A battery energy management system for electric stretching equipment

Abstract

The present invention relates to the technical field of battery energy management, and discloses a battery energy management system for electric tensioning equipment. The system includes a battery energy recovery module, a battery energy supply module, a battery energy dynamic distribution module, and a battery energy regulation module; first, the energy of the tension machine is captured and stored in a supercapacitor, and then the energy of the supercapacitor is stored in a battery pack to form a battery energy storage unit; then, the priority of the energy contribution of the tension machine is considered, and the battery energy storage unit is supplied with energy, and then a multi-strategy fusion sparrow search algorithm is used to optimize the current tolerance limit and voltage tolerance limit in the traction machine energy supply to generate a battery energy dynamic distribution strategy; finally, the working state of the battery energy storage unit is adjusted based on the temperature-voltage closed-loop feedback to complete the battery energy management. The present invention achieves the purpose of battery energy management by performing energy management on the electric tensioning equipment, and the method is accurate and objective.

Description

Technical Field

[0001] The present invention relates to the technical field of battery energy management, and in particular to a battery energy management system for electric stretching equipment. Background Art

[0002] Electric tensioning equipment (such as traction machines and tensioners) is a core component used in the installation of high-voltage transmission lines, submarine cables, and optical cables in the power, communications, and rail transportation sectors. With the development of new energy and smart grids, these equipment are gradually transitioning from traditional fuel-powered to electric-powered systems, with lithium-ion batteries being the common power source.

[0003] In traditional equipment, the traction machine and tension machine are usually powered independently. The tension machine generates a large amount of reverse electrical energy due to braking during the wire-laying operation. The traditional resistance braking method converts the electrical energy into heat energy for dissipation. The battery energy management method of traditional electric tensioning equipment feeds electrical energy back to the battery pack, uses a bidirectional converter to adjust the voltage and store energy, and then calls the tension machine battery energy when the traction machine is under high load. The battery energy is managed according to a single energy distribution strategy and centralized control logic to control the energy flow.

[0004] Electric tensioning equipment has the problem of insufficient battery life under complex working conditions. It relies on large-capacity battery packs for power supply. Frequent charging and discharging leads to reduced battery life and increased operation and maintenance costs. At the same time, the energy of the tension machine is recovered for use in the traction machine. The centralized control lacks real-time and adaptability, and the charging and discharging logic calls are chaotic, which may lead to redundant energy waste or the risk of battery overcharge / over-discharge, which is not conducive to the effective recycling of energy. Summary of the Invention

[0005] In response to the problems in the related art, the present invention provides a battery energy management system for an electric stretching device to overcome the technical problems existing in the existing related art.

[0006] To solve the above technical problems, the present invention is achieved through the following technical solutions:

[0007] The present invention is a battery energy management system for electric stretching equipment, which specifically includes: a battery energy recovery module, a battery energy supply module, a battery energy dynamic distribution module and a battery energy regulation module;

[0008] The battery energy recovery module is used to determine whether the tension machine has entered the energy recovery mode, capture the energy of the tension machine and store it in the supercapacitor first, and then store the supercapacitor energy in the battery pack to form a battery energy storage unit;

[0009] The battery energy supply module is used to predict the output power of the traction machine, obtain a predicted value of the traction machine output power, compare the output power of the battery energy storage unit with the predicted value of the traction machine output power, and enable the battery energy storage unit to supply energy to the traction machine;

[0010] The battery energy dynamic allocation module is configured to optimize the current tolerance limit and the voltage tolerance limit during energy supply by using a multi-strategy fusion sparrow search algorithm when the battery energy storage unit supplies energy to the traction machine, and generate a battery energy dynamic allocation strategy;

[0011] The battery energy regulation module is used to regulate the working state of the battery energy storage unit based on temperature-voltage closed-loop feedback in combination with the battery energy dynamic allocation strategy.

[0012] Preferably, the determining whether the tension machine enters the energy recovery mode includes:

[0013] Obtaining tension machine load parameters and motor operating parameters, wherein the tension machine load parameters include cable tension and drum speed, and the motor operating parameters include motor torque and motor speed, obtaining tension machine operating data, recording the tension machine operating time series, and selecting a first time point and a second time point;

[0014] When the cable tension and the drum speed in the first tension machine operation data are both greater than 0 and the cable tension and the drum speed in the second tension machine operation data are both greater than 0, the tension machine enters the energy recovery mode.

[0015] Preferably, the capturing tension machine energy is first stored in a supercapacitor, comprising:

[0016] Obtain the motor torque and motor speed from the operating data of the first tension machine and the operating data of the second tension machine. When the motor torque is a reverse torque output and the direction of the motor torque is opposite to that of the motor speed, the tension machine enters an active braking condition. Then, calculate the differential of the motor speed from the first time point to the second time point, set a speed threshold, and when the motor torque is 0 and the differential of the motor speed from the first time point to the second time point is less than the speed threshold, the tension machine enters a passive braking condition.

[0017] Obtaining the capacity of the battery pack, setting a first capacity threshold, and storing the energy of the capture tension machine in the supercapacitor when the battery pack capacity is greater than the first capacity threshold;

[0018] In the active braking mode, the motor is switched to the power generation mode to charge the supercapacitor; in the passive braking mode, protective braking is triggered, the parallel crowbar circuit is discharged, and then the supercapacitor is charged.

[0019] Preferably, the supercapacitor energy is stored in a battery pack to form a battery energy storage unit comprising:

[0020] Setting a second capacity threshold, when the battery pack capacity is less than or equal to the first capacity threshold or the supercapacitor capacity is greater than the second capacity threshold, a secondary energy storage method is used to store supercapacitor energy in the battery pack;

[0021] Collecting the voltages of battery cells in the battery pack to obtain a battery cell voltage set;

[0022] Setting a first protection voltage and a second protection voltage, when the battery cell voltage in the battery cell voltage set is lower than the first protection voltage, the supercapacitor charges the battery pack normally;

[0023] When the battery cell voltage in the battery cell voltage set is greater than or equal to the first protection voltage and less than or equal to the second protection voltage, reducing the charging current;

[0024] When the battery cell voltage in the battery cell voltage set is greater than the second protection voltage, disconnecting the supercapacitor to charge the battery pack, thereby completing the charging of the battery pack;

[0025] The supercapacitor and battery pack form a battery energy storage unit.

[0026] Preferably, the predicting the output power of the traction machine and outputting the predicted value of the output power of the traction machine include:

[0027] Obtain the motor current, motor voltage, motor speed, motor temperature, load weight and traction speed of the traction machine to generate a traction machine operation data sample set;

[0028] The BiLSTM neural network is configured to include an input layer, a bidirectional LSTM layer, an attention mechanism layer, a dense layer, and an output layer, with an initial learning rate of 0.001. Historical tractor operation data is obtained, and feature sequences of the historical tractor operation data are extracted and normalized to obtain a sample data set. The sample data set is divided into a sample training set and a sample test set. The sample training set is then input into the BiLSTM neural network for iteration until the BiLSTM neural network converges, thereby obtaining a trained BiLSTM neural network.

[0029] Then, the sample test set is input into the trained BiLSTM neural network, the prediction result is output, and the accuracy threshold is set. When the accuracy of the prediction result is greater than the accuracy threshold, the traction machine output power prediction model is obtained. Otherwise, the weight is adjusted until the accuracy of the prediction result is greater than the accuracy threshold; the traction machine operation data sample set is input into the traction machine output power prediction model, and the traction machine output power prediction value is output.

[0030] Preferably, comparing the output power of the battery energy storage unit with the predicted value of the traction machine output power to enable the battery energy storage unit to supply energy to the traction machine includes:

[0031] Calculate the m The path loss and residual energy factor of the tension machine are calculated using the weighted priority algorithm. m The priority score of each tension machine is calculated; the priority scores of all tension machines are calculated in sequence, and the tension machine with the highest priority score is selected and recorded as the priority tension machine;

[0032] The predicted output power value of the traction machine is compared with the output power of the battery pack in the battery energy storage unit. When the predicted output power value of the traction machine is less than the output power of the battery pack in the battery energy storage unit, the battery pack provides energy; otherwise, the priority tension machine is used to charge the supercapacitor, and the supercapacitor cooperates with the battery pack to provide energy to complete the energy supply of the traction machine.

[0033] Preferably, the optimization of the current tolerance limit and the voltage tolerance limit during energy supply using the multi-strategy fusion sparrow search algorithm includes:

[0034] After the traction machine is powered, the transmission current, transmission resistance, internal resistance of the battery pack in the battery energy storage unit, and voltage are obtained. The transmission resistance loss and the output power of the battery energy storage unit are calculated, and the maximum difference between the output power of the battery energy storage unit and the transmission resistance loss is used as the objective function.

[0035] The current tolerance limit and the voltage tolerance limit are set to constrain the range of the transmission current and the battery pack voltage in the battery energy storage unit; the objective function is used as the fitness function, and the process of solving the objective function is regarded as a search space. There is a sparrow population in the search space, and the current tolerance limit and the voltage tolerance limit are combined into a feasible solution. The individual position of the sparrow in the sparrow population represents the feasible solution, and the process of iterating the individual position of the sparrow is regarded as a process of optimizing the current tolerance limit and the voltage tolerance limit; the sparrow population is divided into discoverers and joiners, and the population is initialized using chaotic mapping, and then the elite reverse learning strategy is introduced to calculate the number of sparrows in the sparrow population. i The reverse solution of the individual positions of sparrows;

[0036] The sine-cosine algorithm is introduced to update the position of the discoverer and then the position of the joiner; the current best fitness function value and the current worst fitness function value are calculated, and the sparrow population is controlled to move in the direction of the sparrow individual corresponding to the current best fitness function value, completing the position update of all current stages and generating the next generation of sparrow population; until the current number of iterations reaches the maximum number of iterations, the iteration is stopped to obtain the final sparrow population, and the position of the sparrow individual corresponding to the best fitness function value is found to obtain the optimized current tolerance limit and the optimized voltage tolerance limit.

[0037] Preferably, generating a dynamic battery energy allocation strategy includes:

[0038] During the traction machine power supply, the transmission current is kept within the optimized current tolerance limit, and the battery pack voltage in the battery energy storage unit is within the optimized voltage tolerance limit. The battery energy storage unit provides energy for the traction machine and generates a dynamic battery energy allocation strategy.

[0039] Preferably, the adjusting the working state of the battery energy storage unit based on the temperature-voltage closed-loop feedback to complete the battery energy management includes:

[0040] Running the battery energy dynamic allocation strategy, the battery energy storage unit supplies power to the traction machine, collecting the battery pack temperature and battery pack voltage in the battery energy storage unit, and calculating the dynamic voltage threshold;

[0041] When the battery energy storage unit is powered, a temperature threshold is set. When the battery pack voltage in the battery energy storage unit is lower than the dynamic voltage threshold, the power supply is stopped. When the battery pack temperature in the battery energy storage unit is higher than the temperature threshold, the power supply current is reduced to form a closed-loop feedback regulation to complete battery energy management.

[0042] The present invention has the following beneficial effects:

[0043] 1. This invention determines whether the tension machine has entered the energy recovery mode, captures the energy of the tension machine and stores it in a supercapacitor first, and then stores the supercapacitor energy in a battery pack to form a battery energy storage unit. By working in coordination with the supercapacitor and the battery pack, it takes into account both instantaneous high-output power recovery and long-term energy storage, solves the problem of traditional reliance on large-capacity battery packs for power supply and insufficient battery life, and solves the problem of traditional single battery packs being susceptible to high voltage shocks, thereby effectively protecting the battery pack.

[0044] 2. This invention predicts the output power of the traction machine by constructing a model, considers the energy contribution priority of the tension machine, and uses a weighted priority algorithm to supply energy to the battery energy storage unit. This method supports multiple tension machines to power the traction machine at the same time, avoiding the delays and conflicts of centralized control and reducing the risk of battery overcharge / over-discharge. At the same time, the neural network model can effectively handle the nonlinear time-varying characteristics of the traction machine output power to ensure system reliability.

[0045] 3. This invention establishes an objective function when the traction machine supplies energy, uses a multi-strategy fusion sparrow search algorithm for optimization, and generates a dynamic battery energy distribution strategy. This algorithm overcomes the shortcomings of traditional algorithms that the population diversity gradually decreases and is prone to falling into local optimal solutions, improves the convergence speed and algorithm performance, dynamically optimizes the charging and discharging logic according to the output power demand, and provides an efficient and reliable energy distribution strategy for the traction equipment, thereby maximizing the energy recovery efficiency and extending the equipment life, which is conducive to the effective recycling of energy.

[0046] 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

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the invention, the following briefly introduces the drawings required for describing the embodiments. 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 any creative work.

[0048] Figure 1 The present invention provides a flow chart of a battery energy management method for an electric stretching device. DETAILED DESCRIPTION

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making any creative efforts shall fall within the scope of protection of the present invention.

[0050] Electric tensioning equipment suffers from insufficient endurance under complex operating conditions. Relying on large-capacity battery packs for power, frequent charging and discharging reduces battery life and increases maintenance costs. Meanwhile, energy from the tensioner is recovered for use in the traction machine. Centralized control lacks real-time and adaptability, leading to chaotic charging and discharging logic, potentially leading to redundant energy waste and the risk of battery overcharge or over-discharge, hindering effective energy recycling. The tensioner maintains tension on construction rigging, such as conductors and traction ropes, while they are deployed or pulled. Under passive operating conditions (where the conductor drives the tension pulley), the electric tensioner generates electricity through reverse motor rotation, converting mechanical energy into electrical energy, which is then stored in the battery pack. This process, based on Faraday's law of electromagnetic induction and the principle of conservation of energy, achieves effective energy recovery and storage through the motor and battery pack. By recovering and storing the electrical energy generated by the electric tensioner under passive operating conditions (i.e., tension conditions), it supplies the traction machine and other electric equipment, achieving effective energy recycling.

[0051] To solve the above technical problems, an embodiment of the present invention provides a battery energy management system for electric tensioning equipment. The purpose is to store the electrical energy generated by multiple electric tensioners when unwinding the conductors to provide power for the traction machines, as well as a control system for the entire electric tensioning equipment in the tension field. Specifically, it includes:

[0052] A battery energy recovery module, a battery energy supply module, a battery energy dynamic distribution module and a battery energy regulation module; the battery energy recovery module is used to determine whether the tension machine has entered the energy recovery mode, capture the energy of the tension machine and store it in the supercapacitor first, and then store the supercapacitor energy in the battery pack to form a battery energy storage unit; the battery energy supply module is used to predict the output power of the traction machine, obtain the predicted value of the traction machine output power, compare the output power of the battery energy storage unit with the predicted value of the traction machine output power, and enable the battery energy storage unit to supply energy to the traction machine; the battery energy dynamic distribution module is used to use a multi-strategy fusion sparrow search algorithm to optimize the current tolerance limit and voltage tolerance limit during energy supply when the battery energy storage unit supplies energy to the traction machine, and generate a battery energy dynamic distribution strategy; the battery energy regulation module is used to combine the battery energy dynamic distribution strategy and adjust the working state of the battery energy storage unit based on temperature-voltage closed-loop feedback.

[0053] In a specific embodiment, three electric tension machine prototypes without batteries, one electric small traction machine prototype without batteries, and one power battery container prototype (battery energy storage unit) can be used to realize that the three tension machines provide energy for the traction machine; among them, one electric tension machine prototype can generate about 40kWh of excess electricity per hour, and the three electric tension machine prototypes working simultaneously can generate about 120kWh of electricity per hour. The electric small traction machine prototype requires between 50-100kWh of electricity per hour during the traction process;

[0054] During the specific implementation of the above embodiment, first, the tension machine controller collects the tension machine load parameters and motor operating parameters, including cable tension, drum speed, motor torque, and motor speed, to determine whether the tension machine has entered the energy recovery mode. The captured tension machine energy is first stored in the supercapacitor, and then the supercapacitor energy is stored in the battery pack to form a battery energy storage unit. This method works in conjunction with the supercapacitor and the battery pack, using the supercapacitor to take into account instantaneous high output power recovery, solving the problem that the traditional single battery pack is susceptible to high voltage shock, effectively protecting the battery pack, and simultaneously recovering the energy of multiple tension machines to solve the problem of insufficient traditional endurance, thereby realizing energy recovery and storage. Secondly, a BiLSTM neural network model is constructed to predict the output power of the traction machine. Considering the energy contribution priority of the tension machine, a weighted priority algorithm is used in combination with the predicted output power value of the traction machine, and the battery energy storage unit collaboratively provides energy to realize the energy supply of the traction machine. This method supports multi-machine energy sharing, that is, supports multiple tension machines to power the traction machine at the same time, avoiding delays and conflicts in centralized control, reducing the risk of battery overcharge / over-discharge, and at the same time, the neural network model can effectively handle the non- When the traction machine is powered, an objective function is established based on maximum energy utilization. The algorithm integrates a chaotic mapping, sine-cosine algorithm, and an elite reverse learning strategy to optimize the current and voltage tolerance limits in the traction machine power supply using a multi-strategy fusion sparrow search algorithm. This ensures energy distribution within the allowable ranges of current and voltage, and generates a dynamic battery energy allocation strategy. This algorithm overcomes the shortcomings of traditional algorithms, such as the gradual decrease in population diversity and the susceptibility to local optimal solutions, by improving convergence speed and algorithm performance. It dynamically optimizes the charge and discharge logic, providing an efficient and reliable energy allocation strategy for the traction equipment, maximizing energy recovery efficiency and extending equipment life. The entire process efficiently searches for the optimal combination of current and voltage tolerance limits, ensuring that current and voltage are within a safe range, significantly improving energy transmission efficiency, and facilitating the effective recycling of energy. Finally, combined with the dynamic battery energy allocation strategy, the operating state of the battery energy storage unit is adjusted based on temperature-voltage closed-loop feedback to complete battery energy management. The temperature-voltage closed-loop feedback regulation mechanism avoids the risk of thermal runaway through real-time feedback and dynamic adjustment, ensuring a safe and reliable energy call process.

[0055] Further, in order to better introduce the technical solution of the embodiment of the present invention, based on the battery energy management system of the above-mentioned electric stretching device, such as Figure 1 As shown, an embodiment of the present invention provides a battery energy management method for an electric stretching device, which specifically includes the following contents:

[0056] S1. Obtain the load parameters of the tension machine and the operating parameters of the motor, determine whether the tension machine has entered the energy recovery mode, capture the energy of the tension machine and store it in the supercapacitor first, and then use the secondary energy storage method and layered protection mechanism to store the supercapacitor energy in the battery pack to form a battery energy storage unit;

[0057] Said S1 comprises the following steps:

[0058] S11. Obtaining tension machine load parameters and motor operating parameters, wherein the tension machine load parameters include cable tension and drum speed, and the motor operating parameters include motor torque and motor speed, to obtain tension machine operating data; then obtaining the time points of collecting the tension machine load parameters and motor operating parameters, generating a tension machine operating time series, finding tension machine operating data corresponding to the time points in the tension machine operating time series, and forming a tension machine operating data set; setting a time window, selecting time points separated by the time window in the tension machine operating time series, recording them as a first time point and a second time point, respectively, obtaining the tension machine operating data at the first time point, recording them as the first tension machine operating data, and then obtaining the tension machine operating data at the second time point, recording them as the second tension machine operating data;

[0059] S12: When the cable tension and the drum speed in the first tension machine operation data are both greater than 0 and the cable tension and the drum speed in the second tension machine operation data are both greater than 0, the tension machine enters the energy recovery mode, capturing the energy of the tension machine and storing it in the supercapacitor. The specific steps are as follows:

[0060] S121. Obtain the motor torque and motor speed from the first tension machine operating data and the second tension machine operating data. When the motor torque is a reverse torque output and the direction of the motor torque is opposite to that of the motor speed, the tension machine enters an active braking condition. Then, calculate the differential of the motor speed within the range from the first time point to the second time point, set a speed threshold, and when the motor torque is 0 and the differential of the motor speed within the range from the first time point to the second time point is less than the speed threshold, the tension machine enters a passive braking condition. The braking conditions of the tension machine operating data sets are determined in sequence.

[0061] S122. Obtain the battery pack capacity and set a first capacity threshold. When the battery pack capacity is greater than the first capacity threshold, capture the tension machine energy and store it in the supercapacitor. In the active braking condition, switch the motor to a power generation mode to charge the supercapacitor. In the passive braking condition, trigger protective braking, connect a parallel crowbar circuit for discharge, and then charge the supercapacitor.

[0062] S13. Setting a second capacity threshold. When the battery pack capacity is less than or equal to the first capacity threshold or the supercapacitor capacity is greater than the second capacity threshold, a secondary energy storage method is used to store supercapacitor energy in the battery pack, and a hierarchical protection mechanism is used to form a battery energy storage unit. The specific steps are as follows:

[0063] S131, collecting battery cell voltages in the battery pack to obtain a battery cell voltage set;

[0064] S132, setting a first protection voltage and a second protection voltage, when the battery cell voltage in the battery cell voltage set is less than the first protection voltage, the supercapacitor charges the battery pack normally; when the battery cell voltage in the battery cell voltage set is greater than or equal to the first protection voltage and less than or equal to the second protection voltage, reducing the charging current; when the battery cell voltage in the battery cell voltage set is greater than the second protection voltage, disconnecting the supercapacitor from charging the battery pack, thereby completing charging of the battery pack;

[0065] S133, combining the supercapacitor and the battery pack into a battery energy storage unit;

[0066] In this embodiment, the tension machine controller collects the tension machine load parameters and motor operating parameters to determine whether the tension machine enters the energy recovery mode, captures the tension machine energy and stores it in the supercapacitor first, and then stores the supercapacitor energy in the battery pack to form a battery energy storage unit; this method cooperates with the supercapacitor and the battery pack to use the supercapacitor to take into account instantaneous high output power recovery, solves the problem that the traditional single battery pack is susceptible to high voltage shock, effectively protects the battery pack, and at the same time recovers the energy of multiple tension machines to solve the problem of insufficient traditional endurance, thereby realizing energy recovery and storage; specifically, for example, the cable tension is collected: 1200N, the drum speed is 60rpm, the motor torque is -50N·m, the motor speed is -55rpm, the time window is set to 5 seconds, and two adjacent time points t1=10:00:00 and t2=10:00:05 are selected; at t1, the motor speed is When the torque is -50 N·m (<0) and the speed is -55 rpm (opposite to the torque direction), active braking is triggered, storing energy in the supercapacitor. The motor torque suddenly drops to 0, and the speed differential dω / dt = -60 rpm / s (exceeding the threshold of -50 rpm / s), triggering passive braking to release instantaneous energy. When the battery pack's current capacity is 85% (> the first capacity threshold of 80%), the supercapacitor is prioritized for energy storage. When the supercapacitor capacity reaches 95% (> the second capacity threshold of 90%), secondary storage is activated. Battery cell voltage detection: Cell 1 voltage is 3.7V (< the first protection voltage of 3.6V), normal charging (current 20A), cell 2 voltage is 4.0V (between 3.6V and 4.2V), current is reduced to 10A, and cell 3 voltage is 4.3V (> the second protection voltage of 4.2V), disconnecting the charging circuit. Energy storage is achieved through the energy recovery system, improving energy utilization.

[0067] S2. Predict the output power of the traction machine based on a neural network and output the predicted output power value of the traction machine. Then, considering the energy contribution priority of the tension machine, a weighted priority algorithm is used in combination with the predicted output power value of the traction machine. The battery energy storage unit synergistically provides energy to realize the energy supply of the traction machine.

[0068] The S2 comprises the following steps:

[0069] S21, obtain the motor current, motor voltage, motor speed, motor temperature, load weight and traction speed of the traction machine to obtain the traction machine operation data, then obtain the time point of collecting the traction machine operation data, obtain the traction machine operation time series, and form the traction machine operation data set; set the sliding window size to , place the sliding window in the tractor operation data set, and select The traction machine operation data corresponding to each time point is combined into a feature sequence, and then normalized to generate a traction machine operation data sample set;

[0070] S22. Train the BiLSTM neural network to obtain a traction machine output power prediction model, and output a traction machine output power prediction value based on the sample data set. The specific steps are as follows:

[0071] S221. Set a BiLSTM neural network including an input layer, a bidirectional LSTM layer, an attention mechanism layer, a dense layer, and an output layer, with an initial learning rate of 0.001; obtain historical tractor operation data, extract a feature sequence of the historical tractor operation data, and perform normalization processing to obtain a sample data set, divide the sample data set into a sample training set and a sample test set, and then input the sample training set into the BiLSTM neural network for iteration until the BiLSTM neural network converges to obtain a trained BiLSTM neural network;

[0072] S222. Input the sample test set into the trained BiLSTM neural network, output a prediction result, set an accuracy threshold, and obtain a tractor output power prediction model when the prediction result accuracy is greater than the accuracy threshold. Otherwise, adjust the weight until the prediction result accuracy is greater than the accuracy threshold. Input the tractor operation data sample set into the tractor output power prediction model, and output a tractor output power prediction value.

[0073] S23, the m The distance between the tension machine and the traction machine is recorded as , the transmission loss is b , then m Path loss of a tension machine , set the m The rated energy generated by the first tension machine is obtained m The remaining energy generated by the first tension machine is used to m The remaining energy generated by the first tension machine and the m The quotient of the rated energy generated by the first tension machine is recorded as m Residual energy factor of a tension machine , using the weighted priority algorithm to calculate the m The priority score of each tension machine is calculated as follows:

[0074] ;

[0075] in, Indicates the m Priority score of the tension machine;

[0076] Calculate the priority scores of all tension machines in sequence and sort them in descending order. Select the tension machine with the highest priority score and record it as the priority tension machine. Compare the predicted output power value of the traction machine with the output power of the battery pack in the battery energy storage unit. When the predicted output power value of the traction machine is less than the output power of the battery pack in the battery energy storage unit, the battery pack provides energy. Otherwise, use the priority tension machine to charge the supercapacitor, and the supercapacitor cooperates with the battery pack to provide energy to complete the traction machine energy supply.

[0077] In this embodiment, the output power of the traction machine is predicted by building a model, considering the energy contribution priority of the tension machine, using a weighted priority algorithm combined with the predicted value of the traction machine output power, and the battery energy storage unit collaboratively provides energy to realize the traction machine energy supply. This method supports multiple tension machines to power the traction machine at the same time, avoiding delays and conflicts in centralized control, and reducing the risk of battery overcharge / over-discharge. At the same time, the neural network model can effectively handle the nonlinear time-varying characteristics of the traction machine output power to ensure system reliability. Specifically, for example, traction machine data is collected every 10 seconds, and the collection is continued for 24 hours to form 8640 sets of time series data, and the window size is set. 15 (2.5 minutes of data), with a window sliding step of 5 seconds, generated 1728 feature sequences. After normalization, these sequences were input into the tractor output power prediction model, which output a predicted tractor output power value of 58.7 kW. The three tensioners T1, T2, and T3 are located at distances of 0.8 km, 1.2 km, and 0.5 km, respectively. Their residual energy factors are 0.9, 0.6, and 0.3, respectively, and their priority scores are 3.68, 2.92, and 6.67, respectively. Tensioner T3 is selected as the priority. The supercapacitor provides a peak output power of 10 kW, and the battery reduces the load to 48.7 kW, providing a total energy of 48.7 + 10 = 58.7 kW.

[0078] S3. When the traction machine is supplying energy, an objective function is established based on the maximum energy utilization rate, and a multi-strategy fusion sparrow search algorithm is used to optimize the current tolerance limit and voltage tolerance limit in the traction machine power supply to generate a dynamic battery energy allocation strategy;

[0079] The S3 includes the following steps:

[0080] S31: After the traction machine supplies energy, the transmission current, transmission resistance, internal resistance and voltage of the battery pack in the battery energy storage unit are obtained, and the transmission resistance loss is calculated. ,in represents the transmission current, Represents the transmission resistance, and then calculates the output power of the battery energy storage unit ,in V Represents the battery pack voltage in the battery energy storage unit, represents the internal resistance of the battery pack in the battery energy storage unit, and takes the maximum difference between the output power of the battery energy storage unit and the transmission resistance loss as the objective function;

[0081] S32. Setting a current tolerance limit and a voltage tolerance limit to constrain the range of the transmission current and the battery pack voltage in the battery energy storage unit, improving the sparrow search algorithm, and obtaining a multi-strategy fusion sparrow search algorithm; in the traction machine energy supply process, using the multi-strategy fusion sparrow search algorithm to optimize the current tolerance limit and the voltage tolerance limit, and obtaining an optimized current tolerance limit and an optimized voltage tolerance limit. The specific steps are as follows:

[0082] S321. The objective function is regarded as the fitness function, and the process of solving the objective function is regarded as the search space. There is a sparrow population in the search space, and the number of sparrow populations is p , the sparrow population dimension is q , the current tolerance limit and the voltage tolerance limit are combined into a feasible solution, the individual position of the sparrow in the sparrow population represents the feasible solution, and the process of iterating the individual position of the sparrow is regarded as the process of optimizing the current tolerance limit and the voltage tolerance limit; the sparrow population is divided into discoverers and joiners, the population is initialized using chaotic mapping, and then the elite reverse learning strategy is introduced to select the initial best position and the initial worst position in the sparrow population, and set represents a random number between the interval [0, 1], Indicates the first i The individual position of the sparrow, calculate the first i The reverse solution of the individual positions of sparrows ,in represents the best individual sparrow position in the sparrow population, Indicates the position of the worst individual sparrow in the sparrow population;

[0083] S322, set the current number of iterations to t , the maximum number of iterations is T , 、 and represents a random number between the interval (0, 1], A random number representing the warning value between the interval [0, 1], A random number representing a safe value between the interval [0.5, 1], Indicates the t At the first iteration, the number of sparrows in the population i Sparrow individual j Dimensional position, find the t The best sparrow individual in the sparrow population at the iteration is denoted as ; Introduce the sine-cosine algorithm to update the discoverer's position. When t +1 iteration in the sparrow population i Sparrow individual j Dimensional Position ,when When you get the position ;

[0084] S323, looking for t The best sparrow individual in the sparrow population at the +1 iteration is recorded as , No. t The worst individual sparrow in the sparrow population at the +1 iteration is denoted as ,set up represents a random number between the interval [-1, 1], C Indicates a random number that obeys the standard normal distribution; at this time, the position of the joiner is updated. When you get the position ,when When you get the position ; Calculate the current best fitness function value and the current worst fitness function value, control the sparrow population to move in the direction of the sparrow individual corresponding to the current best fitness function value, complete the position update of all current stages, and generate the next generation of sparrow population; until the current number of iterations reaches the maximum number of iterations, stop the iteration, obtain the final sparrow population, find the sparrow individual position corresponding to the best fitness function value, and obtain the optimized current tolerance limit and the optimized voltage tolerance limit;

[0085] S33, during the traction machine energy supply, the transmission current is maintained within the optimized current tolerance range, and the battery pack voltage in the battery energy storage unit is maintained within the optimized voltage tolerance range, the battery energy storage unit provides energy for the traction machine, and a dynamic battery energy allocation strategy is generated;

[0086] In this embodiment, by establishing an objective function based on the maximum energy utilization rate when the traction machine is supplying energy, and integrating chaotic mapping, sine-cosine algorithm and elite reverse learning strategy, the sparrow search algorithm with multi-strategy fusion is used to optimize the current tolerance limit and voltage tolerance limit in the traction machine power supply, and generate a dynamic battery energy distribution strategy, which overcomes the shortcomings of the traditional algorithm that the population diversity gradually decreases and is prone to falling into the local optimal solution, improves the convergence speed and algorithm performance; dynamically optimizes the charging and discharging logic, provides an efficient and reliable energy distribution strategy for the traction equipment, and realizes the maximization of energy recovery efficiency and the extension of equipment life; the whole process can efficiently search for the optimal combination of current and voltage tolerance limits, ensuring that the current and voltage are within a safe range The energy transmission efficiency is significantly improved within the range, which is conducive to the effective recycling of energy. Specifically, for example, the current tolerance limit constraint ∈ [1%, 10%] (initial range), the voltage tolerance limit constraint ∈ [1%, 5%] (initial range), the sparrow population size p = 50, the dimension q = 2 (current and voltage tolerance limits), the maximum number of iterations T = 100, the chaotic mapping uses the Logistic mapping, the elite reverse learning strategy introduces the reverse solution, the initial sparrow individual position is set to [8%, 4%], the transmission current is 98A, the battery pack voltage in the battery energy storage unit is 45.5V, the battery pack internal resistance in the battery energy storage unit is 0.05Ω, the transmission resistance is 0.1Ω, and the energy utilization rate is Update individual sparrows in the sparrow population. Finder update: When 0.3 < 0.8, use sinusoidal update. Joiner update: When 20 ≤ p / 2 = 25, update around the optimal position. Complete this iteration. After the maximum number of iterations, converge to [3.2%, 1.8%]. At this point, the energy utilization rate is 94%. The optimization of the current and voltage tolerance limits before and after improves energy utilization while balancing efficiency and equipment life.

[0087] S4. In combination with the battery energy dynamic allocation strategy, the working state of the battery energy storage unit is adjusted based on the temperature-voltage closed-loop feedback to complete the battery energy management;

[0088] The S4 comprises the following steps:

[0089] S41, running the battery energy dynamic allocation strategy, the battery energy storage unit supplies power to the traction machine, collecting the battery pack temperature and battery pack voltage in the battery energy storage unit, obtaining a temperature-voltage array, and then calculating the average temperature of the battery pack in the battery energy storage unit. ; Set the temperature coefficient to , the standard voltage at temperature u is , the battery pack temperature in the battery energy storage unit at temperature u is , at this time the dynamic voltage threshold ;

[0090] S42. When the battery energy storage unit is powered, a temperature threshold is set. When the voltage of the battery pack in the battery energy storage unit is less than the dynamic voltage threshold, the power supply is stopped. When the temperature of the battery pack in the battery energy storage unit is greater than the temperature threshold, the power supply current is reduced to form a closed-loop feedback regulation to complete battery energy management.

[0091] In this embodiment, by combining a dynamic battery energy allocation strategy and adjusting the operating state of the battery energy storage unit based on temperature-voltage closed-loop feedback, real-time feedback and dynamic adjustment are implemented to avoid the risk of thermal runaway, extend battery life, ensure a safe and reliable energy call process, and complete battery energy management. Specifically, for example, a temperature coefficient of 0.05V / °C (assuming an empirical value) is set, the current ambient temperature is 25°C, the standard battery voltage at this temperature is 3.7V, the temperature threshold is set to 50°C, the collected temperatures are 25°C, 28°C, and 31°C, the average temperature is 28°C, and the current monitoring point temperature is 25°C. The dynamic voltage threshold is calculated to be 3.55V. If the voltage of a single cell drops to 3.5V (<3.55V), the system immediately stops power supply to prevent over-discharge. If the temperature rises to 52°C (>50°C), the system automatically reduces the supply current from 100A to 70A and dissipates heat through the cooling system.

[0092] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0093] The preferred embodiments of the invention disclosed above are intended only to help illustrate the invention. These preferred embodiments do not exhaust all details, nor do they limit the invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the content of this specification. These embodiments are selected and described in detail in this specification to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention.

Claims

1. A battery energy management system for an electric tensioning device, characterized in that: include: The battery energy recovery module is used to determine whether the tension machine has entered the energy recovery mode, capture the energy of the tension machine and store it in the supercapacitor first, and then store the supercapacitor energy in the battery pack to form a battery energy storage unit; The battery energy supply module is used to predict the output power of the traction machine, obtain the predicted value of the traction machine output power, compare the output power of the battery energy storage unit with the predicted value of the traction machine output power, and enable the battery energy storage unit to supply energy to the traction machine; A battery energy dynamic allocation module is used to optimize the current tolerance limit and the voltage tolerance limit during energy supply by using a multi-strategy fusion sparrow search algorithm when the battery energy storage unit supplies energy to the traction machine, and generate a battery energy dynamic allocation strategy; The battery energy regulation module is used to combine the battery energy dynamic allocation strategy and regulate the working state of the battery energy storage unit based on the temperature-voltage closed-loop feedback.

2. A battery energy management system for electric stretching equipment according to claim 1, characterized in that: Determining whether the tension machine enters the energy recovery mode includes: Obtaining tension machine load parameters and motor operating parameters, wherein the tension machine load parameters include cable tension and drum speed, and the motor operating parameters include motor torque and motor speed, obtaining tension machine operating data, recording the tension machine operating time series, and selecting a first time point and a second time point; When the cable tension and the drum speed in the first tension machine operation data are both greater than 0 and the cable tension and the drum speed in the second tension machine operation data are both greater than 0, the tension machine enters the energy recovery mode.

3. A battery energy management system for electric stretching equipment according to claim 2, characterized in that: The energy of the capture tension machine is first stored in the super capacitor, comprising: Obtain the motor torque and motor speed from the operating data of the first tension machine and the operating data of the second tension machine. When the motor torque is a reverse torque output and the direction of the motor torque is opposite to that of the motor speed, the tension machine enters an active braking condition. Then, calculate the differential of the motor speed from the first time point to the second time point, set a speed threshold, and when the motor torque is 0 and the differential of the motor speed from the first time point to the second time point is less than the speed threshold, the tension machine enters a passive braking condition. Obtaining the capacity of the battery pack, setting a first capacity threshold, and storing the energy of the capture tension machine in the supercapacitor when the battery pack capacity is greater than the first capacity threshold; In the active braking mode, the motor is switched to the power generation mode to charge the supercapacitor; in the passive braking mode, protective braking is triggered, the parallel crowbar circuit is discharged, and then the supercapacitor is charged.

4. A battery energy management system for electric stretching equipment according to claim 3, characterized in that: The supercapacitor energy is then stored in a battery pack, and the battery energy storage unit comprises: Setting a second capacity threshold, when the battery pack capacity is less than or equal to the first capacity threshold or the supercapacitor capacity is greater than the second capacity threshold, a secondary energy storage method is used to store supercapacitor energy in the battery pack; Collecting the voltages of battery cells in the battery pack to obtain a battery cell voltage set; Setting a first protection voltage and a second protection voltage, when the battery cell voltage in the battery cell voltage set is lower than the first protection voltage, the supercapacitor charges the battery pack normally; When the battery cell voltage in the battery cell voltage set is greater than or equal to the first protection voltage and less than or equal to the second protection voltage, reducing the charging current; When the battery cell voltage in the battery cell voltage set is greater than the second protection voltage, disconnecting the supercapacitor to charge the battery pack, thereby completing the charging of the battery pack; The supercapacitor and battery pack form a battery energy storage unit.

5. A battery energy management system for electric stretching equipment according to claim 4, characterized in that: The predicted output power of the traction machine and the output predicted value of the traction machine output power include: Obtain the motor current, motor voltage, motor speed, motor temperature, load weight and traction speed of the traction machine to generate a traction machine operation data sample set; The BiLSTM neural network is configured to include an input layer, a bidirectional LSTM layer, an attention mechanism layer, a dense layer, and an output layer, with an initial learning rate of 0.

001. Historical tractor operation data is obtained, and feature sequences of the historical tractor operation data are extracted and normalized to obtain a sample data set. The sample data set is divided into a sample training set and a sample test set. The sample training set is then input into the BiLSTM neural network for iteration until the BiLSTM neural network converges, thereby obtaining a trained BiLSTM neural network. Then, the sample test set is input into the trained BiLSTM neural network, the prediction result is output, and the accuracy threshold is set. When the accuracy of the prediction result is greater than the accuracy threshold, the traction machine output power prediction model is obtained. Otherwise, the weight is adjusted until the accuracy of the prediction result is greater than the accuracy threshold; the traction machine operation data sample set is input into the traction machine output power prediction model, and the traction machine output power prediction value is output.

6. A battery energy management system for electric stretching equipment according to claim 5, characterized in that: The comparing the output power of the battery energy storage unit and the predicted value of the traction machine output power to enable the battery energy storage unit to supply energy to the traction machine includes: Calculate the m The path loss and residual energy factor of the tension machine are calculated using the weighted priority algorithm. m The priority score of each tension machine is calculated; the priority scores of all tension machines are calculated in sequence, and the tension machine with the highest priority score is selected and recorded as the priority tension machine; The predicted output power value of the traction machine is compared with the output power of the battery pack in the battery energy storage unit. When the predicted output power value of the traction machine is less than the output power of the battery pack in the battery energy storage unit, the battery pack provides energy; otherwise, the priority tension machine is used to charge the supercapacitor, and the supercapacitor cooperates with the battery pack to provide energy to complete the energy supply of the traction machine.

7. A battery energy management system for electric stretching equipment according to claim 6, characterized in that: The method of using the multi-strategy fusion sparrow search algorithm to optimize the current tolerance limit and the voltage tolerance limit during energy supply includes: After the traction machine is powered, the transmission current, transmission resistance, internal resistance of the battery pack in the battery energy storage unit, and voltage are obtained. The transmission resistance loss and the output power of the battery energy storage unit are calculated, and the maximum difference between the output power of the battery energy storage unit and the transmission resistance loss is used as the objective function. The current tolerance limit and the voltage tolerance limit are set to constrain the range of the transmission current and the battery pack voltage in the battery energy storage unit; the objective function is used as the fitness function, and the process of solving the objective function is regarded as a search space. There is a sparrow population in the search space, and the current tolerance limit and the voltage tolerance limit are combined into a feasible solution. The individual position of the sparrow in the sparrow population represents the feasible solution, and the process of iterating the individual position of the sparrow is regarded as a process of optimizing the current tolerance limit and the voltage tolerance limit; the sparrow population is divided into discoverers and joiners, and the population is initialized using chaotic mapping, and then the elite reverse learning strategy is introduced to calculate the number of sparrows in the sparrow population. i The reverse solution of the individual positions of sparrows; The sine-cosine algorithm is introduced to update the position of the discoverer and then the position of the joiner; the current best fitness function value and the current worst fitness function value are calculated, and the sparrow population is controlled to move in the direction of the sparrow individual corresponding to the current best fitness function value, completing the position update of all current stages and generating the next generation of sparrow population; until the current number of iterations reaches the maximum number of iterations, the iteration is stopped to obtain the final sparrow population, and the position of the sparrow individual corresponding to the best fitness function value is found to obtain the optimized current tolerance limit and the optimized voltage tolerance limit.

8. A battery energy management system for electric stretching equipment according to claim 7, characterized in that: The generation of a dynamic battery energy allocation strategy includes: During the traction machine power supply, the transmission current is kept within the optimized current tolerance limit, and the battery pack voltage in the battery energy storage unit is within the optimized voltage tolerance limit. The battery energy storage unit provides energy for the traction machine and generates a dynamic battery energy allocation strategy.

9. A battery energy management system for electric stretching equipment according to claim 8, characterized in that: The method of adjusting the working state of the battery energy storage unit based on the temperature-voltage closed-loop feedback to complete the battery energy management includes: Running the battery energy dynamic allocation strategy, the battery energy storage unit supplies power to the traction machine, collecting the battery pack temperature and battery pack voltage in the battery energy storage unit, and calculating the dynamic voltage threshold; When the battery energy storage unit is powered, a temperature threshold is set. When the battery pack voltage in the battery energy storage unit is lower than the dynamic voltage threshold, the power supply is stopped. When the battery pack temperature in the battery energy storage unit is higher than the temperature threshold, the power supply current is reduced to form a closed-loop feedback regulation to complete battery energy management.

Images