy storage cabinet wind direction prediction instrument according to claim 6 is characterized in that: The step S5 comprises: S5.1: When the time arrives At this time, adjust the coolant pump so that the coolant flow rate is , and obtain Actual working temperature at the moment ; S5.2: Calculation Always predict operating temperature The actual working temperature The deviation between ; S5.3: Set the temperature deviation threshold to ,judge Is the deviation of the time within the acceptable range? , The deviation of the time is within the acceptable range; if , If the deviation of the time is not within an acceptable range, the heat dissipation adjustment table is dynamically adjusted.

8. The dynamic heat dissipation adjustment method of a wind energy storage cabinet wind direction prediction instrument according to claim 7 is characterized in that: The adjustment of the heat dissipation adjustment table in S5.3 also includes: Calculated based on deviation value and temperature deviation threshold Proportional coefficient of flow rate adjustment at any time ; like , adjust the coolant pump so that the coolant flow rate is ; like , adjust the coolant pump so that the coolant flow rate is .

9. A dynamic heat dissipation adjustment system for a wind energy storage cabinet wind direction predictor, which is used to implement a dynamic heat dissipation adjustment method for a wind energy storage cabinet wind direction predictor as claimed in any one of claims 1 to 8, characterized in that: include: Data collection module, prediction module and adjustment module; The data acquisition module is used to obtain the actual working temperature of the wind direction predictor and related environmental parameter data in real time, pre-process the collected data, and save them in the constructed parameter database; The prediction module is used to predict environmental parameters, generate an environmental prediction sequence, and analyze the relationship between the working temperature of the wind direction predictor and the environmental parameters to predict the predicted working temperature of the environmental prediction sequence; The adjustment module is used to construct a heat dissipation adjustment table according to the predicted working temperature, and compare the predicted working temperature with the safe temperature range of the wind direction predictor to formulate a heat dissipation adjustment strategy; continuously collect the actual working temperature of the wind direction predictor, and make local adjustments to the heat dissipation adjustment table by comparing it with the predicted working temperature.

10. The dynamic heat dissipation adjustment system of the wind direction prediction instrument of the wind energy storage cabinet according to claim 9 is characterized in that: The prediction module includes an environment prediction unit and a temperature prediction unit; The environment prediction unit is used to construct an environment prediction model. Environmental sequence of moments Input the environmental prediction model to generate The environmental prediction sequence at each moment ; The temperature prediction unit is used to construct a temperature model, update the parameters by calculating the gradient values ​​of each parameter in the temperature model; and output the environmental prediction sequence Predicted operating temperature .

11. The dynamic heat dissipation adjustment system of the wind direction prediction instrument of the wind energy storage cabinet according to claim 9, characterized in that: The heat dissipation adjustment strategy and the update strategy are configured in the adjustment module; The heat dissipation adjustment strategy sets the safe temperature range of the wind direction predictor, determines whether the predicted working temperature of the wind direction predictor should be adjusted, and calculates the expected target flow rate of the coolant. , saving the target flow rate into the heat dissipation adjustment table; The update strategy continuously collects the actual operating temperature of the wind direction predictor, calculates the deviation between the actual operating temperature and the predicted operating temperature, and updates the heat dissipation adjustment table.

Images