Hybrid power energy-saving control method suitable for mining equipment

Abstract

The invention discloses a hybrid power energy-saving control method suitable for mining equipment, and belongs to the technical field of mining equipment energy-saving control, and the method comprises the following steps: collecting real-time operation state data, including load power, gradient, speed and battery SOC, of the mining equipment through multiple sensors; based on the collected real-time operation data and historical data, predicting a load demand in a future time window by using a time sequence prediction model; according to the current load prediction result, the power distribution proportion of the internal combustion engine and the motor is dynamically adjusted, so that the hybrid power system meets the operation requirement; and in the deceleration or idling stage of the equipment, the braking energy is recovered through reverse working of the motor, and the recovered energy is stored in the battery or the super capacitor. Through dynamic load prediction, energy distribution optimization and flexible energy recovery, efficient energy saving of mining equipment and reasonable utilization of energy of the internal combustion engine and the motor under complex working conditions are achieved.

Description

Technical Field

[0001] The present invention relates to a hybrid power energy-saving control method applicable to mining equipment, and belongs to the technical field of energy-saving control of mining equipment. Background Art

[0002] In modern mining operations, the energy consumption and operating efficiency of equipment directly affect the production cost and environmental impact of mines. With the continuous rise of energy costs and the increasing strictness of environmental protection policies, traditional high-energy-consuming equipment has become difficult to meet the actual needs. Hybrid power technology has become an important direction for the optimization and upgrading of mining equipment due to its significant advantages in reducing fuel consumption and emissions.

[0003] In the prior art, some control methods based on hybrid power technology have been developed to improve energy utilization efficiency. These technologies can reduce the fuel consumption of internal combustion engines to a certain extent and reduce emissions under relatively stable load conditions through fixed energy distribution strategies and static rule adjustment methods. In addition, the application of some energy recovery technologies can also convert mechanical energy into electrical energy during braking or deceleration stages to achieve partial energy recovery and reuse. These methods have achieved certain results in improving the local operating efficiency of equipment.

[0004] However, the prior art still has deficiencies; firstly, current technologies mostly adopt fixed power distribution strategies or adjustment methods based on static rules, and have poor adaptability to load fluctuations in complex mining conditions, making it difficult to effectively respond to real-time changing energy demands; secondly, the load prediction accuracy is not high, and it cannot provide reliable guidance for dynamic energy management, resulting in a further reduction in energy utilization efficiency; in addition, the existing energy recovery technologies lack flexible distribution and utilization mechanisms and fail to fully utilize the value of the recovered energy. Summary of the Invention

[0005] The purpose of the present invention is to overcome the deficiencies of the above-mentioned prior art and provide a hybrid power energy-saving control method applicable to mining equipment.

[0006] The technical solution provided by the present invention is as follows: A hybrid power energy-saving control method applicable to mining equipment, characterized by comprising the following steps:

[0007] S1, Working condition perception: Collect real-time operating state data of mining equipment through multiple sensors, including load power, slope, speed, and battery SOC;

[0008] S2, Load prediction: Based on the collected real-time operating data and historical data, use a time series prediction model to predict the load demand within a future time window;

[0009] S3. Dynamic energy management: Dynamically adjust the power distribution ratio between the internal combustion engine and the electric motor according to the current load prediction results, so that the hybrid power system meets the operation requirements;

[0010] S4. Energy recovery: During the deceleration or idling stage of the equipment, recover the braking energy through the reverse operation of the electric motor, and store the recovered energy in the battery or supercapacitor;

[0011] S5. Energy distribution: Dynamically distribute the recovered energy based on the current task priority and working condition requirements, and give priority to meeting the high-priority tasks;

[0012] S6. Multi-device collaboration: Through task allocation and energy sharing optimization, achieve collaborative scheduling of multiple devices and overall energy conservation.

[0013] Preferably, the working condition perception includes:

[0014] Obtain the current load power data of the equipment through a load sensor;

[0015] Real-time monitor the terrain slope of the mine through a slope sensor;

[0016] Record the running speed of the equipment through a speed sensor;

[0017] Obtain the battery state data through a battery SOC sensor;

[0018] Use the Kalman filtering algorithm to fuse the multi-sensor data, eliminate noise, and improve the data accuracy.

[0019] Preferably, the load prediction includes:

[0020] Use the historical load power, running speed, terrain slope, and battery SOC as input features;

[0021] Use the long short-term memory network model to process the time series working condition data;

[0022] Capture the long-term and short-term dependencies of the load data through the network structure;

[0023] Output the load demand within the future time window, including the future load power demand, battery SOC change, and running speed change.

[0024] Preferably, the dynamic energy management includes:

[0025] Construct an energy balance model to determine the power distribution ratio between the internal combustion engine and the electric motor;

[0026] Under high-load working conditions, give priority to increasing the output power of the internal combustion engine and reducing the power demand of the electric motor;

[0027] Under low-load conditions, the motor power is preferentially used to reduce the power output of the internal combustion engine;

[0028] The power ratio of the internal combustion engine and the motor is dynamically adjusted through an optimization algorithm to achieve the dynamic energy balance of the hybrid system.

[0029] Preferably, the optimization algorithm includes:

[0030] Define an optimization objective function to minimize the fuel consumption cost and the battery loss cost;

[0031] Set constraint conditions, including the power ranges of the internal combustion engine and the motor and the safe range of the battery SOC;

[0032] Use a reinforcement learning algorithm to optimize the power distribution strategy and adjust the power output ratio of the internal combustion engine and the motor in real time;

[0033] While meeting the equipment load demand, ensure that the battery SOC is maintained within the safe range.

[0034] Preferably, the optimization objective is:

[0035] Minimize the fuel consumption and battery loss of the hybrid system on the premise of meeting the equipment load demand;

[0036] Dynamically adjust the task allocation and energy sharing strategy to optimize the overall energy-saving effect of the equipment group.

[0037] Preferably, the energy recovery includes:

[0038] During the deceleration, braking or idling stage of the equipment, convert mechanical energy into electrical energy through the reverse operation of the motor;

[0039] Dynamically adjust the braking current of the motor to improve the energy recovery efficiency;

[0040] Store the recovered energy in the battery or supercapacitor through the energy management system;

[0041] Dynamically adjust the storage location of the energy recovery according to the battery SOC state.

[0042] Preferably, the energy distribution includes:

[0043] Dynamically adjust the distribution ratio of the recovered energy according to the priority of the equipment tasks;

[0044] Calculate the distribution ratio of the task energy demand through the priority weight model;

[0045] High-priority tasks include starting and large-load climbing conditions, and the recovered energy is preferentially allocated to meet the instantaneous high-power demand;

[0046] The energy requirements of low-priority tasks are met by the remaining recovered energy or the backup power of the battery.

[0047] Preferably, the multi-device collaboration includes:

[0048] Obtaining the load status, task allocation information, and battery SOC status of each device through communication between devices;

[0049] Optimizing the task allocation of multiple devices using a scheduling algorithm;

[0050] Dynamically adjusting the task priority and scheduling scheme by real-time monitoring of the status changes between devices;

[0051] Minimizing the total energy consumption of the entire device group under the condition of meeting the task completion time.

[0052] Preferably, the energy sharing includes:

[0053] Dynamically allocating shared energy between high-SOC devices and low-SOC devices;

[0054] The sharing energy allocation ratio is dynamically adjusted according to the task priority and physical distance between devices;

[0055] Realizing the real-time transmission of shared energy through the communication protocol between devices;

[0056] Guaranteeing the energy requirements of devices with urgent tasks first.

[0057] The beneficial effects of the present invention are as follows: The present invention solves the problems in the prior art of poor adaptability to load changes in complex mine working conditions, lack of dynamic adjustment ability in energy distribution, and inflexible energy recovery and distribution mechanism.

[0058] 1. The present invention adopts a load prediction technical solution based on a long short-term memory network, combines multi-dimensional time series data such as historical load, operating speed, slope, and battery SOC, realizes the ability to dynamically predict future load requirements, and avoids the deficiency of slow response to load fluctuations by static rules in the prior art through accurately capturing the load change trend in complex working conditions, effectively improving the accuracy and real-time performance of load prediction.

[0059] 2. The present invention, through a dynamic energy management module, combines an energy balance model and a reinforcement learning algorithm to adjust the power distribution ratio between the internal combustion engine and the motor in real time, minimizes fuel consumption and battery loss while meeting the device load requirements. Compared with the traditional fixed power distribution strategy, this technology enables the hybrid power system to have stronger adaptability to complex working conditions and significantly improves the energy utilization efficiency.

[0060] 3. The present invention proposes an energy allocation technical solution based on task priorities. By constructing a priority weight model to dynamically adjust the energy allocation ratio, sufficient energy support is provided for high-priority tasks, solving the problems of uneven energy allocation or insufficient energy supply for tasks in the prior art. This solution can also reduce energy waste, make the device operate more efficiently, and meet the diverse energy requirements of different tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 It is a flowchart of the method of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0062] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the specification of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0063] As Figure 1 shown, a hybrid power energy-saving control method applicable to mining equipment includes the following steps:

[0064] S1. Working condition perception: Collect real-time operation state data of mining equipment through multiple sensors, including load power, slope, speed, and battery SOC;

[0065] Combined with the complex working environment of mining equipment, in this step, multiple sensors are deployed and data fusion technology is combined to accurately obtain key parameters such as equipment load power, slope, speed, and battery SOC, so as to effectively cope with the real-time monitoring challenges brought by the complexity of working conditions. Generally, the data output of this step will be directly transmitted to the load prediction module to provide reliable input for the intelligence of energy management;

[0066] In a possible implementation, multiple types of sensors are arranged to monitor the operation state of mining equipment in real time. Specifically:

[0067] The load sensor is installed near the power output device and is used to collect the current load power P of the mining equipment in real time load . The obtained load data can reflect the current operation power demand of the mining equipment, which is of great significance for subsequent load prediction.

[0068] The slope sensor is arranged on the vehicle chassis or key nodes of the equipment and is used to detect the slope information θ in the mining environment in real time. This parameter is an important factor affecting the load demand. For example, when going uphill, the load power demand often increases significantly, while when going downhill, the load demand will decrease.

[0069] The speed sensor is installed on the vehicle transmission device and is used to record the operating speed v of the device. The change in the device's operating speed is usually closely related to the load fluctuation. Through the speed information, the operating state of the device can be inferred, such as whether it is in the acceleration, deceleration, or constant speed stage.

[0070] The battery SOC sensor is used to monitor the current state of the device's battery, specifically the state of charge of the battery. SOC is used to reflect the remaining battery power. Its calculation formula is:

[0071]

[0072] Among them, E remaining is the current remaining capacity of the battery, and E max is the maximum capacity of the battery;

[0073] In some embodiments, in order to improve the accuracy of the data, for the data collected by the above sensors, the present invention introduces the Kalman filtering algorithm to fuse the multi-dimensional data. Generally, there are certain time deviations and noises in the multi-sensor data sources, and direct use may affect the performance of subsequent modules. Through Kalman filtering, the weights can be dynamically adjusted, thereby generating more accurate operating condition data.

[0074] Specifically, the Kalman filtering processing steps are as follows:

[0075] The sensor data collected is input into the filtering model through the Kalman state equation:

[0076]

[0077] Among them, represents the state estimate value after fusion at the current moment, z t is the real-time observation value of the sensor, H is the state transition matrix, and K t is the dynamically adjusted Kalman gain; represents the state estimate value at the previous moment (time t-1), that is, the result of the previous filtering process.

[0078] The calculation formula of the Kalman gain is:

[0079] K t = P t H T (HP t H T + R) -1 ;

[0080] Among them, P t is the prediction error covariance matrix, R is the observation noise covariance; H T is the transpose of the state transition matrix H. H is the matrix that maps the state space to the observation space, HT is the transpose of this matrix, which is often used to relate the error in the observation space to the prediction error;

[0081] The final output of data fusion includes load power P load , slope θ, speed v, battery SOC, all data have high time consistency and accuracy. These data are directly input into the load prediction module;

[0082] As an option, the Kalman filter can be dynamically adjusted by adjusting the parameter P t and R to meet the requirements of data fusion accuracy in different mining conditions. For example, when the mining environment is noisy, increase the value of R to improve the noise tolerance of sensor data; and in scenarios with low noise but severe load fluctuations, reduce P t , enhancing the sensitivity of the filter to rapid changes.

[0083] In some embodiments, in order to meet the needs of different equipment types, the installation position of the load sensor can be adjusted according to the power output mode of the equipment. For example, for a tracked equipment, the load sensor can be installed near the drive wheel, while for a wheeled equipment, it can be installed on the drive shaft. This can obtain load data more accurately.

[0084] In addition, in order to improve the adaptability of the system, the present embodiment can also preset multiple sensor calibration models for different types of mining equipment. For example, for sensors with different slopes, calibration curves can be established according to the mine terrain characteristics to correct the nonlinear error of the sensor.

[0085] In a possible implementation, the output data of this embodiment will be automatically transmitted to the load prediction module. load , θ, v and SOC data are updated at fixed time intervals to provide continuous time series input for the subsequent load prediction model.

[0086] S2. Load forecasting: Based on the collected real-time operation status data and historical data, the time series forecasting model is used to predict the load demand in the future time window;

[0087] This step uses the time series prediction model to predict future load demand. Generally speaking, this prediction can effectively solve the problem of frequent load fluctuations and high real-time requirements in mining conditions, and provide a decision-making basis for subsequent energy management;

[0088] In one possible implementation, the load forecasting model is mainly built based on the long short-term memory network (LSTM) model, combining historical operating data and real-time sensor data for forecasting. Specifically:

[0089] First, the input features of the prediction model include the following:

[0090] The load power P collected in real time load , which is provided by the load sensor in step S1 and is used to characterize the current operating power demand of the device;

[0091] The operating speed v of the device, which is used to reflect the dynamic state of the device, such as acceleration or deceleration;

[0092] The slope information θ of the mine terrain, which is collected by the slope sensor and is a key environmental variable determining the load change;

[0093] The SOC value of the battery, which is used to indicate the current state of charge of the battery and helps predict the energy demand of the device.

[0094] In some embodiments, the historical time window is set to the data of the most recent T h = 30 seconds, including the above features of multiple time steps. These data are organized in the form of a time series and used as the input of the prediction model.

[0095] In the design of the model, the LSTM network structure consists of three parts: an input layer, a hidden layer, and an output layer.

[0096] The input layer receives the time series features and maps them into a high-dimensional feature representation.

[0097] The hidden layer consists of two layers of LSTM cells and can capture long-term and short-term dependencies. For example, the slow growth of the load demand during continuous climbing of the device and the sharp drop in the load when suddenly turning to flat ground, these complex dynamic characteristics can be captured by the LSTM;

[0098] Specifically, the state update formula in the hidden layer of the LSTM is as follows:

[0099] f t = σ(W f ·[h t-1 ,x t +b f )

[0100] i t = σ(W i ·[h t-1 ,x t +b i )

[0101]

[0102] h t = o t ·tanh(C t )

[0103] Among them: f t is the activation value of the forget gate; i t is the activation value of the input gate; is the candidate cell state; C t is the current cell state; h t is the current hidden state; W f , W i , W C is the weight matrix of the model, W f is the weight matrix of the forget gate, which maps the previous hidden state and the current input feature to the activation value of the forget gate, W i is the weight matrix of the input gate, which maps the previous hidden state and the current input feature to the activation value of the input gate, W C is the weight matrix of the candidate cell state, which maps the previous hidden state and the current input feature to the candidate cell state; b f , b i , b C is the bias term; is the candidate cell state, representing the new potential information generated at the current moment, b f is the bias term of the forget gate, which, together with the result of the linear combination of the inputs of the forget gate, determines the activation value of the forget gate, b i is the bias term of the input gate, which, together with the result of the linear combination of the inputs of the input gate, determines the activation value of the input gate, b C is the bias term of the candidate cell state, which, together with the result of the linear combination of the inputs of the candidate cell state, determines the output of the candidate cell state; h t-1 is the hidden state at the previous moment (t - 1); o t is the activation value of the output gate; σ represents the Sigmoid activation function, which is used to calculate the activation value of the gate; tanh represents the hyperbolic tangent activation function, which is used to calculate the candidate cell state and the hidden state;

[0104] The parameter x in the above formula t represents the input feature at time step t, including P load , θ, v, and SOC, etc.,

[0105] In some embodiments, the model output is not limited to a single load power prediction value, but also includes the change trend of the device operating state. For example, in the prediction result, the decline rate of the device speed within the next 10 seconds and the rapid consumption trend of the battery SOC can be output simultaneously.

[0106] To ensure the accuracy of the prediction model, the model training process is optimized in this embodiment. The training data mainly comes from the historical operation data of mining equipment, covering a variety of working conditions, including complex situations such as flat driving, steep slope climbing, and downhill braking. Generally, the model uses the mean square error (MSE) as the loss function, and its formula is:

[0107]

[0108] where y i is the actual load demand, is the load value predicted by the model, and N1 is the number of samples.

[0109] As an option, to enhance the generalization ability of the model, the stochastic gradient descent (SGD) optimization algorithm is adopted during the model training process, and a learning rate decay mechanism is introduced to enable the model to adapt to the variable characteristics of the load demand.

[0110] In a possible implementation manner, the load prediction module of this embodiment also supports online update.

[0111] Specifically:

[0112] The newly collected data in real time will be dynamically added to the model's training dataset;

[0113] The model updates the weights using a sliding window mechanism to adapt to the new load change characteristics;

[0114] The online update frequency is set according to the equipment operation environment, for example, it is updated once every 5 minutes.

[0115] The output result is directly transmitted to the dynamic energy management module to provide support for the subsequent power distribution of the internal combustion engine and the electric motor. Generally, the prediction data is updated once per second to ensure that the energy management module can respond to the changes in the load demand in a timely manner.

[0116] S3. Dynamic energy management: According to the current load prediction result, dynamically adjust the power distribution ratio of the internal combustion engine and the electric motor to make the hybrid system meet the operation requirements;

[0117] Dynamic energy management is one of the core steps. It directly adjusts the power output ratio of the internal combustion engine and the electric motor in real time according to the load prediction result in step S2 to meet the operation requirements of the equipment, while reducing fuel consumption and battery loss. Generally, the load demand of mining equipment changes frequently under complex working conditions, and the fixed energy distribution strategy cannot meet the actual needs. Therefore, a dynamic and intelligent energy distribution method is required. In this step, an energy balance model is constructed, and the power distribution is adjusted in real time by combining an optimization algorithm to achieve the dynamic balance of the hybrid system.

[0118] In this embodiment, the specific technical implementation of dynamic energy management includes the following:

[0119] In a possible implementation, this step first calculates the power demand and distribution ratio of the device based on the energy balance model. Specifically:

[0120] The total power demand P total is directly determined by the output of the load prediction module, and the calculation formula is:

[0121] P total = P enaine + P motor ;

[0122] Where: P engine is the output power of the internal combustion engine; P motor is the output power of the electric motor;

[0123] To ensure the health status of the battery, the change relationship of the battery SOC needs to be monitored in real time. The update formula of the battery SOC is:

[0124]

[0125] Where: SOC t is the SOC value of the current battery; Δt is the time interval; E battery,max is the maximum energy storage capacity of the battery;

[0126] As an option, for different load conditions, the present invention designs two typical energy distribution strategies:

[0127] Specifically, when the device is in a high-load condition (such as climbing a slope or starting under heavy load), the output power P engine of the internal combustion engine is preferentially increased to reduce the load on the electric motor. This strategy can effectively reduce the instantaneous discharge pressure of the battery, thereby extending the battery life.

[0128] Under low-load conditions (such as driving on flat ground or running without load), the output power P motor of the electric motor is preferentially utilized, and at the same time, the power output P engine of the internal combustion engine is reduced. This can not only reduce fuel consumption but also effectively utilize the remaining energy of the battery.

[0129] In some embodiments, to ensure the real-time performance and global optimization of the energy distribution strategy, this step adopts an optimization algorithm based on reinforcement learning (RL). Generally, the reinforcement learning algorithm minimizes the fuel consumption cost C fuel and the battery loss cost C battery by dynamically adjusting the power distribution ratio between the internal combustion engine and the electric motor. Its optimization objective function can be expressed as:

[0130]

[0131] Wherein: C fuel (P engine ) is the fuel consumption cost of the internal combustion engine, which is directly related to the power output of the internal combustion engine; C battery (P motor ) is the battery loss cost, mainly considering the impact of the charge and discharge process of the battery on its life.

[0132] In a possible implementation, the reinforcement learning algorithm realizes dynamic optimization through the following three parts:

[0133] State: including the current load power demand P total , the efficiency parameters of the internal combustion engine, etc.;

[0134] Action: Different options for adjusting the power ratio of the internal combustion engine and the electric motor, such as reducing the power of the internal combustion engine, increasing the power of the electric motor, etc.;

[0135] Reward function: Dynamically defined based on the reduction of total energy consumption and the satisfaction of equipment operation requirements.

[0136] Through the above optimization method, this step can quickly adjust the power distribution ratio according to the load prediction result and perform adaptive optimization when the subsequent working conditions change.

[0137] To enhance the robustness of dynamic energy management, in some embodiments, the constraint conditions for power distribution are further defined. Generally, the output power of the internal combustion engine and the electric motor needs to meet the following ranges:

[0138] P engine,min ≤P engine ≤P engine,max ;

[0139] P motor,min ≤P motor ≤P motor,max ;

[0140] At the same time, the battery SOC needs to be maintained within a safe range:

[0141] SOC min ≤SOC t ≤SOC max ;

[0142] For example, when the SOC is lower than SOC min , the system preferentially uses the internal combustion engine to provide power and charges the battery.

[0143] As an improvement, in order to further enhance the flexibility of energy allocation, this embodiment supports setting different optimization goals in different mission modes. For example, in a climbing mission, the optimization goal may focus on maximizing power output, while in a low-load mode, the optimization goal may focus on minimizing fuel consumption.

[0144] In general, the output of the dynamic energy management module includes the adjusted internal combustion engine power P engine and motor power P motor ,These data are transmitted to the energy recovery and distribution module in real time to further optimize energy recovery and utilization;

[0145] S4, Energy recovery: During the deceleration or idling stage of the equipment, the braking energy is recovered by the reverse operation of the motor, and the recovered energy is stored in the battery or supercapacitor;

[0146] This step relies on the real-time power distribution data output by the dynamic energy management module in step S3, and dynamically adjusts the energy recovery strategy in combination with the operating conditions of the equipment to ensure the efficiency and stability of energy recovery. Generally, this step works in conjunction with the subsequent energy distribution module to achieve efficient use of energy.

[0147] In this embodiment, the specific technical implementation of energy recovery includes the following:

[0148] In one possible implementation, energy recovery is first started during the braking or deceleration phase of the equipment, and the conversion of mechanical energy into electrical energy is achieved through the reverse operation of the electric motor. Specifically:

[0149] During the operation of the equipment, when it is detected that the equipment has entered a deceleration, braking or idling condition, the system automatically switches the motor to a power generation mode, that is, the motor runs in reverse to recover mechanical energy. In this mode, the braking force is mainly provided by the electromagnetic resistance generated by the reverse rotation of the motor. The magnitude of the braking force is adjusted in real time by the control system according to the equipment operating status and load requirements.

[0150] As an option, in order to improve the efficiency of energy recovery, this embodiment introduces a dynamic adjustment mechanism for braking current. Generally, the braking current of the motor directly affects its power generation efficiency and recovery power. By adjusting the braking current, the conversion efficiency of mechanical energy to electrical energy can be optimized.

[0151] Specifically, the energy recovery power P regen The calculation formula of (t) is:

[0152] P regen (t) = I regen (t)·V regen (t)·η regen ;

[0153] Where: I regen (t) is the braking current of the motor during the recovery stage; V regen (t) is the operating voltage of the motor during the recovery stage; η regen is the efficiency factor of energy recovery, which is affected by factors such as the operating state and temperature of the motor.

[0154] In some embodiments, the system monitors the temperature and operating state of the motor in real time through sensors. When it detects that the temperature is too high or the load is abnormal, it automatically reduces the braking current I regen (t) to protect the operating stability of the motor.

[0155] To achieve optimized storage of the recovered energy, this embodiment designs an energy dynamic allocation mechanism based on storage requirements. Generally, the recovered electrical energy can be stored in a battery or a supercapacitor. The battery is suitable for long-term energy storage, while the supercapacitor is suitable for short-term high-power output scenarios.

[0156] Specifically, this embodiment dynamically adjusts the storage location of the recovered energy through the following logic:

[0157] When the device is operating at low load, the recovered energy is preferentially stored in the battery to increase the battery SOC;

[0158] When the device is operating at high load, the recovered energy is preferentially stored in the supercapacitor to provide instantaneous high-power output for the motor subsequently.

[0159] In a possible implementation, the allocation ratio of the recovered energy is calculated by the following formula:

[0160]

[0161] Where: E recovery is the total recovered energy; w battery is the battery storage weight, which is affected by the battery SOC; w capacitor is the supercapacitor storage weight, which is affected by the task urgency and instantaneous power demand.

[0162] To further improve the energy recovery efficiency, a segmented optimization strategy is adopted in this embodiment. In some embodiments, the recovery modes are divided according to different working conditions. For example, during a long downhill stage, the system tends to a stable continuous recovery mode; while during a short-term rapid deceleration stage, the system preferentially selects an instantaneous high-power recovery mode.

[0163] In a possible implementation, the energy recovery module also dynamically adjusts the power generation load of the motor in combination with the slope information θ of the vehicle. For example, when a small slope is detected, the power generation load of the motor is reduced to reduce the braking force; while when the slope is large, the power generation load is increased to increase the braking force and maximize the recovery efficiency.

[0164] As an improvement, in order to further improve the adaptability of the energy recovery module, this embodiment supports multi-scenario switching. Specifically:

[0165] In the low-speed operation scenario of mining equipment, the system mainly operates in a low-power recovery mode to reduce the impact of energy recovery on equipment operation;

[0166] In the high-speed operation scenario, the system enables a high-power recovery mode to quickly accumulate energy and reduce the inertia of the equipment.

[0167] The output of this embodiment includes the braking current, recovery power, and storage location that are adjusted in real time. These data are directly transmitted to the energy distribution module to support the subsequent task energy supply.

[0168] S5. Energy distribution: Dynamically distribute the recovered energy based on the current task priority and working condition requirements, and give priority to high-priority tasks;

[0169] S5 combines the energy recovered in step S4 with the power distribution requirements of the dynamic energy management module in step S3, and reasonably distributes it according to the priority of the current task of the equipment and the working condition requirements. Generally, during the operation of mining equipment, the task requirements are diverse, and the optimization of the energy distribution strategy is directly related to the operation efficiency and energy-saving effect of the equipment. In this step, through the priority weight model and dynamic adjustment mechanism, it is ensured that high-priority tasks obtain sufficient energy support while avoiding energy waste.

[0170] In this embodiment, the specific technical implementation of energy distribution includes the following:

[0171] In a possible implementation, the energy distribution module realizes the efficient utilization of energy based on the dynamic adjustment mechanism of task priority. Specifically:

[0172] Energy distribution takes the task priority as the core basis. The priority weight w of each task i is jointly determined by the real-time task requirements and operating conditions of the equipment. The dynamic calculation formula for task priority is:

[0173]

[0174] Where: is the current load power demand of task i1; Δt i is the remaining completion time of the task; T iis the total scheduled completion time of the task; α i , β i are the priority adjustment weight coefficients, and the specific values can be preset according to the task type; α i is the controlled load power is the weight in the calculation of task priority, indicating the importance of the current load power to the task priority; β i is the weight in the calculation of task priority for the ratio of the remaining completion time to the total completion time , indicating the influence degree of the remaining task time on the priority;

[0175] As an option, in high-load tasks (such as equipment ramping or heavy-load starting), the priority weight w i will be significantly increased to ensure that such tasks can be preferentially allocated with recovered energy. In low-load tasks (such as idling driving), the priority weight is reduced, and the recovered energy is more used to improve the battery SOC.

[0176] Specifically, this embodiment realizes the dynamic allocation of energy through the following formula:

[0177]

[0178] Where: E alloc,i is the energy allocated to task i; E recovery is the total recovered energy in step S4; N2 is the total number of tasks of the current device;

[0179] In some embodiments, the system optimizes the allocation strategy for different task types and working conditions. For example, when the device is in a multi-task running mode, the system preferentially allocates energy to emergency demand scenarios such as ramping tasks and large-load tasks; for tasks with a long duration but a low load power (such as flat ground transportation), the energy allocation ratio is relatively reduced.

[0180] To further improve the allocation efficiency, this embodiment introduces a dynamic adjustment mechanism. Generally, the system adjusts the priority weight coefficients α i and β i according to the real-time changes in the task requirements of the device. For example, when the load power of a certain task suddenly increases, the system immediately raises its priority to ensure the timeliness of energy supply.

[0181] In a possible implementation manner, the energy allocation module also combines the battery SOC status and the supercapacitor status of the device. For example:

[0182] When the battery SOC is lower than the safety threshold SOC min , the recovered energy is preferentially allocated to the battery to increase the SOC level;

[0183] When the energy storage of the supercapacitor approaches the full-load state, reduce the allocation ratio to the supercapacitor and supply more energy to the battery.

[0184] As an improvement, the energy distribution module of this embodiment supports the energy re-distribution function between tasks. For example, when a high-priority task is completed, its unused energy can be dynamically re-allocated to other running tasks, thus avoiding energy waste.

[0185] To verify the rationality of energy distribution, this embodiment establishes a feedback mechanism for the distribution effect through real-time monitoring of the device operating state. The feedback data includes:

[0186] The energy usage of each task;

[0187] The dynamic change of the battery SOC;

[0188] The energy storage level of the supercapacitor.

[0189] These feedback data are used as important references for adjusting the priority weights and distribution strategies.

[0190] Specifically, during the actual operation of the device, some embodiments support the distribution optimization of the following scenarios:

[0191] Startup phase: When the device starts up, due to the high instantaneous power demand, the recovered energy is preferentially allocated to the startup task;

[0192] Climbing phase: When the device is running on a steep slope, most of the recovered energy is used for the motor power output to reduce the load on the internal combustion engine;

[0193] No-load operation: When the device is operating under no-load or light-load conditions, the recovered energy is preferentially stored in the battery or supercapacitor.

[0194] The output of this embodiment includes the energy distribution ratio of each task and the dynamic adjustment scheme of the storage device. The output data is directly transmitted to the multi-device cooperation module for further optimizing the energy sharing and task coordination between devices.

[0195] S6. Multi-device cooperation: Through task allocation and energy sharing optimization, achieve collaborative scheduling and overall energy saving of multiple devices;

[0196] This step is based on the energy distribution results of each device in step S5, combined with the real-time task status, energy demand and operating conditions between multiple devices, and through task allocation optimization and energy sharing mechanism, improve the overall operating efficiency of the device group. Generally, in the group operation of mining equipment, due to the task complexity and uneven energy distribution of the equipment, it may lead to resource waste or local energy shortage. Therefore, this step realizes the efficient cooperation of multiple devices through collaborative scheduling and dynamic energy sharing between devices.

[0197] In this embodiment, the specific technical implementation of multi-device collaboration includes the following:

[0198] In a possible implementation, multi-device collaboration is based on device-to-device communication, and the operating states of each device are obtained through real-time data interaction. Specifically:

[0199] Through the V2V communication protocol, devices can share their real-time operating parameters, including the current load power P load,i , the battery SOC state SOC i , the task execution progress Δt i / T i and the remaining allocation situation E of the recovered energy residual,i . These data are centrally processed to determine the task allocation and energy sharing strategies for multi-device collaboration.

[0200] As an option, in a multi-task scenario, the optimization of task allocation is achieved through a scheduling algorithm. Generally, the ant colony optimization algorithm (ACO) is used to optimize the task paths and load allocations of devices. The goal of the ant colony optimization algorithm is to minimize the total energy consumption of the device group while ensuring that tasks are completed on time.

[0201] Specifically, the core formula of the ant colony optimization algorithm is as follows:

[0202]

[0203] where: τ ij (t) is the pheromone intensity on the task path from device i2 to device j; ρ is the pheromone evaporation coefficient; Q is a constant used to adjust the pheromone intensity; L best is the total energy consumption of the current optimal path.

[0204] Through the above optimization process, the system can dynamically adjust the task priorities and path planning of devices. For example, when the load power P of a certain device load,i exceeds the set threshold, the scheduling algorithm will reallocate some tasks to neighboring low-load devices to balance the task load.

[0205] In a possible implementation, multi-device collaboration also supports the dynamic energy sharing function. Specifically: when it is detected that the battery SOC of a certain device is lower than the safety threshold SOC min , the system will trigger the energy sharing mechanism, and the high-SOC device will transmit energy to the low-SOC device through the energy sharing module to ensure the normal operation of its tasks.

[0206] The dynamic allocation ratio of energy sharing is calculated by the following formula:

[0207]

[0208] Wherein: is the energy shared from device i2 to device j; SOC i , SOC j is the current battery SOC of devices i and j; is the physical distance between device i2 and device j; η transfer is the energy transfer efficiency factor.

[0209] In some embodiments, to improve the stability of energy sharing, the system preferentially selects devices with a shorter distance for energy transfer. For example, when multiple devices are detected to be in a low SOC state, the energy sharing module preferentially supplies energy to the nearest device based on the transfer efficiency to reduce the loss during energy transfer.

[0210] To achieve the global optimization of multiple devices, a distributed edge computing module is further introduced in this embodiment. Generally, the edge computing unit deployed on each device is responsible for processing the allocation optimization of local tasks and transmitting the optimization results to other devices through V2V communication. The advantage of distributed computing is that it reduces the computing burden on the central server and improves the real-time performance of scheduling and energy sharing.

[0211] In a possible implementation, this step also supports the optimization of the following special scenarios:

[0212] High-load scenario: When the entire device group is in a high-load state, the recovered energy is preferentially concentrated and allocated to critical task devices, and through a dynamic task offloading mechanism, some low-priority tasks are postponed or cancelled;

[0213] Emergency task scenario: For high-priority emergency tasks, such as ore transportation or emergency repair of equipment, the system dynamically adjusts the energy allocation ratio of other devices to ensure the energy requirements of the emergency tasks. The outputs of this embodiment include the optimized task allocation scheme and energy sharing plan, and these outputs are transmitted to the control modules of each device in real time to guide their specific task operations.

[0214] It should be understood that the parts not elaborated in detail in this specification belong to the prior art. The above embodiments merely describe the preferred implementation manners of the present invention and do not limit the scope of the present invention. Without departing from the design spirit of the present invention, various deformations and improvements made by those of ordinary skill in the art to the technical solutions of the present invention shall fall within the protection scope determined by the claims of the present invention.

Claims

1. A hybrid power energy-saving control method suitable for mining equipment, characterized in that: The following steps are involved: S1. Working condition perception: Collect real-time operating status data of mining equipment through multiple sensors, including load power, slope, speed and battery SOC; S2. Load forecasting: Based on the collected real-time operation data and historical data, the time series forecasting model is used to predict the load demand in the future time window; S3, Dynamic Energy Management: Dynamically adjust the power distribution ratio of the internal combustion engine and the electric motor according to the current load forecast results so that the hybrid system can meet the operating requirements; S4, Energy recovery: During the deceleration or idling stage of the equipment, the braking energy is recovered by the reverse operation of the motor, and the recovered energy is stored in the battery or supercapacitor; S5, Energy allocation: Based on the current task priority and working condition requirements, the recovered energy is dynamically allocated to give priority to high-priority tasks; S6. Multi-device collaboration: Through task allocation and energy sharing optimization, collaborative scheduling of multiple devices and overall energy saving are achieved.

2. A hybrid power energy-saving control method suitable for mining equipment according to claim 1, characterized in that: The working condition perception includes: Obtain the current load power data of the equipment through the load sensor; Real-time monitoring of mine terrain slope through slope sensors; Record the running speed of the equipment through the speed sensor; Obtain battery status data through the battery SOC sensor; The Kalman filter algorithm is used to fuse multi-sensor data to eliminate noise and improve data accuracy.

3. A hybrid power energy-saving control method suitable for mining equipment according to claim 1, characterized in that: The load prediction includes: Using historical load power, operating speed, terrain slope, and battery SOC as input features; Use long short-term memory network model to process time series operating condition data; Capturing the long-term and short-term dependencies of load data through network structures; Output the load demand in the future time window, including future load power demand, battery SOC change and operating speed change.

4. A hybrid power energy-saving control method suitable for mining equipment according to claim 1, characterized in that: The dynamic energy management includes: Construct an energy balance model to determine the power distribution ratio between the internal combustion engine and the electric motor; Under high load conditions, priority is given to increasing the output power of the internal combustion engine and reducing the power demand of the electric motor; Under low load conditions, the electric motor power is used first, reducing the power output of the internal combustion engine; The power ratio of the internal combustion engine and the electric motor is dynamically adjusted through an optimization algorithm to achieve dynamic energy balance of the hybrid system.

5. A hybrid power energy-saving control method suitable for mining equipment according to claim 4, characterized in that: The optimization algorithm includes: Define the optimization objective function to minimize the fuel consumption cost and battery loss cost; Set constraints, including the power range of the internal combustion engine and electric motor and the safe range of the battery SOC; Use reinforcement learning algorithms to optimize the power allocation strategy and adjust the power output ratio of the internal combustion engine and the electric motor in real time; While meeting the equipment load requirements, ensure that the battery SOC remains within a safe range.

6. A hybrid power energy-saving control method suitable for mining equipment according to claim 5, characterized in that: The optimization goal is: Minimize fuel consumption and battery loss of the hybrid system while meeting the equipment load requirements; Dynamically adjust task allocation and energy sharing strategies to optimize the overall energy saving effect of the device group.

7. A hybrid power energy-saving control method suitable for mining equipment according to claim 1, characterized in that: The energy recovery comprises: During the deceleration, braking or idling stage of the equipment, the mechanical energy is converted into electrical energy through the reverse work of the motor; Dynamically adjust the braking current of the motor to improve energy recovery efficiency; The recovered energy is stored in batteries or supercapacitors through an energy management system; The storage location of energy recovery is dynamically adjusted according to the battery SOC state.

8. A hybrid power energy-saving control method applicable to mining equipment according to claim 1, characterized in that: The energy distribution includes: Dynamically adjust the allocation ratio of recovered energy according to the priority of equipment tasks; The allocation ratio of task energy requirements is calculated through a priority weight model; High-priority tasks include startup and high-load climbing conditions, with priority given to allocating recovered energy to meet instantaneous high power demands; The energy requirements of low priority tasks are met by residual recovered energy or battery backup power.

9. A hybrid power energy-saving control method applicable to mining equipment according to claim 1, characterized in that: The multi-device collaboration includes: Obtain the load status, task allocation information and battery SOC status of each device through inter-device communication; Use scheduling algorithms to optimize task allocation for multiple devices; By monitoring the status changes between devices in real time, dynamically adjust the task priority and scheduling plan; Minimize the total energy consumption of the entire device group while meeting the task completion time.

10. A hybrid power energy-saving control method applicable to mining equipment according to claim 1, characterized in that: The energy sharing includes: Dynamically allocate shared energy between high SOC devices and low SOC devices; The shared energy allocation ratio is dynamically adjusted according to the task priority and physical distance between devices; Realize real-time transmission of shared energy through communication protocols between devices; Prioritize energy needs of mission-critical equipment.

Images