An intelligent control method for an integrated electrically-driven hydraulic support variable frequency pump station

Abstract

The application discloses an intelligent control method for an integrated electrically-driven hydraulic support variable-frequency pump station, and belongs to the field of smart power grids and power system automation, which comprises the following steps: S1, independently integrating a board-mounted electrically-driven pump station on each hydraulic support; S2, after starting, loading a preset fuzzy PID rule base and an efficiency MAP model on a controller arranged on the board-mounted electrically-driven pump station; S3, the controller receiving an action instruction from an upper computer or a local computer and generating a target pressure P0; S4, the controller collecting column pressures P of each hydraulic support in real time and generating an optimal motor speed N; and S5, a frequency converter driving the operation of the motor and the pump according to the optimal motor speed N. The intelligent control method for the integrated electrically-driven hydraulic support variable-frequency pump station introduces a brand-new distributed electrically-driven architecture and combines an AI intelligent control algorithm, so that the power source of a single hydraulic support is optimized.

Description

Technical Field

[0001] This invention relates to the field of smart grids and power system automation, specifically to an intelligent control method for an integrated electro-hydraulic support variable frequency pump station. Background Technology

[0002] Traditional hydraulic supports typically employ a centralized hydraulic drive architecture. This involves installing a high-power emulsion pump station in the working face tunnel or on the surface, supplying emulsion to each support on the working face via high-pressure pipelines hundreds of meters long. The supports themselves have no independent power source; all their movements depend on external fluid supply.

[0003] The existing technology has the following problems: (1) High-pressure emulsions have a pressure loss of up to 5-8 MPa in long-distance pipelines, and this part of the energy is completely dissipated in the form of heat. At the same time, in order to ensure the liquid supply pressure of the end support, the pump station needs to maintain continuous operation at a high pressure of 30-35 MPa all year round. Even when most supports are stationary and maintaining pressure, the pump station motor is still running idle or inefficiently, resulting in energy waste.

[0004] (2) Emulsion (a mixture of water and emulsified oil) is prone to bacterial growth, deterioration and pollution, and needs to be replaced and treated regularly, which pollutes the mine environment and has high treatment costs.

[0005] (3) Poor system reliability: The massive pipeline system is a high-risk area for failures. Any leak or burst in the main pipeline, as well as the failure of the pump station itself, will cause the entire working face to stop working.

[0006] (4) Low control accuracy: All supports share the same hydraulic source, which makes it impossible to accurately "supply fluid on demand" according to the actual load requirements of a single support, resulting in poor pressure control accuracy, pressure overshoot and fluctuation, which affects the support effect. Summary of the Invention

[0007] This invention aims to overcome the problems of low energy efficiency, environmental pollution, poor reliability, and imprecise control in traditional centralized hydraulic support systems. It provides an intelligent control method for integrated electric-driven hydraulic support variable frequency pump stations. By introducing a novel distributed electric drive architecture and combining it with AI intelligent control algorithms, the power source of a single hydraulic support is optimized.

[0008] This invention provides an intelligent control method for an integrated electro-hydraulic support variable frequency pump station, comprising the following steps: S1. An onboard electric drive pump station is independently integrated on each hydraulic support; S2. After startup, the controller set on the onboard electric drive pump station loads the preset fuzzy PID rule base and efficiency MAP model. S3. The controller receives action commands from the host computer or local computer and generates the target pressure P0 based on the action commands. S4. The controller collects the column pressure P of each hydraulic support in real time, calculates the error and error change rate based on the target pressure P0 and the column pressure P; calculates the base motor speed N0 through a preset fuzzy PID rule library; and then searches for the optimal efficiency operating point of the motor under the current column pressure P using the gradient ascent method with the base motor speed N0 as the search starting point through the efficiency MAP model, outputs the optimal motor speed N, and sends the optimal motor speed N as a motion command to the frequency converter. S5. The frequency converter drives the motor and pump according to the optimal motor speed N; S6. The controller determines whether the hydraulic support has entered the pressure holding state: when the deviation between the column pressure P and the target pressure P0 is |P-P0|≤0.5MPa and the duration of this state is t≥3s, it is determined that the pressure holding state has been entered and proceeds to step S7; otherwise, it returns to step S3. S7. The controller sends a speed reduction command to the frequency converter. The frequency converter reduces the motor speed to zero according to the speed reduction command and returns to step S6.

[0009] The intelligent control method for an integrated electro-hydraulic support variable frequency pump station according to the present invention, in a preferred embodiment, further includes the following steps in step S4: S41. The controller collects the column pressure P of the hydraulic support in real time with a sampling period of 10ms, and calculates the error e and the error change rate Δe based on the column pressure P and the target pressure P0. S42, The controller uses a preset fuzzy PID rule base to analyze K. p K i K d The parameters are tuned online to calculate the base motor speed N0; where K p K is the initial controller proportional parameter. i K is the initial controller integral parameter. d These are the initial controller differential parameters; S43. Using the efficiency MAP model calibrated by actual measurement for hydraulic oil medium under pressure-speed operating conditions, with the basic motor speed N0 as the center, the gradient ascent method is used to iteratively search for the optimal efficiency operating point of the motor under the current column pressure P, and the optimal motor speed N is output.

[0010] The intelligent control method for an integrated electro-hydraulic support variable frequency pump station described in this invention, as a preferred embodiment, calculates the error e(t) as follows: ; Where t is time, e(t) is the calculation error at time t, and P(t) is the column pressure at time t.

[0011] The intelligent control method for an integrated electro-hydraulic support variable frequency pump station described in this invention, as a preferred embodiment, has an error change rate Δe(t) as follows: ; Where t is time, e(t) is the calculation error at time t, and e(t-1) is the calculation error at time t-1.

[0012] The intelligent control method for an integrated electro-driven hydraulic support frequency conversion pump station according to the present invention, in a preferred embodiment, further includes the following steps in step S42: S421. Map the membership functions of the calculation error e(t) and the error change rate Δe(t) to fuzzy linguistic variables; S422. Establish a fuzzy PID rule base based on expert experience, and convert the fuzzy output obtained by reasoning based on the fuzzy PID rule base into a numerical value using the centroid method. S423, Calculate the base motor speed N0: ; Where t is time, N0(t) is the base motor speed at time t, and K p (t) represents the proportional parameter of the controller at time t, K i (t) represents the controller integral parameter at time t, K d (t) represents the differential parameter of the controller at time t; ; ; ; Among them, K p K is the initial controller proportional parameter. i K is the initial controller integral parameter. d Let ΔK be the initial controller differential parameter. p For the fuzzy output of the controller's proportional parameters, ΔK i For the fuzzy output of the controller integral parameters, ΔK d This is the fuzzy output of the controller's differential parameters.

[0013] The intelligent control method for an integrated electric-driven hydraulic support frequency conversion pump station described in this invention preferably uses a Gaussian function or a trigonometric function as the membership function.

[0014] In the intelligent control method for an integrated electro-hydraulic support variable frequency pump station described in this invention, as a preferred embodiment, the efficiency MAP model in step S43 is a function: ; Where n is the motor speed and P is the column pressure; The efficiency MAP model is a two-dimensional efficiency surface obtained by collecting no less than 500 sets of (n, P, η) data points through bench tests within the pressure-speed operating range of 10~35MPa and 500~3000rpm using hydraulic oil as the working medium, and by multinomial fitting or neural network modeling. The efficiency MAP model differs from the general hydraulic pump efficiency model that uses emulsions or water as the medium. Its viscosity-temperature characteristics, volumetric efficiency, and mechanical efficiency parameters are calibrated specifically for hydraulic oil media.

[0015] The intelligent control method for an integrated electro-hydraulic support variable frequency pump station described in this invention, as a preferred method, uses an efficiency MAP model obtained by polynomial fitting or neural network modeling based on experimental data.

[0016] The intelligent control method for an integrated electric drive hydraulic support variable frequency pump station described in this invention, as a preferred embodiment, involves an iterative process in step S43 where the gradient ascent method searches for the optimal motor speed N, which is used to find the local maximum value of the efficiency MAP model under the column pressure P.

[0017] The intelligent control method for an integrated electro-hydraulic support variable frequency pump station described in this invention, as a preferred embodiment, includes the following iterative process: S431. Calculate the gradient: Calculate the partial derivative of the efficiency MAP value with respect to the rotational speed n at the current operating point (n, P): ; S432, Iterative Update: Update the rotation speed n according to the gradient direction to find a better rotation speed point: ; Where α is the learning rate and k is the number of iterations; S433, Boundary Constraints: The updated rotational speed n is within the allowable operating range of the motor; S434, Final Output: The optimal motor speed N after iterative updates is sent to the frequency converter as a motion command.

[0018] The present invention has the following beneficial effects: (1) Energy saving: It fundamentally eliminates pressure loss in long-distance pipelines. Through "on-demand start-up" and "efficiency MAP optimization", it ensures that the pump station always operates in the most efficient range. Compared with the traditional centralized liquid supply system, the comprehensive energy saving of a single unit can reach more than 20%, and the energy saving effect of the entire working face is more significant.

[0019] (2) Green and environmentally friendly: Clean, closed-loop hydraulic oil replaces the easily polluted and frequently treated emulsion, achieving zero emissions and fully meeting the requirements for green mine construction.

[0020] (3) Extremely high reliability and safety: The distributed architecture eliminates the risk of single point of failure, and the failure of the power system of any single support will not affect other supports. At the same time, the high-pressure liquid supply pipeline is eliminated, eliminating the major safety hazard of pipeline bursts causing injury or production stoppage.

[0021] (3) Control performance: It realizes independent and precise liquid supply to a single support, and the pressure control accuracy can be improved to within ±0.5MPa, resulting in more stable support effect and longer equipment life. Attached Figure Description

[0022] Figure 1 Flowchart of intelligent control method for integrated electric drive hydraulic support frequency conversion pump station; Figure 2 A flowchart illustrating the intelligent control method for an integrated electric-driven hydraulic support frequency converter pump station. Figure 3 A flowchart illustrating the iterative process of intelligent control methods for integrated electric-driven hydraulic support frequency conversion pump stations. Detailed Implementation

[0023] Example 1 like Figure 1 As shown, an intelligent control method for an integrated electro-hydraulic support variable frequency pump station includes the following steps: S1. An onboard electric drive pump station is independently integrated on each hydraulic support; S2. After startup, the controller set on the onboard electric drive pump station loads the preset fuzzy PID rule base and efficiency MAP model. S3. The controller receives action commands from the host computer or local computer and generates the target pressure P0 based on the action commands. S4. The controller collects the column pressure P of each hydraulic support in real time with a sampling period of 10ms. It calculates the error and error rate of change based on the target pressure P0 and the column pressure P, and then applies this to K using a preset fuzzy PID rule base. p K i K d The parameters are tuned online, and the base motor speed N0 is calculated, where K p K is the initial controller proportional parameter. i K is the initial controller integral parameter. dThe initial controller differential parameters are used; then, using the efficiency MAP model η=f(n,P) calibrated by actual measurements for hydraulic oil under pressure-speed conditions, with the base motor speed N0 as the search starting point, the gradient ascent method is used to iteratively find the optimal efficiency operating point of the motor under the current column pressure P, outputting the optimal motor speed N, and sending the optimal motor speed N as a motion command to the frequency converter; the pressure range of the pressure-speed operating range is 10~35MPa, and the speed range is 500~3000rpm; Figure 2 As shown, it further includes the following steps: S41. The controller collects the column pressure P and calculates the error e and the error change rate Δe based on the column pressure P and the target pressure P0. The calculation error e(t) is: ; Where t is time, e(t) is the calculation error at time t, and P(t) is the column pressure at time t; The rate of change of error Δe(t) is: ; Where t is time, e(t) is the calculation error at time t, and e(t-1) is the calculation error at time t-1; S42. The controller calculates the basic motor speed N0 using a fuzzy PID rule base; this further includes the following steps: S421. Map the membership functions of the calculation error e(t) and the error change rate Δe(t) to fuzzy linguistic variables, such as {negative large (NB), negative medium (NM), zero (ZE), positive medium (PM), positive large (PB)}; the membership functions are Gaussian functions or trigonometric functions; S422. Establish a fuzzy PID rule base based on expert experience, and convert the fuzzy output obtained by reasoning from the fuzzy PID rule base into a numerical value using the centroid method; the fuzzy PID rule base is shown in the table below:

[0024] S423, Calculate the base motor speed N0: ; Where t is time, N0(t) is the base motor speed at time t, and K p (t) represents the proportional parameter of the controller at time t, K i (t) represents the controller integral parameter at time t, K d (t) represents the differential parameter of the controller at time t; ; ; ; Among them, K p K is the initial controller proportional parameter. i K is the initial controller integral parameter. d Let ΔK be the initial controller differential parameter. p For the fuzzy output of the controller's proportional parameters, ΔK i For the fuzzy output of the controller integral parameters, ΔK d This is the fuzzy output of the controller's differential parameters.

[0025] S43. Using the efficiency MAP model, with the basic motor speed N0 as the center, use the gradient ascent method to search for the optimal motor speed N. The efficiency MAP model is a function: ; Where n is the motor speed and P is the column pressure; The efficiency MAP model is a two-dimensional efficiency surface obtained by collecting no less than 500 sets of (n, P, η) data points through bench tests within the pressure-speed operating range of 10~35MPa and 500~3000rpm using hydraulic oil as the working medium, and by multinomial fitting or neural network modeling. The efficiency MAP model differs from the general hydraulic pump efficiency model that uses emulsions or water as the medium. Its viscosity-temperature characteristics, volumetric efficiency, and mechanical efficiency parameters are calibrated specifically for hydraulic oil media. The gradient ascent method for finding the optimal motor speed N is an iterative process used to find the local maximum value of the efficiency MAP model under column pressure P. like Figure 3 As shown, the iterative process includes the following steps: S431. Calculate the gradient: Calculate the partial derivative of the efficiency MAP value with respect to the rotational speed n at the current operating point (n, P): ;

[0026] S432, Iterative Update: Update the rotation speed n according to the gradient direction to find a better rotation speed point: ; Where α is the learning rate and k is the number of iterations; S433, Boundary Constraints: The updated rotational speed n is within the allowable operating range of the motor; S434, Final Output: The optimal motor speed N after iterative updates is sent to the frequency converter as a motion command; S5. The frequency converter drives the motor and pump according to the optimal motor speed N; S6. The controller determines whether the hydraulic support has entered the pressure holding state: when the deviation between the column pressure P and the target pressure P0 is |P-P0|≤0.5MPa and the duration of this state is t≥3s, it is determined that the pressure holding state has been entered and proceeds to step S7; otherwise, it returns to step S3. S7. The controller sends a speed reduction command to the frequency converter. The frequency converter reduces the motor speed to zero according to the speed reduction command and returns to step S6.

[0027] The hardware system of this embodiment includes: an independent onboard electric drive pump station integrated on each hydraulic support, consisting of a servo motor, frequency converter, fixed displacement hydraulic pump, small hydraulic oil tank, pressure sensor and edge controller. The system uses hydraulic oil as the working medium to form an independent closed hydraulic circuit.

[0028] The control system employs a two-layer intelligent control algorithm based on fuzzy PID and efficiency MAP optimization, running in the edge controller, including: The bottom pressure regulation layer employs a fuzzy PID control algorithm, using the error e and the rate of change Δe between the target pressure P0 and the actual pressure P of the support column as inputs. Through a built-in fuzzy rule library (e.g., "If the error is large and changes rapidly, then significantly increase the PID parameter"), the proportional parameter K of the PID controller is adjusted online in real time. p Integral parameter K i Differential parameter K d This enables the controller to respond quickly to load changes and effectively suppress pressure overshoot and fluctuations.

[0029] Upper-level energy efficiency optimization layer: Employing the efficiency MAP optimization algorithm, this module uses a two-dimensional efficiency map (EfficiencyMAP) of the total efficiency η of the "motor-pump" system with respect to the motor speed n and the system pressure P, which is pre-established through experimental calibration or online identification. After the underlying fuzzy PID controller calculates the base speed n that meets the current pressure control requirements, this module does not execute it directly. Instead, it uses this as a benchmark to perform a small-range search on the MAP using the gradient ascent method to find the speed point n with the highest energy efficiency, i.e., the optimal motor speed N, as the final motor speed command.

[0030] During actual operation, at time t, the controller calculates the base rotation speed n = 1200 rpm and the current column pressure P = 25 MPa. The gradient calculated by the efficiency MAP model at this point =0.005% / rpm; Set the learning rate α = 1000; The rotational speed after the first iteration is n(1) = 1200 + 1000 × 0.005 = 1205 rpm; The controller fine-tunes the command from 1200rpm to 1205rpm, thereby improving the operating efficiency of the pump station and achieving energy saving.

[0031] Example 2 The method and hardware system of Example 1 are compared and verified with the traditional variable frequency constant pressure liquid supply scheme as follows: A 30-day industrial comparative test was conducted on a fully mechanized mining face (260m long, 130 supports, rated working resistance 14000kN) in a certain mine. The comparison scheme is as follows: Comparison with Option A: Traditional centralized variable frequency constant pressure liquid supply solution (pump station constant pressure 32MPa, pipeline end pressure approximately 25MPa). Comparison Scheme B: Distributed onboard electric drive pump station + constant pressure PID control (target pressure fixed at 35MPa); Scheme C of this invention: Distributed onboard electric drive pump station + fuzzy PID + efficiency MAP gradient optimization control.

[0032] The experimental results are shown in the table below:

[0033] As shown in the table above, the pressure control accuracy of Scheme C in this invention reaches ±0.4 MPa, which is approximately 81% higher than the centralized constant pressure scheme and approximately 67% higher than the constant pressure PID scheme; the energy saving rate per unit reaches 40.2%, of which pipeline pressure loss elimination contributes approximately 22% and efficiency MAP optimization contributes approximately 18%; the motor completely stops during the pressure holding stage, eliminating the ineffective energy consumption of idling during pressure holding in traditional schemes. The above data strongly support the claim of this invention that "the comprehensive energy saving per unit can reach more than 20%", and the actual energy saving effect exceeds expectations.

[0034] The above description is illustrative only and not restrictive of the present invention. Those skilled in the art will understand that any modifications, variations or equivalents that can be made without departing from the spirit and scope defined by the claims will fall within the protection scope of the present invention.

Claims

1. An intelligent control method for an integrated electro-hydraulic support variable frequency pump station, characterized in that: Includes the following steps: S1. An onboard electric drive pump station is independently integrated on each hydraulic support; S2. After startup, the controller on the onboard electric drive pump station loads the preset fuzzy PID rule base and efficiency MAP model. S3. The controller receives action commands from the host computer or local computer, and generates a target pressure P0 based on the action commands. S4. The controller collects the column pressure P of each hydraulic support in real time, and calculates the error and error change rate based on the target pressure P0 and the column pressure P. The base motor speed N0 is calculated using a pre-defined fuzzy PID rule base. Then, using the efficiency MAP model, with the basic motor speed N0 as the search starting point, the gradient ascent method is used to search for the optimal efficiency operating point of the motor under the current column pressure P, output the optimal motor speed N, and send the optimal motor speed N as a motion command to the frequency converter. S5. The frequency converter drives the operation of the motor and the pump according to the optimal speed N of the motor; S6. The controller determines whether the hydraulic support has entered the pressure holding state: when the deviation between the column pressure P and the target pressure P0 is |P-P0|≤0.5MPa and the duration of this state is t≥3s, it is determined that the pressure holding state has been entered, and proceeds to step S7. Otherwise, return to step S3; S7. The controller sends a speed reduction command to the frequency converter, and the frequency converter reduces the motor speed to zero according to the speed reduction command, and returns to step S6.

2. The intelligent control method for an integrated electro-hydraulic support frequency conversion pump station according to claim 1, characterized in that: Step S4 further includes the following steps: S41. The controller collects the column pressure P of the hydraulic support in real time with a sampling period of 10ms, and calculates the error e and the error change rate Δe based on the column pressure P and the target pressure P0. S42, The controller uses a preset fuzzy PID rule base to analyze K. p K i K d The parameters are tuned online, and the base motor speed N0 is calculated, where K p K is the initial controller proportional parameter. i K is the initial controller integral parameter. d These are the initial controller differential parameters; S43. Using the efficiency MAP model calibrated by actual measurement for hydraulic oil medium under pressure-speed operating conditions, starting from the basic motor speed N0, the gradient ascent method is used to iteratively search for the optimal efficiency operating point of the motor under the current column pressure P, and the optimal motor speed N is output.

3. The intelligent control method for an integrated electro-hydraulic support frequency conversion pump station according to claim 2, characterized in that: The calculation error e(t) is: ； Where t is time, e(t) is the calculation error at time t, and P(t) is the column pressure at time t.

4. The intelligent control method for an integrated electro-hydraulic support frequency conversion pump station according to claim 3, characterized in that: The error change rate Δe(t) is: ； Where t is time, e(t) is the calculation error at time t, and e(t-1) is the calculation error at time t-1.

5. The intelligent control method for an integrated electro-hydraulic support frequency conversion pump station according to claim 4, characterized in that: Step S42 further includes the following steps: S421. Map the calculated error e(t) and the error change rate Δe(t) membership function into fuzzy linguistic variables; S422. Based on expert experience, establish the fuzzy PID rule base, and convert the fuzzy output obtained by reasoning based on the fuzzy PID rule base into a numerical value using the centroid method; S423. Calculate the speed N0 of the basic motor: ； Where t is time, N0(t) is the basic motor speed at time t, and K p (t) represents the proportional parameter of the controller at time t, K i (t) represents the controller integral parameter at time t, K d (t) represents the differential parameter of the controller at time t; ； ； ； Among them, K p K is the initial controller proportional parameter. i K is the initial controller integral parameter. d Let ΔK be the initial controller differential parameter. p For the fuzzy output of the controller's proportional parameters, ΔK i For the fuzzy output of the controller integral parameters, ΔK d This is the fuzzy output of the controller's differential parameters.

6. The intelligent control method for an integrated electro-hydraulic support frequency conversion pump station according to claim 5, characterized in that: The membership function is a Gaussian function or a trigonometric function.

7. The intelligent control method for an integrated electro-hydraulic support frequency conversion pump station according to claim 2, characterized in that: The efficiency MAP model described in step S43 is a function: ； Where n is the motor speed and P is the column pressure; The efficiency MAP model is a two-dimensional efficiency surface obtained by collecting no less than 500 sets of (n, P, η) data points through bench tests within the pressure-speed operating range of 10~35MPa and 500~3000rpm using hydraulic oil as the working medium, and by multinomial fitting or neural network modeling. The efficiency MAP model differs from the general hydraulic pump efficiency model that uses emulsions or water as the medium. Its viscosity-temperature characteristics, volumetric efficiency, and mechanical efficiency parameters are calibrated specifically for hydraulic oil media.

8. The intelligent control method for an integrated electro-hydraulic support frequency conversion pump station according to claim 7, characterized in that: The efficiency MAP model is obtained by multinomial fitting or neural network modeling of experimental data.

9. The intelligent control method for an integrated electro-hydraulic support frequency conversion pump station according to claim 2, characterized in that: The gradient ascent method described in step S43 to search for the optimal motor speed N is an iterative process used to find the local maximum value of the efficiency MAP model under the column pressure P.

10. The intelligent control method for an integrated electro-hydraulic support frequency conversion pump station according to claim 9, characterized in that: The iterative process includes the following steps: S431. Calculate the gradient: Calculate the partial derivative of the efficiency MAP value with respect to the rotational speed n at the current operating point (n, P): ； S432, Iterative Update: Update the rotation speed n according to the gradient direction to find a better rotation speed point: ； Where α is the learning rate and k is the number of iterations; S433, Boundary Constraint: The updated rotational speed n is within the allowable operating range of the motor; S434. Final output: The optimal motor speed N after the iterative update is sent to the frequency converter as the motion command.

Images