Fully-mechanized coal mining liquid supply system speed regulation method based on model prediction and multi-objective optimization

Abstract

The invention discloses a fully-mechanized coal mining liquid supply system speed regulation method based on model prediction and multi-objective optimization, and belongs to the technical field of fully-mechanized coal mining liquid supply control, and the method comprises the following steps: S1, collecting the real-time state information of a pump station; s2, according to the real-time state information of the pump station, utilizing the prediction model to predict the outlet pressure of the pump station; s3, constructing a multi-objective optimization function, and S4, controlling the pump station outlet pressure according to the predicted pump station outlet pressure and the multi-objective optimization function. According to the method, the health of the pump machine is guaranteed, and the service life of the pump machine can be effectively prolonged, the maintenance cost is reduced and the reliability and economy of the system are improved by monitoring the health state of the pump machine in real time and taking the health state into an optimization target.

Description

Technical Field

[0001] The present invention belongs to the technical field of fully-mechanized mining fluid supply control, and in particular relates to a speed regulation method for a fully-mechanized mining fluid supply system based on model prediction and multi-objective optimization. Background Art

[0002] In fully mechanized mining faces, emulsion pumping stations are the power source for hydraulic supports. The stability and response speed of their hydraulic pressure directly impact the proper operation of the supports and coal mining efficiency. Traditional pumping station speed regulation methods primarily utilize PID control, which adjusts the pumping station motor frequency to control the outlet pressure. However, PID control suffers from the following limitations in practical applications: First, parameter tuning is difficult: PID parameters are sensitive to changes in system model parameters, making it difficult to achieve ideal control results in the complex and ever-changing environment of fully mechanized mining faces. Second, response speed is limited: PID control is inherently based on error feedback, and its response to system disturbances is delayed, making it difficult to meet the rapid motion requirements of hydraulic supports. Third, single-target control: PID control typically optimizes for a single objective (such as pressure deviation), making it difficult to balance multiple objectives such as pressure fluctuations, system energy efficiency, and equipment life. Fourth, it lacks foresight: PID control cannot predict future system state changes, making it difficult to respond to sudden load disturbances.

[0003] In recent years, with the increasing demand for automation and intelligent control in fully mechanized mining faces, pump control speed regulation methods based on model predictive control have gradually attracted attention. By leveraging system models to predict future system states and optimizing control inputs based on the predicted results, more precise and rapid system control can be achieved. However, existing pump station speed regulation methods based on model predictive control are mostly limited to single-objective optimization and fail to fully consider the actual operating conditions of fully mechanized mining faces, such as the randomness and periodicity of support movements and the performance degradation caused by long-term pump operation. Summary of the Invention

[0004] In order to solve the problems of insufficient control accuracy, slow response speed, difficulty in balancing multi-objective optimization, lack of foresight and insufficient consideration of the health status of the pump machine in the existing comprehensive mining pump station speed control method, the present invention proposes a comprehensive mining liquid supply system speed control method based on model prediction and multi-objective optimization.

[0005] The technical solution of the present invention is: a speed regulation method for a fully mechanized mining fluid supply system based on model prediction and multi-objective optimization comprises the following steps: S1. Collect real-time status information of the pumping station; S2. Predict the pump station outlet pressure using a prediction model based on the real-time status information of the pump station; S3, construct a multi-objective optimization function; S4. Control the pump station outlet pressure according to the predicted pump station outlet pressure and the multi-objective optimization function.

[0006] Furthermore, in S1, the real-time status information includes the pump station outlet pressure, flow rate, motor frequency, motor current, motor power, emulsion temperature and support action status.

[0007] Furthermore, in S2, the pump station outlet pressure is predicted using a prediction model composed of a mechanism model and a data-driven model.

[0008] Furthermore, the expression of the mechanism model is: Where Q(t) represents the pump outlet flow rate, K1 represents the first pump characteristic coefficient, K2 represents the second pump characteristic coefficient, n(t) represents the pump speed, and P(t) represents the pump outlet pressure. represents the nonlinear disturbance term, a i represents the autoregressive coefficient of the disturbance term, b j represents the sliding average coefficient of the control input, represents the nonlinear disturbance term at time ti, and u(tj) represents the control input at time tj; The data-driven model includes an input layer, a bidirectional LSTM layer, an attention mechanism module, and an output layer connected in sequence. The loss function L of the data-driven model is expressed as: ; In the formula, λ1 represents the weight coefficient of mean square error, λ2 represents the weight coefficient of continuous sorting probability score, λ3 represents the weight coefficient of KL divergence, L MSE represents the mean square error, L CRPS represents the continuous sorting probability score, L KL represents KL divergence; The expression of the prediction model is: ; ; ; Where, represents the prediction model output, represents the output of the mechanism model, represents the data-driven model output, e represents the model fusion error, w m (t) represents the dynamic weight of the mechanism model, w d (t) represents the dynamic weight of the data-driven model, k represents the adjustment coefficient, and C m (t) represents the confidence level of the mechanism model; The expression of the residual compensation function of the prediction model is: Where ΔP(t) represents the pressure prediction residual, β0 represents the steady-state residual coefficient, β1 represents the dynamic residual coefficient, β2 represents the pressure change rate residual coefficient, and P(t) represents the system pressure measurement value. represents the derivative of the system pressure measurement, represents the residual noise term.

[0009] Furthermore, in S3, the multi-objective optimization function includes a pressure deviation minimization function, a frequency adjustment amplitude minimization function, an energy efficiency maximization function, and a health risk minimization function.

[0010] Furthermore, the expression of the pressure deviation minimization function J1 is: ; Where P(k) represents the actual pressure at the pump station outlet at time k, P set (k) represents the set pressure at the pump station outlet at time k, and N represents the total time; The expression of the frequency adjustment amplitude minimization function J2 is: ; Where τ(k) represents the time response delay of the system at time k, and Δf(k) represents the system frequency deviation; The expression of energy efficiency maximization function J3 is: ; Where η(k) represents the efficiency coefficient of the system at time k, P(k) represents the active power of the system at time k, and Q(k) represents the reactive power of the system at time k. The expression of health risk minimization function J4 is: ; Where, T safe represents the safety temperature threshold of the device, and T(k) represents the temperature of the device at time k.

[0011] Furthermore, in S4, based on the predicted pump station outlet pressure, the multi-objective optimization function is solved to obtain the optimal control input sequence, and the optimal control input sequence is output to the pump station to adjust the motor frequency and complete the control of the pump station outlet pressure.

[0012] Furthermore, the expression of the motor frequency f is: ; Where, f base Indicates the reference frequency, P max Indicates the maximum allowable outlet pressure, Pmin Indicates the minimum allowable outlet pressure, P o Indicates the actual outlet pressure, Δf max Indicates the maximum allowable adjustment amount.

[0013] The beneficial effects of the present invention are: (1) The control accuracy of the present invention is high. Through model-based prediction and multi-objective optimization, it can achieve more accurate control of the pump station outlet pressure, effectively reduce pressure deviation and fluctuation, and improve the stability and reliability of the hydraulic support operation; (2) The present invention has a fast response speed and can adjust the motor frequency in advance according to the predicted support action state, thereby improving the system's response speed to load disturbances and meeting the requirements of rapid action of the hydraulic support; (3) The present invention performs multi-objective optimization, which can simultaneously consider multiple objectives such as pressure control, energy efficiency optimization, equipment life, and pump health, thereby achieving comprehensive performance optimization of the pump station system and improving the overall operating efficiency of the system; (4) The present invention has strong adaptability and can flexibly adjust the weight coefficients in the multi-objective optimization function according to different working conditions and needs. It has strong adaptability and can adapt to the complex and changeable environment of the fully mechanized mining working face; (5) The present invention has a high degree of intelligence, adopts model predictive control and multi-objective optimization algorithm, has a high degree of intelligence, and is in line with the development trend of automation and intelligence of fully mechanized mining working faces; (6) The present invention realizes the health protection of the pump. By monitoring the health status of the pump in real time and incorporating it into the optimization target, it can effectively extend the service life of the pump, reduce maintenance costs, and improve the reliability and economy of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flow chart of the speed regulation method for the fully mechanized mining fluid supply system based on model prediction and multi-objective optimization. DETAILED DESCRIPTION

[0015] The embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0016] like Figure 1 As shown, the present invention provides a speed control method for a fully mechanized mining fluid supply system based on model prediction and multi-objective optimization, comprising the following steps: S1. Collect real-time status information of the pumping station; S2. Predict the pump station outlet pressure using a prediction model based on the real-time status information of the pump station; S3, construct a multi-objective optimization function; S4. Control the pump station outlet pressure according to the predicted pump station outlet pressure and the multi-objective optimization function.

[0017] In an embodiment of the present invention, the pump station system includes multiple pump groups including an emulsion pump, a spray pump, and a clean water pump, and each pump group is equipped with a frequency converter and corresponding sensors.

[0018] Sensors include pressure sensors, flow sensors, vibration sensors, and temperature sensors, etc., which are used to monitor the operating status of the pumping station system in real time.

[0019] As the core of the control system, the industrial computer connects the industrial computer, inverter and sensors, and is responsible for data acquisition, processing, model prediction, optimization calculation and control output.

[0020] In the embodiment of the present invention, in S1, the real-time status information includes the pump station outlet pressure, flow rate, motor frequency, motor current, motor power, emulsion temperature and support action status.

[0021] Real-time status information is used to monitor the operating status of the pump station system in real time, including but not limited to the pump station outlet pressure P, which reflects the output capacity of the pump station; the flow rate Q, which reflects the liquid supply capacity of the pump station; the motor frequency f, which reflects the operating speed of the pump station; the motor current I, which reflects the load condition of the motor; the motor power P, which reflects the energy consumption of the pump station; the emulsion temperature T, which reflects the viscosity of the emulsion and the operating status of the pump station; the support action status S, which reflects the change of the load, such as raising the column, lowering the column and pushing, etc.; the health status indicators H such as pump vibration and temperature, which reflect the operating health of the pump.

[0022] In the embodiment of the present invention, in S2, the pump station outlet pressure is predicted using a prediction model composed of a mechanism model and a data-driven model.

[0023] In the embodiment of the present invention, the expression of the mechanism model is: Where Q(t) represents the pump outlet flow rate, K1 represents the first pump characteristic coefficient, K2 represents the second pump characteristic coefficient, n(t) represents the pump speed, and P(t) represents the pump outlet pressure. represents the nonlinear disturbance term, a i represents the autoregressive coefficient of the disturbance term, b j represents the sliding average coefficient of the control input, represents the nonlinear disturbance term at time ti, and u(tj) represents the control input at time tj. The pump speed is linearly related to the motor frequency. The first and second pump characteristic coefficients are calibrated through factory testing, with typical values ​​ranging from K1 to 0.03-0.05 and K2 to 0.15-0.25.

[0024] The data-driven model consists of an input layer, a bidirectional LSTM layer, an attention mechanism module, and an output layer connected in sequence. The loss function L of the data-driven model is expressed as: ; In the formula, λ1 represents the weight coefficient of mean square error, λ2 represents the weight coefficient of continuous sorting probability score, λ3 represents the weight coefficient of KL divergence, L MSE represents the mean square error, L CRPS represents the continuous sorting probability score, L KL represents the KL divergence; λ1=0.6, λ2=0.3, λ3=0.1.

[0025] Input layer: 8-dimensional vector features [P(t), Q(t), f(t), I(t), T(t), S(t), H(t), t]; Bidirectional LSTM layer (64 units): captures temporal features; GRU layer (32 units): extracts nonlinear patterns; Attention mechanism module: , softmax(·) represents the activation function, h i represents the hidden state at the i-th moment, W represents the first learnable parameter, V represents the second learnable parameter, and b represents the third learnable parameter; Output layer: N-step pressure prediction, N=5 steps, corresponding to the typical response time of the hydraulic system.

[0026] The expression of the prediction model is: ; ; ; Where, represents the prediction model output, represents the output of the mechanism model, represents the data-driven model output, e represents the model fusion error, w m (t) represents the dynamic weight of the mechanism model, w d (t) represents the dynamic weight of the data-driven model, k represents the adjustment coefficient, and C m (t) represents the confidence level of the mechanism model; Using the real-time status information obtained in step S1, the future pump station outlet pressure is predicted. This model can be developed using mechanistic modeling, data-driven modeling, or a combination of both. For example, a mechanistic model can establish a nonlinear mathematical model of the pump station system based on fundamental principles such as fluid mechanics and electromechanics. A data-driven model utilizes historical operating data and deep learning to establish a predictive model for the pump station system. Data-driven models can effectively capture the nonlinear relationships and dynamic characteristics of the system.

[0027] Hybrid models can combine mechanistic models with data-driven models, using the mechanistic model to provide physical constraints on the system and the data-driven model to compensate for inaccurate mechanistic model parameters or the effects of unmodeled dynamics. They also need to consider the predicted support motion state information. For example, based on historical support motion data, pattern recognition or time series analysis methods can be used to predict support motion patterns (such as raising, lowering, and shifting) over a period of time and estimate the corresponding flow demand, thereby more accurately predicting changes in pump station outlet pressure.

[0028] The expression of the residual compensation function of the prediction model is: Where ΔP(t) represents the pressure prediction residual, β0 represents the steady-state residual coefficient, β1 represents the dynamic residual coefficient, β2 represents the pressure change rate residual coefficient, and P(t) represents the system pressure measurement value. represents the derivative of the system pressure measurement, represents the residual noise term.

[0029] In an embodiment of the present invention, in S3, the multi-objective optimization function includes a pressure deviation minimization function, a frequency adjustment amplitude minimization function, an energy efficiency maximization function, and a health risk minimization function.

[0030] In the embodiment of the present invention, the expression of the pressure deviation minimization function J1 is: ; Where P(k) represents the actual pressure at the pump station outlet at time k, P set (k) represents the set pressure at the pump station outlet at time k, and N represents the total time; The expression of the frequency adjustment amplitude minimization function J2 is: ; Where τ(k) represents the time response delay of the system at time k, and Δf(k) represents the system frequency deviation; The expression of energy efficiency maximization function J3 is: ; Where η(k) represents the efficiency coefficient of the system at time k, P(k) represents the active power of the system at time k, and Q(k) represents the reactive power of the system at time k. The expression of health risk minimization function J4 is: ; Where, T safe represents the safety temperature threshold of the device, and T(k) represents the temperature of the device at time k.

[0031] In an embodiment of the present invention, in S4, based on the predicted pump station outlet pressure, the multi-objective optimization function is solved to obtain the optimal control input sequence, and the optimal control input sequence is output to the pump station, the motor frequency is adjusted, and the control of the pump station outlet pressure is completed.

[0032] The optimal solution is selected from the Pareto optimal solution set, and a set of optimal control input sequences is selected according to actual needs [f t1 ,f t2 ,…,f tn ], where f tk where (k=1,2,...,n) represents the motor frequency setpoint recommended to be applied in the next k control cycles.

[0033] Control instruction sending: The industrial control computer sends the optimized control input sequence to the frequency converter to adjust the frequency of the pump group motor to achieve balanced speed regulation of the pump station system.

[0034] Control effect monitoring: Real-time monitoring of the operating status of the pumping station system, evaluation of the control effect, and adjustment of control parameters as needed.

[0035] In the embodiment of the present invention, the motor frequency f is expressed as: ; Where, f base Indicates the reference frequency, P max Indicates the maximum allowable outlet pressure, P min Indicates the minimum allowable outlet pressure, P o Indicates the actual outlet pressure, Δf max Indicates the maximum allowable adjustment amount.

[0036] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and it should be understood that the scope of protection of the present invention is not limited to such specific descriptions and embodiments. Those skilled in the art can make various other specific variations and combinations based on the technical teachings disclosed in the present invention without departing from the essence of the present invention, and such variations and combinations are still within the scope of protection of the present invention.

Claims

1. A speed control method for a fully mechanized mining fluid supply system based on model prediction and multi-objective optimization, characterized in that: The following steps are involved: S1. Collect real-time status information of the pumping station; S2. Predict the pump station outlet pressure using a prediction model based on the real-time status information of the pump station; S3, construct a multi-objective optimization function; S4. Control the pump station outlet pressure according to the predicted pump station outlet pressure and the multi-objective optimization function.

2. The speed control method for a fully mechanized mining liquid supply system based on model prediction and multi-objective optimization according to claim 1 is characterized in that: In S1, the real-time status information includes the pump station outlet pressure, flow rate, motor frequency, motor current, motor power, emulsion temperature and support action status.

3. The speed control method for a fully mechanized mining liquid supply system based on model prediction and multi-objective optimization according to claim 1 is characterized in that: In S2, the pump station outlet pressure is predicted using a prediction model composed of a mechanism model and a data-driven model.

4. The speed control method for a fully mechanized mining liquid supply system based on model prediction and multi-objective optimization according to claim 3 is characterized in that: The expression of the mechanism model is: Where Q(t) represents the pump outlet flow rate, K1 represents the first pump characteristic coefficient, K2 represents the second pump characteristic coefficient, n(t) represents the pump speed, and P(t) represents the pump outlet pressure. represents the nonlinear disturbance term, a i represents the autoregressive coefficient of the disturbance term, b j represents the sliding average coefficient of the control input, represents the nonlinear disturbance term at time ti, and u(tj) represents the control input at time tj; The data-driven model includes an input layer, a bidirectional LSTM layer, an attention mechanism module, and an output layer connected in sequence. The loss function L of the data-driven model is expressed as: ； In the formula, λ1 represents the weight coefficient of mean square error, λ2 represents the weight coefficient of continuous sorting probability score, λ3 represents the weight coefficient of KL divergence, L MSE represents the mean square error, L CRPS represents the continuous sorting probability score, L KL represents KL divergence; The expression of the prediction model is: ； ； ； Where, represents the prediction model output, represents the output of the mechanism model, represents the data-driven model output, e represents the model fusion error, w m (t) represents the dynamic weight of the mechanism model, w d (t) represents the dynamic weight of the data-driven model, k represents the adjustment coefficient, and C m (t) represents the confidence level of the mechanism model; The expression of the residual compensation function of the prediction model is: Where ΔP(t) represents the pressure prediction residual, β0 represents the steady-state residual coefficient, β1 represents the dynamic residual coefficient, β2 represents the pressure change rate residual coefficient, and P(t) represents the system pressure measurement value. represents the derivative of the system pressure measurement, represents the residual noise term.

5. The speed control method for a fully mechanized mining liquid supply system based on model prediction and multi-objective optimization according to claim 1 is characterized in that: In S3, the multi-objective optimization function includes a pressure deviation minimization function, a frequency adjustment amplitude minimization function, an energy efficiency maximization function, and a health risk minimization function.

6. The speed control method for a fully mechanized mining liquid supply system based on model prediction and multi-objective optimization according to claim 5 is characterized in that: The expression of the pressure deviation minimization function J1 is: ； Where P(k) represents the actual pressure at the pump station outlet at time k, P set (k) represents the set pressure at the pump station outlet at time k, and N represents the total time; The expression of the frequency adjustment amplitude minimization function J2 is: ； Where τ(k) represents the time response delay of the system at time k, and Δf(k) represents the system frequency deviation; The expression of the energy efficiency maximization function J3 is: ； Where η(k) represents the efficiency coefficient of the system at time k, P(k) represents the active power of the system at time k, and Q(k) represents the reactive power of the system at time k. The expression of the health risk minimization function J4 is: ； Where, T safe represents the safety temperature threshold of the device, and T(k) represents the temperature of the device at time k.

7. The speed control method for a fully mechanized mining fluid supply system based on model prediction and multi-objective optimization according to claim 1 is characterized in that: In S4, based on the predicted pump station outlet pressure, the multi-objective optimization function is solved to obtain the optimal control input sequence, and the optimal control input sequence is output to the pump station, the motor frequency is adjusted, and the control of the pump station outlet pressure is completed.

8. The speed control method for a fully mechanized mining liquid supply system based on model prediction and multi-objective optimization according to claim 7 is characterized in that: The expression of the motor frequency f is: ； Where, f base Indicates the reference frequency, P max Indicates the maximum allowable outlet pressure, P min Indicates the minimum allowable outlet pressure, P o Indicates the actual outlet pressure, Δf max Indicates the maximum allowable adjustment amount.

Images